train_rhn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/zake7749/.local/lib/python3.5/site-packages/h5py/__init__.py:34: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n",
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "\n",
    "from sotoxic.data_helper.data_transformer import DataTransformer\n",
    "from sotoxic.data_helper.data_loader import DataLoader\n",
    "from sotoxic.train.trainer import PyTorchModelTrainer\n",
    "from sotoxic.config import dataset_config "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "VOCAB_SIZE = 100000\n",
    "MAX_SEQUENCE_LENGTH = 300\n",
    "EMBEDDING_SIZE = 300"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prepare training and testing data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading the dataset\n",
      "Doing preprocessing...\n",
      "Preprocessed.\n"
     ]
    }
   ],
   "source": [
    "print(\"Loading the dataset\")\n",
    "data_transformer = DataTransformer(max_num_words=VOCAB_SIZE, max_sequence_length=MAX_SEQUENCE_LENGTH, char_level=False)\n",
    "data_loader = DataLoader()\n",
    "train_sequences, training_labels, test_sequences = data_transformer.prepare_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[686,\n",
       "  76,\n",
       "  1,\n",
       "  132,\n",
       "  124,\n",
       "  173,\n",
       "  28,\n",
       "  705,\n",
       "  4323,\n",
       "  9377,\n",
       "  1162,\n",
       "  81,\n",
       "  350,\n",
       "  44,\n",
       "  81,\n",
       "  12,\n",
       "  13593,\n",
       "  50,\n",
       "  6506,\n",
       "  15,\n",
       "  59,\n",
       "  2539,\n",
       "  143,\n",
       "  3,\n",
       "  2736,\n",
       "  31,\n",
       "  111,\n",
       "  1190,\n",
       "  16064,\n",
       "  2418,\n",
       "  6,\n",
       "  51,\n",
       "  20,\n",
       "  12,\n",
       "  239,\n",
       "  1,\n",
       "  349,\n",
       "  29,\n",
       "  1,\n",
       "  46,\n",
       "  136,\n",
       "  3,\n",
       "  38,\n",
       "  3789,\n",
       "  85],\n",
       " [337,\n",
       "  7767,\n",
       "  42,\n",
       "  2818,\n",
       "  14,\n",
       "  628,\n",
       "  3656,\n",
       "  3,\n",
       "  38,\n",
       "  4700,\n",
       "  2584,\n",
       "  22,\n",
       "  90,\n",
       "  46,\n",
       "  929,\n",
       "  3682,\n",
       "  987,\n",
       "  674,\n",
       "  10173],\n",
       " [429,\n",
       "  363,\n",
       "  3,\n",
       "  38,\n",
       "  118,\n",
       "  12,\n",
       "  250,\n",
       "  2,\n",
       "  77,\n",
       "  291,\n",
       "  10,\n",
       "  50,\n",
       "  9,\n",
       "  14,\n",
       "  551,\n",
       "  8,\n",
       "  2391,\n",
       "  496,\n",
       "  478,\n",
       "  99,\n",
       "  6,\n",
       "  568,\n",
       "  2,\n",
       "  39,\n",
       "  314,\n",
       "  132,\n",
       "  364,\n",
       "  4,\n",
       "  28,\n",
       "  46,\n",
       "  42,\n",
       "  183,\n",
       "  2,\n",
       "  440,\n",
       "  58,\n",
       "  34,\n",
       "  1,\n",
       "  2227,\n",
       "  88,\n",
       "  1,\n",
       "  696,\n",
       "  474],\n",
       " [58,\n",
       "  3,\n",
       "  134,\n",
       "  92,\n",
       "  55,\n",
       "  311,\n",
       "  1363,\n",
       "  15,\n",
       "  2063,\n",
       "  3,\n",
       "  5531,\n",
       "  23,\n",
       "  1,\n",
       "  109,\n",
       "  2330,\n",
       "  53,\n",
       "  18,\n",
       "  485,\n",
       "  15,\n",
       "  25,\n",
       "  5,\n",
       "  3714,\n",
       "  4,\n",
       "  1310,\n",
       "  4,\n",
       "  9972,\n",
       "  3,\n",
       "  64,\n",
       "  1,\n",
       "  273,\n",
       "  83,\n",
       "  115,\n",
       "  12892,\n",
       "  35,\n",
       "  9,\n",
       "  44,\n",
       "  16,\n",
       "  41,\n",
       "  11,\n",
       "  1,\n",
       "  1470,\n",
       "  131,\n",
       "  1245,\n",
       "  2128,\n",
       "  418,\n",
       "  1245,\n",
       "  304,\n",
       "  3,\n",
       "  40,\n",
       "  20,\n",
       "  9,\n",
       "  485,\n",
       "  15,\n",
       "  23,\n",
       "  48,\n",
       "  47,\n",
       "  290,\n",
       "  68,\n",
       "  108,\n",
       "  23,\n",
       "  7,\n",
       "  17,\n",
       "  55,\n",
       "  4270,\n",
       "  13,\n",
       "  2227,\n",
       "  305,\n",
       "  15,\n",
       "  273,\n",
       "  25,\n",
       "  101,\n",
       "  2,\n",
       "  20,\n",
       "  10,\n",
       "  198,\n",
       "  51,\n",
       "  194,\n",
       "  39,\n",
       "  72,\n",
       "  36,\n",
       "  525,\n",
       "  2,\n",
       "  18,\n",
       "  5,\n",
       "  6072,\n",
       "  15,\n",
       "  45,\n",
       "  13,\n",
       "  367,\n",
       "  35,\n",
       "  3,\n",
       "  604,\n",
       "  36,\n",
       "  83,\n",
       "  18,\n",
       "  5,\n",
       "  2118,\n",
       "  369,\n",
       "  5,\n",
       "  2841,\n",
       "  2937,\n",
       "  71,\n",
       "  10,\n",
       "  420,\n",
       "  11,\n",
       "  1,\n",
       "  478,\n",
       "  597,\n",
       "  1398,\n",
       "  95,\n",
       "  3085,\n",
       "  5258],\n",
       " [7, 1690, 16, 28, 2776, 55, 1052, 7, 564, 37, 9, 15],\n",
       " [2728, 29, 39, 19, 96, 79, 1, 2348, 96, 46],\n",
       " [2525, 142, 7, 2077, 303, 15, 28, 135],\n",
       " [21,\n",
       "  208,\n",
       "  2,\n",
       "  1,\n",
       "  3581,\n",
       "  43,\n",
       "  57,\n",
       "  350,\n",
       "  51,\n",
       "  20,\n",
       "  12,\n",
       "  20,\n",
       "  10,\n",
       "  119,\n",
       "  25,\n",
       "  7,\n",
       "  33,\n",
       "  18,\n",
       "  727],\n",
       " [263,\n",
       "  23,\n",
       "  1,\n",
       "  252,\n",
       "  598,\n",
       "  24,\n",
       "  1407,\n",
       "  2,\n",
       "  7,\n",
       "  463,\n",
       "  3,\n",
       "  38,\n",
       "  12,\n",
       "  9316,\n",
       "  2,\n",
       "  332,\n",
       "  204,\n",
       "  11,\n",
       "  1,\n",
       "  1365,\n",
       "  44,\n",
       "  32,\n",
       "  2796,\n",
       "  15,\n",
       "  39,\n",
       "  13,\n",
       "  208,\n",
       "  3,\n",
       "  38,\n",
       "  1082,\n",
       "  1135,\n",
       "  9,\n",
       "  10,\n",
       "  18,\n",
       "  58,\n",
       "  1230,\n",
       "  35,\n",
       "  47,\n",
       "  40,\n",
       "  79,\n",
       "  10,\n",
       "  13,\n",
       "  450,\n",
       "  19,\n",
       "  5,\n",
       "  259,\n",
       "  3,\n",
       "  17,\n",
       "  57,\n",
       "  2,\n",
       "  1,\n",
       "  4880,\n",
       "  8618,\n",
       "  26,\n",
       "  10,\n",
       "  609,\n",
       "  5,\n",
       "  1275,\n",
       "  10,\n",
       "  652,\n",
       "  2,\n",
       "  2269,\n",
       "  8618,\n",
       "  54,\n",
       "  8,\n",
       "  5,\n",
       "  632,\n",
       "  7656,\n",
       "  9,\n",
       "  1003,\n",
       "  7,\n",
       "  48,\n",
       "  474,\n",
       "  36,\n",
       "  229,\n",
       "  18,\n",
       "  145,\n",
       "  303,\n",
       "  22,\n",
       "  3510,\n",
       "  11,\n",
       "  11415],\n",
       " [11624, 15, 14, 218, 6, 54, 16, 1893, 2, 133, 4, 78105],\n",
       " [281,\n",
       "  79,\n",
       "  906,\n",
       "  13,\n",
       "  65930,\n",
       "  90,\n",
       "  13,\n",
       "  1476,\n",
       "  65930,\n",
       "  3,\n",
       "  430,\n",
       "  1,\n",
       "  147,\n",
       "  2967,\n",
       "  9,\n",
       "  1,\n",
       "  147,\n",
       "  8,\n",
       "  86,\n",
       "  126,\n",
       "  173,\n",
       "  281,\n",
       "  79,\n",
       "  26,\n",
       "  36,\n",
       "  8,\n",
       "  48,\n",
       "  686,\n",
       "  25,\n",
       "  906,\n",
       "  19,\n",
       "  2,\n",
       "  76,\n",
       "  98,\n",
       "  79,\n",
       "  11,\n",
       "  45,\n",
       "  2640,\n",
       "  281,\n",
       "  79,\n",
       "  11,\n",
       "  764,\n",
       "  2,\n",
       "  1,\n",
       "  7833,\n",
       "  281,\n",
       "  79,\n",
       "  349,\n",
       "  7,\n",
       "  229,\n",
       "  65,\n",
       "  332,\n",
       "  74,\n",
       "  15,\n",
       "  1,\n",
       "  147,\n",
       "  684,\n",
       "  5,\n",
       "  464,\n",
       "  686,\n",
       "  25,\n",
       "  906,\n",
       "  13,\n",
       "  76,\n",
       "  197,\n",
       "  14,\n",
       "  147,\n",
       "  11,\n",
       "  409,\n",
       "  8,\n",
       "  1879,\n",
       "  22,\n",
       "  281,\n",
       "  79,\n",
       "  51,\n",
       "  107,\n",
       "  2,\n",
       "  1,\n",
       "  147,\n",
       "  684,\n",
       "  6,\n",
       "  77,\n",
       "  10,\n",
       "  2,\n",
       "  425,\n",
       "  5,\n",
       "  281,\n",
       "  79,\n",
       "  906,\n",
       "  23,\n",
       "  7,\n",
       "  17,\n",
       "  701,\n",
       "  61,\n",
       "  281,\n",
       "  79,\n",
       "  296,\n",
       "  394,\n",
       "  1410,\n",
       "  9,\n",
       "  7,\n",
       "  17,\n",
       "  2073,\n",
       "  1,\n",
       "  281,\n",
       "  79,\n",
       "  906,\n",
       "  15,\n",
       "  133,\n",
       "  45,\n",
       "  149,\n",
       "  7,\n",
       "  40,\n",
       "  129,\n",
       "  5,\n",
       "  153,\n",
       "  4,\n",
       "  147,\n",
       "  45,\n",
       "  7,\n",
       "  17,\n",
       "  592,\n",
       "  30,\n",
       "  1758,\n",
       "  15,\n",
       "  1,\n",
       "  28,\n",
       "  398,\n",
       "  158,\n",
       "  10,\n",
       "  8,\n",
       "  1772,\n",
       "  31,\n",
       "  1,\n",
       "  97,\n",
       "  308,\n",
       "  4,\n",
       "  55,\n",
       "  78,\n",
       "  7,\n",
       "  16,\n",
       "  1697,\n",
       "  11,\n",
       "  6,\n",
       "  80,\n",
       "  3163,\n",
       "  147,\n",
       "  29,\n",
       "  1,\n",
       "  3582,\n",
       "  872,\n",
       "  181,\n",
       "  9,\n",
       "  55,\n",
       "  281,\n",
       "  79,\n",
       "  344,\n",
       "  701,\n",
       "  143,\n",
       "  377,\n",
       "  83,\n",
       "  624,\n",
       "  6,\n",
       "  2534,\n",
       "  103,\n",
       "  27,\n",
       "  686,\n",
       "  33,\n",
       "  18,\n",
       "  140,\n",
       "  47,\n",
       "  702,\n",
       "  143,\n",
       "  44,\n",
       "  17,\n",
       "  57,\n",
       "  701,\n",
       "  19,\n",
       "  897,\n",
       "  15,\n",
       "  410,\n",
       "  13,\n",
       "  270,\n",
       "  113,\n",
       "  23,\n",
       "  7,\n",
       "  17,\n",
       "  55,\n",
       "  244,\n",
       "  51,\n",
       "  214,\n",
       "  84,\n",
       "  31,\n",
       "  1,\n",
       "  296,\n",
       "  249,\n",
       "  244,\n",
       "  128,\n",
       "  7,\n",
       "  46,\n",
       "  710,\n",
       "  6885,\n",
       "  112,\n",
       "  13,\n",
       "  65930,\n",
       "  90,\n",
       "  13,\n",
       "  1476,\n",
       "  65930,\n",
       "  3,\n",
       "  612,\n",
       "  9,\n",
       "  1,\n",
       "  736,\n",
       "  684,\n",
       "  507,\n",
       "  68,\n",
       "  12,\n",
       "  2129,\n",
       "  60,\n",
       "  322,\n",
       "  1,\n",
       "  166,\n",
       "  35,\n",
       "  1,\n",
       "  249,\n",
       "  663,\n",
       "  8,\n",
       "  2145,\n",
       "  23,\n",
       "  7,\n",
       "  67,\n",
       "  12,\n",
       "  447,\n",
       "  14,\n",
       "  736,\n",
       "  198,\n",
       "  80,\n",
       "  7,\n",
       "  33,\n",
       "  115,\n",
       "  2,\n",
       "  2129,\n",
       "  1,\n",
       "  2305,\n",
       "  4,\n",
       "  1,\n",
       "  249,\n",
       "  23,\n",
       "  7,\n",
       "  3671,\n",
       "  10,\n",
       "  29,\n",
       "  5,\n",
       "  387,\n",
       "  80,\n",
       "  5,\n",
       "  158,\n",
       "  2,\n",
       "  1,\n",
       "  387,\n",
       "  29,\n",
       "  54,\n",
       "  10,\n",
       "  24,\n",
       "  519,\n",
       "  785,\n",
       "  22,\n",
       "  5,\n",
       "  7552,\n",
       "  4,\n",
       "  9,\n",
       "  387,\n",
       "  625,\n",
       "  4,\n",
       "  79,\n",
       "  4,\n",
       "  98,\n",
       "  166,\n",
       "  8,\n",
       "  772,\n",
       "  1752,\n",
       "  99,\n",
       "  184,\n",
       "  23,\n",
       "  1,\n",
       "  249,\n",
       "  2178,\n",
       "  8,\n",
       "  254,\n",
       "  29,\n",
       "  1,\n",
       "  387,\n",
       "  2604,\n",
       "  80,\n",
       "  87,\n",
       "  249,\n",
       "  53,\n",
       "  65,\n",
       "  18,\n",
       "  3415,\n",
       "  19,\n",
       "  96,\n",
       "  19,\n",
       "  295,\n",
       "  1,\n",
       "  112,\n",
       "  51,\n",
       "  144,\n",
       "  5,\n",
       "  765,\n",
       "  249,\n",
       "  3672,\n",
       "  258,\n",
       "  23,\n",
       "  1,\n",
       "  736,\n",
       "  68,\n",
       "  12,\n",
       "  17,\n",
       "  47,\n",
       "  223,\n",
       "  23,\n",
       "  7,\n",
       "  322,\n",
       "  605,\n",
       "  1,\n",
       "  458,\n",
       "  2968,\n",
       "  25,\n",
       "  782,\n",
       "  80,\n",
       "  1,\n",
       "  258,\n",
       "  40,\n",
       "  18,\n",
       "  126,\n",
       "  2,\n",
       "  916,\n",
       "  10,\n",
       "  173,\n",
       "  1,\n",
       "  176,\n",
       "  1391,\n",
       "  794,\n",
       "  23,\n",
       "  7,\n",
       "  207,\n",
       "  1,\n",
       "  296,\n",
       "  1307,\n",
       "  1,\n",
       "  410,\n",
       "  31,\n",
       "  281,\n",
       "  79,\n",
       "  79,\n",
       "  5,\n",
       "  258,\n",
       "  103,\n",
       "  19,\n",
       "  25,\n",
       "  47,\n",
       "  4,\n",
       "  1,\n",
       "  61,\n",
       "  725,\n",
       "  420,\n",
       "  31,\n",
       "  147,\n",
       "  249,\n",
       "  725,\n",
       "  281,\n",
       "  79,\n",
       "  63,\n",
       "  147,\n",
       "  249,\n",
       "  725,\n",
       "  13,\n",
       "  1,\n",
       "  481,\n",
       "  153,\n",
       "  4,\n",
       "  249,\n",
       "  725,\n",
       "  9,\n",
       "  7,\n",
       "  40,\n",
       "  79,\n",
       "  23,\n",
       "  7,\n",
       "  17,\n",
       "  701,\n",
       "  61,\n",
       "  1465,\n",
       "  394,\n",
       "  1410,\n",
       "  9,\n",
       "  7,\n",
       "  17,\n",
       "  2073,\n",
       "  87,\n",
       "  112,\n",
       "  6,\n",
       "  857,\n",
       "  84,\n",
       "  149,\n",
       "  7,\n",
       "  40,\n",
       "  129,\n",
       "  5,\n",
       "  153,\n",
       "  4,\n",
       "  1465,\n",
       "  7,\n",
       "  17,\n",
       "  701,\n",
       "  30,\n",
       "  427,\n",
       "  14,\n",
       "  158,\n",
       "  1208,\n",
       "  6,\n",
       "  4701,\n",
       "  344,\n",
       "  83,\n",
       "  18,\n",
       "  140,\n",
       "  47,\n",
       "  702,\n",
       "  143,\n",
       "  44,\n",
       "  17,\n",
       "  57,\n",
       "  857,\n",
       "  19,\n",
       "  897,\n",
       "  15,\n",
       "  410,\n",
       "  13,\n",
       "  270,\n",
       "  113,\n",
       "  23,\n",
       "  1,\n",
       "  147,\n",
       "  8,\n",
       "  1445,\n",
       "  173,\n",
       "  5,\n",
       "  262,\n",
       "  176,\n",
       "  794,\n",
       "  466,\n",
       "  281,\n",
       "  79,\n",
       "  80,\n",
       "  1,\n",
       "  147,\n",
       "  33,\n",
       "  18,\n",
       "  140,\n",
       "  2324,\n",
       "  808,\n",
       "  143,\n",
       "  23,\n",
       "  7,\n",
       "  17,\n",
       "  55,\n",
       "  244,\n",
       "  51,\n",
       "  214,\n",
       "  84,\n",
       "  31,\n",
       "  1,\n",
       "  296,\n",
       "  249,\n",
       "  244,\n",
       "  128,\n",
       "  7,\n",
       "  46,\n",
       "  710],\n",
       " [22498, 18, 5, 363, 6, 2141, 530, 10, 335, 139, 1, 2682],\n",
       " [429,\n",
       "  37,\n",
       "  8,\n",
       "  10,\n",
       "  46,\n",
       "  37,\n",
       "  8,\n",
       "  10,\n",
       "  27,\n",
       "  4227,\n",
       "  412,\n",
       "  4,\n",
       "  59,\n",
       "  683,\n",
       "  47219,\n",
       "  60,\n",
       "  16,\n",
       "  95,\n",
       "  31,\n",
       "  5518,\n",
       "  535,\n",
       "  4557,\n",
       "  47220,\n",
       "  60,\n",
       "  3495,\n",
       "  71,\n",
       "  55,\n",
       "  47,\n",
       "  60,\n",
       "  5171,\n",
       "  84,\n",
       "  244,\n",
       "  20459,\n",
       "  87,\n",
       "  528,\n",
       "  1154,\n",
       "  6,\n",
       "  5788,\n",
       "  262,\n",
       "  1308,\n",
       "  31,\n",
       "  683,\n",
       "  214,\n",
       "  78106,\n",
       "  2,\n",
       "  1480,\n",
       "  71,\n",
       "  66,\n",
       "  954,\n",
       "  88,\n",
       "  251,\n",
       "  39,\n",
       "  6245,\n",
       "  1354],\n",
       " [142,\n",
       "  7,\n",
       "  404,\n",
       "  3938,\n",
       "  1419,\n",
       "  6,\n",
       "  1354,\n",
       "  31,\n",
       "  39,\n",
       "  2141,\n",
       "  367,\n",
       "  1,\n",
       "  77,\n",
       "  508,\n",
       "  271,\n",
       "  1536,\n",
       "  4824,\n",
       "  582,\n",
       "  8,\n",
       "  12,\n",
       "  155,\n",
       "  2,\n",
       "  13153,\n",
       "  21,\n",
       "  581,\n",
       "  10,\n",
       "  33,\n",
       "  1082,\n",
       "  92,\n",
       "  10,\n",
       "  138,\n",
       "  49,\n",
       "  7,\n",
       "  16,\n",
       "  2969,\n",
       "  21,\n",
       "  616,\n",
       "  19,\n",
       "  27,\n",
       "  395,\n",
       "  85,\n",
       "  1,\n",
       "  77,\n",
       "  508,\n",
       "  8,\n",
       "  478,\n",
       "  14,\n",
       "  8,\n",
       "  266,\n",
       "  1,\n",
       "  494,\n",
       "  125,\n",
       "  3294,\n",
       "  34,\n",
       "  1095,\n",
       "  8565,\n",
       "  42,\n",
       "  514,\n",
       "  19,\n",
       "  4,\n",
       "  1293,\n",
       "  66,\n",
       "  3648,\n",
       "  8,\n",
       "  320,\n",
       "  136,\n",
       "  42,\n",
       "  8,\n",
       "  1,\n",
       "  75,\n",
       "  791,\n",
       "  2949,\n",
       "  1195,\n",
       "  60,\n",
       "  67,\n",
       "  12,\n",
       "  5159,\n",
       "  9,\n",
       "  517,\n",
       "  58,\n",
       "  320,\n",
       "  88,\n",
       "  66,\n",
       "  32565,\n",
       "  27,\n",
       "  7008,\n",
       "  3,\n",
       "  2641,\n",
       "  2,\n",
       "  384,\n",
       "  14,\n",
       "  77,\n",
       "  11,\n",
       "  5890,\n",
       "  4,\n",
       "  15854,\n",
       "  1,\n",
       "  753,\n",
       "  4,\n",
       "  27,\n",
       "  395,\n",
       "  9,\n",
       "  8,\n",
       "  1263,\n",
       "  2,\n",
       "  138,\n",
       "  31,\n",
       "  1,\n",
       "  251,\n",
       "  508,\n",
       "  6,\n",
       "  12,\n",
       "  2470,\n",
       "  1419,\n",
       "  303,\n",
       "  353,\n",
       "  35,\n",
       "  15855,\n",
       "  354,\n",
       "  23,\n",
       "  7,\n",
       "  3290,\n",
       "  5,\n",
       "  781,\n",
       "  4,\n",
       "  2020,\n",
       "  117,\n",
       "  7,\n",
       "  40,\n",
       "  20,\n",
       "  14,\n",
       "  52,\n",
       "  40,\n",
       "  17,\n",
       "  5,\n",
       "  3883,\n",
       "  130,\n",
       "  15,\n",
       "  1,\n",
       "  424,\n",
       "  6,\n",
       "  2310,\n",
       "  1,\n",
       "  338,\n",
       "  16065],\n",
       " [396,\n",
       "  6,\n",
       "  1,\n",
       "  1540,\n",
       "  238,\n",
       "  492,\n",
       "  164,\n",
       "  965,\n",
       "  22,\n",
       "  39,\n",
       "  163,\n",
       "  2584,\n",
       "  164,\n",
       "  3945,\n",
       "  117,\n",
       "  10,\n",
       "  68,\n",
       "  12,\n",
       "  1265,\n",
       "  3,\n",
       "  207,\n",
       "  1,\n",
       "  581,\n",
       "  24,\n",
       "  313,\n",
       "  39,\n",
       "  6,\n",
       "  26,\n",
       "  49,\n",
       "  3,\n",
       "  157,\n",
       "  1,\n",
       "  883,\n",
       "  24,\n",
       "  3487,\n",
       "  6,\n",
       "  3,\n",
       "  6947,\n",
       "  90],\n",
       " [34157,\n",
       "  1051,\n",
       "  11,\n",
       "  2401,\n",
       "  34157,\n",
       "  19626,\n",
       "  24,\n",
       "  871,\n",
       "  255,\n",
       "  324,\n",
       "  80,\n",
       "  550,\n",
       "  1225,\n",
       "  5532,\n",
       "  54,\n",
       "  378,\n",
       "  34157,\n",
       "  1324,\n",
       "  921,\n",
       "  271,\n",
       "  1742,\n",
       "  22,\n",
       "  1,\n",
       "  23126,\n",
       "  1,\n",
       "  735,\n",
       "  43604,\n",
       "  2,\n",
       "  18,\n",
       "  2100,\n",
       "  2,\n",
       "  10402,\n",
       "  1856,\n",
       "  173,\n",
       "  6764,\n",
       "  5,\n",
       "  17193,\n",
       "  11,\n",
       "  1865,\n",
       "  42,\n",
       "  24,\n",
       "  596,\n",
       "  255,\n",
       "  324,\n",
       "  938,\n",
       "  74,\n",
       "  22,\n",
       "  66,\n",
       "  159,\n",
       "  3571,\n",
       "  19626,\n",
       "  1537,\n",
       "  6,\n",
       "  302,\n",
       "  35,\n",
       "  275,\n",
       "  42,\n",
       "  8,\n",
       "  851,\n",
       "  11,\n",
       "  6150,\n",
       "  42,\n",
       "  118,\n",
       "  8,\n",
       "  73,\n",
       "  93,\n",
       "  42,\n",
       "  18,\n",
       "  2155,\n",
       "  80,\n",
       "  9035,\n",
       "  6246,\n",
       "  6,\n",
       "  73,\n",
       "  93,\n",
       "  42,\n",
       "  18,\n",
       "  898,\n",
       "  78,\n",
       "  66,\n",
       "  3904,\n",
       "  2089,\n",
       "  1,\n",
       "  15595,\n",
       "  43604,\n",
       "  8,\n",
       "  967,\n",
       "  255,\n",
       "  324,\n",
       "  6150,\n",
       "  624,\n",
       "  34157,\n",
       "  649,\n",
       "  462,\n",
       "  9378,\n",
       "  23,\n",
       "  21,\n",
       "  891,\n",
       "  34,\n",
       "  9,\n",
       "  7514,\n",
       "  967,\n",
       "  107,\n",
       "  2,\n",
       "  21,\n",
       "  6,\n",
       "  161,\n",
       "  888,\n",
       "  66,\n",
       "  388,\n",
       "  4,\n",
       "  1623,\n",
       "  28,\n",
       "  462],\n",
       " [2500, 20, 12, 138, 253, 25, 64, 4, 17820, 151, 15179],\n",
       " [415, 46, 1728, 51732, 98416],\n",
       " [1, 114, 124, 48, 442, 76, 12, 1297, 2, 425, 4616, 15, 43605, 2, 425, 58, 99],\n",
       " [20,\n",
       "  12,\n",
       "  257,\n",
       "  2,\n",
       "  1574,\n",
       "  7,\n",
       "  3,\n",
       "  63,\n",
       "  9,\n",
       "  7,\n",
       "  16,\n",
       "  521,\n",
       "  154,\n",
       "  431,\n",
       "  496,\n",
       "  204,\n",
       "  629,\n",
       "  62,\n",
       "  6,\n",
       "  23,\n",
       "  7,\n",
       "  20,\n",
       "  396,\n",
       "  96,\n",
       "  26,\n",
       "  23,\n",
       "  12,\n",
       "  6,\n",
       "  7,\n",
       "  40,\n",
       "  40668,\n",
       "  530,\n",
       "  14,\n",
       "  22,\n",
       "  39,\n",
       "  80,\n",
       "  91,\n",
       "  188,\n",
       "  3,\n",
       "  32,\n",
       "  49,\n",
       "  2,\n",
       "  214,\n",
       "  7,\n",
       "  2,\n",
       "  152,\n",
       "  5,\n",
       "  2771,\n",
       "  138,\n",
       "  31,\n",
       "  1,\n",
       "  7768,\n",
       "  3683,\n",
       "  3851,\n",
       "  25951,\n",
       "  6,\n",
       "  1,\n",
       "  901,\n",
       "  420,\n",
       "  11,\n",
       "  10,\n",
       "  1589,\n",
       "  105,\n",
       "  901,\n",
       "  1265,\n",
       "  785,\n",
       "  11,\n",
       "  59,\n",
       "  25951,\n",
       "  8,\n",
       "  36,\n",
       "  204,\n",
       "  9,\n",
       "  7,\n",
       "  64,\n",
       "  52,\n",
       "  40,\n",
       "  20,\n",
       "  22,\n",
       "  1,\n",
       "  25951,\n",
       "  1496,\n",
       "  98417,\n",
       "  10]]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_sequences[:20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading the pre-trained word embedding.\n",
      "Total 2000000 word vectors.\n",
      "Null word embeddings: 22573\n",
      "Loaded\n"
     ]
    }
   ],
   "source": [
    "print(\"Loading the pre-trained word embedding.\")\n",
    "embeddings_index = data_loader.load_embedding('features/crawl-300d-2M.vec')\n",
    "embedding_matrix = data_transformer.build_embedding_matrix(embeddings_index)\n",
    "print(\"Loaded\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'sotoxic.train.trainer' from '/home/zake7749/PycharmProjects/CCP/sotoxic/train/trainer.py'>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import importlib\n",
    "import sotoxic.models.pytorch.rhn as rhn\n",
    "import sotoxic.train.trainer as trn\n",
    "importlib.reload(rhn)\n",
    "importlib.reload(trn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_recurrent_higway_classifier():\n",
    "    embedding = nn.Embedding(VOCAB_SIZE, EMBEDDING_SIZE)\n",
    "    embedding.weight.data.copy_(torch.from_numpy(embedding_matrix))\n",
    "    embedding.weight.requires_grad=False\n",
    "    return rhn.RecurrentHighwayClassifier(\n",
    "        input_size=EMBEDDING_SIZE,\n",
    "        hidden_size=60, \n",
    "        embedding=embedding,\n",
    "        recurrence_length=2,\n",
    "        recurrent_dropout=0.25)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "trainer = trn.PyTorchModelTrainer(model_stamp=\"FASTTEXT_RHN_64_64\", epoch_num=300, learning_rate=1e-3,\n",
    "                                  verbose_round=80, shuffle_inputs=False, early_stopping_round=12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Choose the torch base model.\n",
      "Training on fold 0\n",
      "Epoch:1 Batch:0 Log-loss0.6764200329780579\n",
      "Epoch:1 Batch:80 Log-loss0.09407692402601242\n",
      "Epoch:1 Batch:160 Log-loss0.09060367941856384\n",
      "Epoch:1 Batch:240 Log-loss0.0713363066315651\n",
      "Epoch:1 Batch:320 Log-loss0.055740322917699814\n",
      "Epoch:1 Batch:400 Log-loss0.0804063081741333\n",
      "Epoch:1 Batch:480 Log-loss0.0346609428524971\n",
      "Epoch:1 Batch:560 Log-loss0.07526189088821411\n",
      "Epoch average log loss:0.084349685122392\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.04926930949574479, best_val_loss:0.04926930949574479, best_auc:0.9753638758367783\n",
      "Epoch:2 Batch:0 Log-loss0.047581274062395096\n",
      "Epoch:2 Batch:80 Log-loss0.03739865869283676\n",
      "Epoch:2 Batch:160 Log-loss0.02832009084522724\n",
      "Epoch:2 Batch:240 Log-loss0.059971634298563004\n",
      "Epoch:2 Batch:320 Log-loss0.05598978325724602\n",
      "Epoch:2 Batch:400 Log-loss0.05800679326057434\n",
      "Epoch:2 Batch:480 Log-loss0.052891626954078674\n",
      "Epoch:2 Batch:560 Log-loss0.035993900150060654\n",
      "Epoch average log loss:0.0525463173265702\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.04612638649813335, best_val_loss:0.04612638649813335, best_auc:0.9797382016064354\n",
      "Epoch:3 Batch:0 Log-loss0.07543372362852097\n",
      "Epoch:3 Batch:80 Log-loss0.04882766678929329\n",
      "Epoch:3 Batch:160 Log-loss0.054363515228033066\n",
      "Epoch:3 Batch:240 Log-loss0.047235164791345596\n",
      "Epoch:3 Batch:320 Log-loss0.06173953041434288\n",
      "Epoch:3 Batch:400 Log-loss0.06666798144578934\n",
      "Epoch:3 Batch:480 Log-loss0.04930858686566353\n",
      "Epoch:3 Batch:560 Log-loss0.04699297621846199\n",
      "Epoch average log loss:0.04982911641044276\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.04442986261486526, best_val_loss:0.04442986261486526, best_auc:0.9825073813718466\n",
      "Epoch:4 Batch:0 Log-loss0.039802540093660355\n",
      "Epoch:4 Batch:80 Log-loss0.059110600501298904\n",
      "Epoch:4 Batch:160 Log-loss0.0460125207901001\n",
      "Epoch:4 Batch:240 Log-loss0.0500623844563961\n",
      "Epoch:4 Batch:320 Log-loss0.031892646104097366\n",
      "Epoch:4 Batch:400 Log-loss0.054569777101278305\n",
      "Epoch:4 Batch:480 Log-loss0.03769831359386444\n",
      "Epoch:4 Batch:560 Log-loss0.041556309908628464\n",
      "Epoch average log loss:0.04749914327237223\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.042027401884490494, best_val_loss:0.042027401884490494, best_auc:0.9849503597375687\n",
      "Epoch:5 Batch:0 Log-loss0.04480680823326111\n",
      "Epoch:5 Batch:80 Log-loss0.034178946167230606\n",
      "Epoch:5 Batch:160 Log-loss0.05515052378177643\n",
      "Epoch:5 Batch:240 Log-loss0.04813845455646515\n",
      "Epoch:5 Batch:320 Log-loss0.055473726242780685\n",
      "Epoch:5 Batch:400 Log-loss0.03612774983048439\n",
      "Epoch:5 Batch:480 Log-loss0.05094030499458313\n",
      "Epoch:5 Batch:560 Log-loss0.047896113246679306\n",
      "Epoch average log loss:0.045954089844599366\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.04071574760304105, best_val_loss:0.04071574760304105, best_auc:0.9870454816194533\n",
      "Epoch:6 Batch:0 Log-loss0.04592417553067207\n",
      "Epoch:6 Batch:80 Log-loss0.05148691311478615\n",
      "Epoch:6 Batch:160 Log-loss0.03629542514681816\n",
      "Epoch:6 Batch:240 Log-loss0.06964091956615448\n",
      "Epoch:6 Batch:320 Log-loss0.0331006720662117\n",
      "Epoch:6 Batch:400 Log-loss0.05769770219922066\n",
      "Epoch:6 Batch:480 Log-loss0.03334672376513481\n",
      "Epoch:6 Batch:560 Log-loss0.04764517396688461\n",
      "Epoch average log loss:0.044654396429125755\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.04001704288960626, best_val_loss:0.04001704288960626, best_auc:0.9882854413382541\n",
      "Epoch:7 Batch:0 Log-loss0.029853269457817078\n",
      "Epoch:7 Batch:80 Log-loss0.033736906945705414\n",
      "Epoch:7 Batch:160 Log-loss0.03548197075724602\n",
      "Epoch:7 Batch:240 Log-loss0.05883863940834999\n",
      "Epoch:7 Batch:320 Log-loss0.0485190711915493\n",
      "Epoch:7 Batch:400 Log-loss0.03409387916326523\n",
      "Epoch:7 Batch:480 Log-loss0.04048483446240425\n",
      "Epoch:7 Batch:560 Log-loss0.04065375030040741\n",
      "Epoch average log loss:0.043201821892788365\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.03904053562464595, best_val_loss:0.03904053562464595, best_auc:0.9892711468347871\n",
      "Epoch:8 Batch:0 Log-loss0.04094434157013893\n",
      "Epoch:8 Batch:80 Log-loss0.037278275936841965\n",
      "Epoch:8 Batch:160 Log-loss0.04122139886021614\n",
      "Epoch:8 Batch:240 Log-loss0.04711175337433815\n",
      "Epoch:8 Batch:320 Log-loss0.04756763577461243\n",
      "Epoch:8 Batch:400 Log-loss0.029176458716392517\n",
      "Epoch:8 Batch:480 Log-loss0.035748064517974854\n",
      "Epoch:8 Batch:560 Log-loss0.04836246743798256\n",
      "Epoch average log loss:0.042695283996207374\n",
      "In Epoch8, val_loss:0.03916865664410435, best_val_loss:0.03904053562464595, best_auc:0.9892711468347871\n",
      "Epoch:9 Batch:0 Log-loss0.043418336659669876\n",
      "Epoch:9 Batch:80 Log-loss0.04289316013455391\n",
      "Epoch:9 Batch:160 Log-loss0.04656998813152313\n",
      "Epoch:9 Batch:240 Log-loss0.04370303452014923\n",
      "Epoch:9 Batch:320 Log-loss0.02825174294412136\n",
      "Epoch:9 Batch:400 Log-loss0.05991822108626366\n",
      "Epoch:9 Batch:480 Log-loss0.03249562904238701\n",
      "Epoch:9 Batch:560 Log-loss0.03251785412430763\n",
      "Epoch average log loss:0.04206259113603405\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.03876278873239226, best_val_loss:0.03876278873239226, best_auc:0.9897019018201202\n",
      "Epoch:10 Batch:0 Log-loss0.04228666424751282\n",
      "Epoch:10 Batch:80 Log-loss0.04314977303147316\n",
      "Epoch:10 Batch:160 Log-loss0.034742891788482666\n",
      "Epoch:10 Batch:240 Log-loss0.04812990128993988\n",
      "Epoch:10 Batch:320 Log-loss0.04136407747864723\n",
      "Epoch:10 Batch:400 Log-loss0.025030769407749176\n",
      "Epoch:10 Batch:480 Log-loss0.03244735673069954\n",
      "Epoch:10 Batch:560 Log-loss0.04643313214182854\n",
      "Epoch average log loss:0.04147509969105678\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.03851200443300018, best_val_loss:0.03851200443300018, best_auc:0.9901377008404991\n",
      "Epoch:11 Batch:0 Log-loss0.04146170616149902\n",
      "Epoch:11 Batch:80 Log-loss0.053498104214668274\n",
      "Epoch:11 Batch:160 Log-loss0.040703464299440384\n",
      "Epoch:11 Batch:240 Log-loss0.04785778746008873\n",
      "Epoch:11 Batch:320 Log-loss0.04854751005768776\n",
      "Epoch:11 Batch:400 Log-loss0.03147374838590622\n",
      "Epoch:11 Batch:480 Log-loss0.04659843444824219\n",
      "Epoch:11 Batch:560 Log-loss0.04250236228108406\n",
      "Epoch average log loss:0.04126149103644171\n",
      "In Epoch11, val_loss:0.038879015785034575, best_val_loss:0.03851200443300018, best_auc:0.9901377008404991\n",
      "Epoch:12 Batch:0 Log-loss0.03712557628750801\n",
      "Epoch:12 Batch:80 Log-loss0.041291918605566025\n",
      "Epoch:12 Batch:160 Log-loss0.08089765161275864\n",
      "Epoch:12 Batch:240 Log-loss0.03415997698903084\n",
      "Epoch:12 Batch:320 Log-loss0.05799022316932678\n",
      "Epoch:12 Batch:400 Log-loss0.03835301101207733\n",
      "Epoch:12 Batch:480 Log-loss0.04691047593951225\n",
      "Epoch:12 Batch:560 Log-loss0.03046753630042076\n",
      "Epoch average log loss:0.0408158015592822\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.03794116695033534, best_val_loss:0.03794116695033534, best_auc:0.9901346830293855\n",
      "Epoch:13 Batch:0 Log-loss0.05102552846074104\n",
      "Epoch:13 Batch:80 Log-loss0.02731751836836338\n",
      "Epoch:13 Batch:160 Log-loss0.03600749745965004\n",
      "Epoch:13 Batch:240 Log-loss0.03918173164129257\n",
      "Epoch:13 Batch:320 Log-loss0.04957057535648346\n",
      "Epoch:13 Batch:400 Log-loss0.044353459030389786\n",
      "Epoch:13 Batch:480 Log-loss0.050627049058675766\n",
      "Epoch:13 Batch:560 Log-loss0.04249121621251106\n",
      "Epoch average log loss:0.04051390608905681\n",
      "In Epoch13, val_loss:0.03801531511161917, best_val_loss:0.03794116695033534, best_auc:0.9901346830293855\n",
      "Epoch:14 Batch:0 Log-loss0.036575231701135635\n",
      "Epoch:14 Batch:80 Log-loss0.032333336770534515\n",
      "Epoch:14 Batch:160 Log-loss0.049467410892248154\n",
      "Epoch:14 Batch:240 Log-loss0.040553562343120575\n",
      "Epoch:14 Batch:320 Log-loss0.04318680986762047\n",
      "Epoch:14 Batch:400 Log-loss0.03236261382699013\n",
      "Epoch:14 Batch:480 Log-loss0.06088517978787422\n",
      "Epoch:14 Batch:560 Log-loss0.04661327227950096\n",
      "Epoch average log loss:0.04027959303930402\n",
      "In Epoch14, val_loss:0.03797812082417173, best_val_loss:0.03794116695033534, best_auc:0.9901346830293855\n",
      "Epoch:15 Batch:0 Log-loss0.04220352694392204\n",
      "Epoch:15 Batch:80 Log-loss0.05214868113398552\n",
      "Epoch:15 Batch:160 Log-loss0.03501560539007187\n",
      "Epoch:15 Batch:240 Log-loss0.04288938641548157\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:15 Batch:320 Log-loss0.030456768348813057\n",
      "Epoch:15 Batch:400 Log-loss0.0304322000592947\n",
      "Epoch:15 Batch:480 Log-loss0.04947105050086975\n",
      "Epoch:15 Batch:560 Log-loss0.03121676668524742\n",
      "Epoch average log loss:0.03985342569981835\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch15, val_loss:0.03766852826388454, best_val_loss:0.03766852826388454, best_auc:0.990284062556752\n",
      "Epoch:16 Batch:0 Log-loss0.03523236885666847\n",
      "Epoch:16 Batch:80 Log-loss0.04171955958008766\n",
      "Epoch:16 Batch:160 Log-loss0.03461119160056114\n",
      "Epoch:16 Batch:240 Log-loss0.048311617225408554\n",
      "Epoch:16 Batch:320 Log-loss0.038581714034080505\n",
      "Epoch:16 Batch:400 Log-loss0.031511932611465454\n",
      "Epoch:16 Batch:480 Log-loss0.03388978913426399\n",
      "Epoch:16 Batch:560 Log-loss0.04285668209195137\n",
      "Epoch average log loss:0.039643369408856544\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch16, val_loss:0.03758368035311024, best_val_loss:0.03758368035311024, best_auc:0.9901440471698403\n",
      "Epoch:17 Batch:0 Log-loss0.032700762152671814\n",
      "Epoch:17 Batch:80 Log-loss0.02879558503627777\n",
      "Epoch:17 Batch:160 Log-loss0.03951933979988098\n",
      "Epoch:17 Batch:240 Log-loss0.040075939148664474\n",
      "Epoch:17 Batch:320 Log-loss0.03330705687403679\n",
      "Epoch:17 Batch:400 Log-loss0.031178727746009827\n",
      "Epoch:17 Batch:480 Log-loss0.03478436917066574\n",
      "Epoch:17 Batch:560 Log-loss0.035119540989398956\n",
      "Epoch average log loss:0.03956325358178999\n",
      "In Epoch17, val_loss:0.03825140556539517, best_val_loss:0.03758368035311024, best_auc:0.9901440471698403\n",
      "Epoch:18 Batch:0 Log-loss0.03675508499145508\n",
      "Epoch:18 Batch:80 Log-loss0.04728022590279579\n",
      "Epoch:18 Batch:160 Log-loss0.03821094334125519\n",
      "Epoch:18 Batch:240 Log-loss0.05057370290160179\n",
      "Epoch:18 Batch:320 Log-loss0.04730428382754326\n",
      "Epoch:18 Batch:400 Log-loss0.04004756361246109\n",
      "Epoch:18 Batch:480 Log-loss0.037844911217689514\n",
      "Epoch:18 Batch:560 Log-loss0.032352812588214874\n",
      "Epoch average log loss:0.03915557430924049\n",
      "In Epoch18, val_loss:0.0376485194814342, best_val_loss:0.03758368035311024, best_auc:0.9901440471698403\n",
      "Epoch:19 Batch:0 Log-loss0.024929923936724663\n",
      "Epoch:19 Batch:80 Log-loss0.05298177897930145\n",
      "Epoch:19 Batch:160 Log-loss0.03634621575474739\n",
      "Epoch:19 Batch:240 Log-loss0.030359650030732155\n",
      "Epoch:19 Batch:320 Log-loss0.036767106503248215\n",
      "Epoch:19 Batch:400 Log-loss0.03607554733753204\n",
      "Epoch:19 Batch:480 Log-loss0.02750535123050213\n",
      "Epoch:19 Batch:560 Log-loss0.0413665808737278\n",
      "Epoch average log loss:0.03921299163318638\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch19, val_loss:0.037267903869533166, best_val_loss:0.037267903869533166, best_auc:0.990399690513044\n",
      "Epoch:20 Batch:0 Log-loss0.034237802028656006\n",
      "Epoch:20 Batch:80 Log-loss0.05007334426045418\n",
      "Epoch:20 Batch:160 Log-loss0.033780600875616074\n",
      "Epoch:20 Batch:240 Log-loss0.03857364133000374\n",
      "Epoch:20 Batch:320 Log-loss0.05805135890841484\n",
      "Epoch:20 Batch:400 Log-loss0.05021004378795624\n",
      "Epoch:20 Batch:480 Log-loss0.058453842997550964\n",
      "Epoch:20 Batch:560 Log-loss0.034603990614414215\n",
      "Epoch average log loss:0.03874110936054161\n",
      "In Epoch20, val_loss:0.037582170460810756, best_val_loss:0.037267903869533166, best_auc:0.990399690513044\n",
      "Epoch:21 Batch:0 Log-loss0.047929417341947556\n",
      "Epoch:21 Batch:80 Log-loss0.04436241462826729\n",
      "Epoch:21 Batch:160 Log-loss0.04364759847521782\n",
      "Epoch:21 Batch:240 Log-loss0.035156525671482086\n",
      "Epoch:21 Batch:320 Log-loss0.025918342173099518\n",
      "Epoch:21 Batch:400 Log-loss0.03881867229938507\n",
      "Epoch:21 Batch:480 Log-loss0.023292921483516693\n",
      "Epoch:21 Batch:560 Log-loss0.027374932542443275\n",
      "Epoch average log loss:0.0387386004540271\n",
      "In Epoch21, val_loss:0.03753228258548166, best_val_loss:0.037267903869533166, best_auc:0.990399690513044\n",
      "Epoch:22 Batch:0 Log-loss0.05115697160363197\n",
      "Epoch:22 Batch:80 Log-loss0.047385167330503464\n",
      "Epoch:22 Batch:160 Log-loss0.029025472700595856\n",
      "Epoch:22 Batch:240 Log-loss0.0546819269657135\n",
      "Epoch:22 Batch:320 Log-loss0.03428877517580986\n",
      "Epoch:22 Batch:400 Log-loss0.034371282905340195\n",
      "Epoch:22 Batch:480 Log-loss0.041002240031957626\n",
      "Epoch:22 Batch:560 Log-loss0.03473769873380661\n",
      "Epoch average log loss:0.03850810405731733\n",
      "In Epoch22, val_loss:0.03736843508519175, best_val_loss:0.037267903869533166, best_auc:0.990399690513044\n",
      "Epoch:23 Batch:0 Log-loss0.03158576413989067\n",
      "Epoch:23 Batch:80 Log-loss0.048538610339164734\n",
      "Epoch:23 Batch:160 Log-loss0.0354345478117466\n",
      "Epoch:23 Batch:240 Log-loss0.050980400294065475\n",
      "Epoch:23 Batch:320 Log-loss0.043934594839811325\n",
      "Epoch:23 Batch:400 Log-loss0.03882073983550072\n",
      "Epoch:23 Batch:480 Log-loss0.037687547504901886\n",
      "Epoch:23 Batch:560 Log-loss0.04900965839624405\n",
      "Epoch average log loss:0.03833134936700974\n",
      "In Epoch23, val_loss:0.03756084080039639, best_val_loss:0.037267903869533166, best_auc:0.990399690513044\n",
      "Epoch:24 Batch:0 Log-loss0.02768946997821331\n",
      "Epoch:24 Batch:80 Log-loss0.031722404062747955\n",
      "Epoch:24 Batch:160 Log-loss0.021872369572520256\n",
      "Epoch:24 Batch:240 Log-loss0.03798697516322136\n",
      "Epoch:24 Batch:320 Log-loss0.037174519151449203\n",
      "Epoch:24 Batch:400 Log-loss0.026079200208187103\n",
      "Epoch:24 Batch:480 Log-loss0.03231441602110863\n",
      "Epoch:24 Batch:560 Log-loss0.02602890320122242\n",
      "Epoch average log loss:0.03841097147004413\n",
      "In Epoch24, val_loss:0.03735023559702597, best_val_loss:0.037267903869533166, best_auc:0.990399690513044\n",
      "Epoch:25 Batch:0 Log-loss0.04134552553296089\n",
      "Epoch:25 Batch:80 Log-loss0.030070418491959572\n",
      "Epoch:25 Batch:160 Log-loss0.04004358872771263\n",
      "Epoch:25 Batch:240 Log-loss0.03437092527747154\n",
      "Epoch:25 Batch:320 Log-loss0.029749682173132896\n",
      "Epoch:25 Batch:400 Log-loss0.029591700062155724\n",
      "Epoch:25 Batch:480 Log-loss0.032342951744794846\n",
      "Epoch:25 Batch:560 Log-loss0.04937488213181496\n",
      "Epoch average log loss:0.03832612745796463\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch25, val_loss:0.03725959461773621, best_val_loss:0.03725959461773621, best_auc:0.9905751249641485\n",
      "Epoch:26 Batch:0 Log-loss0.0320991650223732\n",
      "Epoch:26 Batch:80 Log-loss0.02465318702161312\n",
      "Epoch:26 Batch:160 Log-loss0.04539009556174278\n",
      "Epoch:26 Batch:240 Log-loss0.0364583395421505\n",
      "Epoch:26 Batch:320 Log-loss0.03470206260681152\n",
      "Epoch:26 Batch:400 Log-loss0.021839803084731102\n",
      "Epoch:26 Batch:480 Log-loss0.02721996046602726\n",
      "Epoch:26 Batch:560 Log-loss0.04744965583086014\n",
      "Epoch average log loss:0.03815448995779402\n",
      "In Epoch26, val_loss:0.03748247352951362, best_val_loss:0.03725959461773621, best_auc:0.9905751249641485\n",
      "Epoch:27 Batch:0 Log-loss0.05059082806110382\n",
      "Epoch:27 Batch:80 Log-loss0.03617523983120918\n",
      "Epoch:27 Batch:160 Log-loss0.03388464078307152\n",
      "Epoch:27 Batch:240 Log-loss0.045082781463861465\n",
      "Epoch:27 Batch:320 Log-loss0.039641112089157104\n",
      "Epoch:27 Batch:400 Log-loss0.026605019345879555\n",
      "Epoch:27 Batch:480 Log-loss0.03162379935383797\n",
      "Epoch:27 Batch:560 Log-loss0.02381027862429619\n",
      "Epoch average log loss:0.03809526336845011\n",
      "In Epoch27, val_loss:0.03728883046542259, best_val_loss:0.03725959461773621, best_auc:0.9905751249641485\n",
      "Epoch:28 Batch:0 Log-loss0.027494648471474648\n",
      "Epoch:28 Batch:80 Log-loss0.0328642912209034\n",
      "Epoch:28 Batch:160 Log-loss0.041541460901498795\n",
      "Epoch:28 Batch:240 Log-loss0.02953500859439373\n",
      "Epoch:28 Batch:320 Log-loss0.05289639160037041\n",
      "Epoch:28 Batch:400 Log-loss0.03328942880034447\n",
      "Epoch:28 Batch:480 Log-loss0.03690456226468086\n",
      "Epoch:28 Batch:560 Log-loss0.029628336429595947\n",
      "Epoch average log loss:0.037877332658639976\n",
      "In Epoch28, val_loss:0.03726707154300652, best_val_loss:0.03725959461773621, best_auc:0.9905751249641485\n",
      "Epoch:29 Batch:0 Log-loss0.03290620818734169\n",
      "Epoch:29 Batch:80 Log-loss0.047255802899599075\n",
      "Epoch:29 Batch:160 Log-loss0.05807851254940033\n",
      "Epoch:29 Batch:240 Log-loss0.046099480241537094\n",
      "Epoch:29 Batch:320 Log-loss0.03777880594134331\n",
      "Epoch:29 Batch:400 Log-loss0.024298369884490967\n",
      "Epoch:29 Batch:480 Log-loss0.03934828191995621\n",
      "Epoch:29 Batch:560 Log-loss0.044723477214574814\n",
      "Epoch average log loss:0.03771931560976165\n",
      "In Epoch29, val_loss:0.03727894076481766, best_val_loss:0.03725959461773621, best_auc:0.9905751249641485\n",
      "Epoch:30 Batch:0 Log-loss0.04600619152188301\n",
      "Epoch:30 Batch:80 Log-loss0.043856654316186905\n",
      "Epoch:30 Batch:160 Log-loss0.037016626447439194\n",
      "Epoch:30 Batch:240 Log-loss0.03210769221186638\n",
      "Epoch:30 Batch:320 Log-loss0.034346308559179306\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:30 Batch:400 Log-loss0.03785456344485283\n",
      "Epoch:30 Batch:480 Log-loss0.04306377097964287\n",
      "Epoch:30 Batch:560 Log-loss0.03783402219414711\n",
      "Epoch average log loss:0.037760935231511084\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch30, val_loss:0.03717460307236107, best_val_loss:0.03717460307236107, best_auc:0.9905650709821402\n",
      "Epoch:31 Batch:0 Log-loss0.03407754749059677\n",
      "Epoch:31 Batch:80 Log-loss0.03229319304227829\n",
      "Epoch:31 Batch:160 Log-loss0.04111488163471222\n",
      "Epoch:31 Batch:240 Log-loss0.03453874588012695\n",
      "Epoch:31 Batch:320 Log-loss0.04419414699077606\n",
      "Epoch:31 Batch:400 Log-loss0.022545315325260162\n",
      "Epoch:31 Batch:480 Log-loss0.0527334101498127\n",
      "Epoch:31 Batch:560 Log-loss0.03118349239230156\n",
      "Epoch average log loss:0.0374804664968646\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch31, val_loss:0.037053314777036495, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:32 Batch:0 Log-loss0.03002392314374447\n",
      "Epoch:32 Batch:80 Log-loss0.03180627152323723\n",
      "Epoch:32 Batch:160 Log-loss0.041562750935554504\n",
      "Epoch:32 Batch:240 Log-loss0.040453020483255386\n",
      "Epoch:32 Batch:320 Log-loss0.046540383249521255\n",
      "Epoch:32 Batch:400 Log-loss0.04631750285625458\n",
      "Epoch:32 Batch:480 Log-loss0.04207080230116844\n",
      "Epoch:32 Batch:560 Log-loss0.03805829957127571\n",
      "Epoch average log loss:0.037372083677577654\n",
      "In Epoch32, val_loss:0.037352713915214455, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:33 Batch:0 Log-loss0.03571286424994469\n",
      "Epoch:33 Batch:80 Log-loss0.03244723752140999\n",
      "Epoch:33 Batch:160 Log-loss0.04622158408164978\n",
      "Epoch:33 Batch:240 Log-loss0.03204644098877907\n",
      "Epoch:33 Batch:320 Log-loss0.0406779944896698\n",
      "Epoch:33 Batch:400 Log-loss0.036384496837854385\n",
      "Epoch:33 Batch:480 Log-loss0.03604878857731819\n",
      "Epoch:33 Batch:560 Log-loss0.03552064672112465\n",
      "Epoch average log loss:0.03761971047081585\n",
      "In Epoch33, val_loss:0.03720720155294187, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:34 Batch:0 Log-loss0.04851221665740013\n",
      "Epoch:34 Batch:80 Log-loss0.031270239502191544\n",
      "Epoch:34 Batch:160 Log-loss0.035059090703725815\n",
      "Epoch:34 Batch:240 Log-loss0.03978835418820381\n",
      "Epoch:34 Batch:320 Log-loss0.033714089542627335\n",
      "Epoch:34 Batch:400 Log-loss0.05361483618617058\n",
      "Epoch:34 Batch:480 Log-loss0.02664881944656372\n",
      "Epoch:34 Batch:560 Log-loss0.0485730916261673\n",
      "Epoch average log loss:0.037691329223369915\n",
      "In Epoch34, val_loss:0.03720132165064912, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:35 Batch:0 Log-loss0.034390099346637726\n",
      "Epoch:35 Batch:80 Log-loss0.034696053713560104\n",
      "Epoch:35 Batch:160 Log-loss0.034581899642944336\n",
      "Epoch:35 Batch:240 Log-loss0.04639191925525665\n",
      "Epoch:35 Batch:320 Log-loss0.030271196737885475\n",
      "Epoch:35 Batch:400 Log-loss0.026636535301804543\n",
      "Epoch:35 Batch:480 Log-loss0.057059675455093384\n",
      "Epoch:35 Batch:560 Log-loss0.041622232645750046\n",
      "Epoch average log loss:0.037603429428833934\n",
      "In Epoch35, val_loss:0.037229388040780055, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:36 Batch:0 Log-loss0.03321629390120506\n",
      "Epoch:36 Batch:80 Log-loss0.02217547595500946\n",
      "Epoch:36 Batch:160 Log-loss0.03575199842453003\n",
      "Epoch:36 Batch:240 Log-loss0.05054885521531105\n",
      "Epoch:36 Batch:320 Log-loss0.03659089654684067\n",
      "Epoch:36 Batch:400 Log-loss0.03731194883584976\n",
      "Epoch:36 Batch:480 Log-loss0.03526708856225014\n",
      "Epoch:36 Batch:560 Log-loss0.025951135903596878\n",
      "Epoch average log loss:0.037446620032590416\n",
      "In Epoch36, val_loss:0.03713843954224116, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:37 Batch:0 Log-loss0.043581772595644\n",
      "Epoch:37 Batch:80 Log-loss0.043725743889808655\n",
      "Epoch:37 Batch:160 Log-loss0.0368051752448082\n",
      "Epoch:37 Batch:240 Log-loss0.03364334627985954\n",
      "Epoch:37 Batch:320 Log-loss0.05272936448454857\n",
      "Epoch:37 Batch:400 Log-loss0.03895464166998863\n",
      "Epoch:37 Batch:480 Log-loss0.03659231960773468\n",
      "Epoch:37 Batch:560 Log-loss0.038140539079904556\n",
      "Epoch average log loss:0.03737750336128686\n",
      "In Epoch37, val_loss:0.037168635334601574, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:38 Batch:0 Log-loss0.026306241750717163\n",
      "Epoch:38 Batch:80 Log-loss0.02992173098027706\n",
      "Epoch:38 Batch:160 Log-loss0.028798798099160194\n",
      "Epoch:38 Batch:240 Log-loss0.04359207674860954\n",
      "Epoch:38 Batch:320 Log-loss0.04920399188995361\n",
      "Epoch:38 Batch:400 Log-loss0.05222843587398529\n",
      "Epoch:38 Batch:480 Log-loss0.04216761514544487\n",
      "Epoch:38 Batch:560 Log-loss0.02712905779480934\n",
      "Epoch average log loss:0.037467412214859254\n",
      "In Epoch38, val_loss:0.0371574252692068, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:39 Batch:0 Log-loss0.04335412010550499\n",
      "Epoch:39 Batch:80 Log-loss0.04821021854877472\n",
      "Epoch:39 Batch:160 Log-loss0.03850442171096802\n",
      "Epoch:39 Batch:240 Log-loss0.046786416321992874\n",
      "Epoch:39 Batch:320 Log-loss0.021278807893395424\n",
      "Epoch:39 Batch:400 Log-loss0.04383312538266182\n",
      "Epoch:39 Batch:480 Log-loss0.033819179981946945\n",
      "Epoch:39 Batch:560 Log-loss0.05293872207403183\n",
      "Epoch average log loss:0.037137713612589454\n",
      "In Epoch39, val_loss:0.03712907108363973, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:40 Batch:0 Log-loss0.03758209943771362\n",
      "Epoch:40 Batch:80 Log-loss0.03181033208966255\n",
      "Epoch:40 Batch:160 Log-loss0.05496115982532501\n",
      "Epoch:40 Batch:240 Log-loss0.04450702667236328\n",
      "Epoch:40 Batch:320 Log-loss0.04095461219549179\n",
      "Epoch:40 Batch:400 Log-loss0.0347493477165699\n",
      "Epoch:40 Batch:480 Log-loss0.03965042158961296\n",
      "Epoch:40 Batch:560 Log-loss0.03862376511096954\n",
      "Epoch average log loss:0.03725563861163599\n",
      "In Epoch40, val_loss:0.037154928622577625, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:41 Batch:0 Log-loss0.033861469477415085\n",
      "Epoch:41 Batch:80 Log-loss0.044767867773771286\n",
      "Epoch:41 Batch:160 Log-loss0.03917510434985161\n",
      "Epoch:41 Batch:240 Log-loss0.047185901552438736\n",
      "Epoch:41 Batch:320 Log-loss0.048432666808366776\n",
      "Epoch:41 Batch:400 Log-loss0.03713146597146988\n",
      "Epoch:41 Batch:480 Log-loss0.036028701812028885\n",
      "Epoch:41 Batch:560 Log-loss0.040290478616952896\n",
      "Epoch average log loss:0.03709756258342947\n",
      "In Epoch41, val_loss:0.037192942878823264, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:42 Batch:0 Log-loss0.03774182125926018\n",
      "Epoch:42 Batch:80 Log-loss0.040579862892627716\n",
      "Epoch:42 Batch:160 Log-loss0.05055759847164154\n",
      "Epoch:42 Batch:240 Log-loss0.03739194944500923\n",
      "Epoch:42 Batch:320 Log-loss0.03608046472072601\n",
      "Epoch:42 Batch:400 Log-loss0.03478196635842323\n",
      "Epoch:42 Batch:480 Log-loss0.034119319170713425\n",
      "Epoch:42 Batch:560 Log-loss0.03807288035750389\n",
      "Epoch average log loss:0.0372748884937859\n",
      "In Epoch42, val_loss:0.03718166056658168, best_val_loss:0.037053314777036495, best_auc:0.9905358257982354\n",
      "Epoch:43 Batch:0 Log-loss0.05145172402262688\n",
      "Epoch:43 Batch:80 Log-loss0.044457316398620605\n",
      "Epoch:43 Batch:160 Log-loss0.036253366619348526\n",
      "Epoch:43 Batch:240 Log-loss0.04659366235136986\n",
      "Epoch:43 Batch:320 Log-loss0.03872975334525108\n",
      "Epoch:43 Batch:400 Log-loss0.032819394022226334\n",
      "Epoch:43 Batch:480 Log-loss0.03846406191587448\n",
      "Epoch:43 Batch:560 Log-loss0.04788227379322052\n",
      "Epoch average log loss:0.03720481875352562\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_640.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 1\n",
      "Epoch:1 Batch:0 Log-loss0.68040531873703\n",
      "Epoch:1 Batch:80 Log-loss0.10298522561788559\n",
      "Epoch:1 Batch:160 Log-loss0.06282658129930496\n",
      "Epoch:1 Batch:240 Log-loss0.06139310821890831\n",
      "Epoch:1 Batch:320 Log-loss0.04335877671837807\n",
      "Epoch:1 Batch:400 Log-loss0.05434329807758331\n",
      "Epoch:1 Batch:480 Log-loss0.04293043911457062\n",
      "Epoch:1 Batch:560 Log-loss0.05770263820886612\n",
      "Epoch average log loss:0.08422579417736935\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.05111843037142374, best_val_loss:0.05111843037142374, best_auc:0.9728146245265809\n",
      "Epoch:2 Batch:0 Log-loss0.03986598923802376\n",
      "Epoch:2 Batch:80 Log-loss0.05376465618610382\n",
      "Epoch:2 Batch:160 Log-loss0.04102429002523422\n",
      "Epoch:2 Batch:240 Log-loss0.058652475476264954\n",
      "Epoch:2 Batch:320 Log-loss0.04778014495968819\n",
      "Epoch:2 Batch:400 Log-loss0.03809560835361481\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:2 Batch:480 Log-loss0.06170211359858513\n",
      "Epoch:2 Batch:560 Log-loss0.056031085550785065\n",
      "Epoch average log loss:0.05337644493473428\n",
      "In Epoch2, val_loss:0.05235306843447438, best_val_loss:0.05111843037142374, best_auc:0.9728146245265809\n",
      "Epoch:3 Batch:0 Log-loss0.05006115138530731\n",
      "Epoch:3 Batch:80 Log-loss0.05657699704170227\n",
      "Epoch:3 Batch:160 Log-loss0.04490293189883232\n",
      "Epoch:3 Batch:240 Log-loss0.03619351238012314\n",
      "Epoch:3 Batch:320 Log-loss0.07174747437238693\n",
      "Epoch:3 Batch:400 Log-loss0.04430321231484413\n",
      "Epoch:3 Batch:480 Log-loss0.05191109701991081\n",
      "Epoch:3 Batch:560 Log-loss0.041987329721450806\n",
      "Epoch average log loss:0.05025121001194098\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.04729695975633182, best_val_loss:0.04729695975633182, best_auc:0.977834065065673\n",
      "Epoch:4 Batch:0 Log-loss0.07252666354179382\n",
      "Epoch:4 Batch:80 Log-loss0.04659726843237877\n",
      "Epoch:4 Batch:160 Log-loss0.06561022251844406\n",
      "Epoch:4 Batch:240 Log-loss0.035077840089797974\n",
      "Epoch:4 Batch:320 Log-loss0.04080243781208992\n",
      "Epoch:4 Batch:400 Log-loss0.04680924117565155\n",
      "Epoch:4 Batch:480 Log-loss0.06649228185415268\n",
      "Epoch:4 Batch:560 Log-loss0.03508764132857323\n",
      "Epoch average log loss:0.04787512851014201\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.04546401137941824, best_val_loss:0.04546401137941824, best_auc:0.9804340964284456\n",
      "Epoch:5 Batch:0 Log-loss0.02574349194765091\n",
      "Epoch:5 Batch:80 Log-loss0.04704449698328972\n",
      "Epoch:5 Batch:160 Log-loss0.039885807782411575\n",
      "Epoch:5 Batch:240 Log-loss0.047452691942453384\n",
      "Epoch:5 Batch:320 Log-loss0.06685565412044525\n",
      "Epoch:5 Batch:400 Log-loss0.045626431703567505\n",
      "Epoch:5 Batch:480 Log-loss0.0384640097618103\n",
      "Epoch:5 Batch:560 Log-loss0.04943135380744934\n",
      "Epoch average log loss:0.04681213934839304\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.04414090868596412, best_val_loss:0.04414090868596412, best_auc:0.9827767252508027\n",
      "Epoch:6 Batch:0 Log-loss0.046746764332056046\n",
      "Epoch:6 Batch:80 Log-loss0.0494605153799057\n",
      "Epoch:6 Batch:160 Log-loss0.047802578657865524\n",
      "Epoch:6 Batch:240 Log-loss0.05385662987828255\n",
      "Epoch:6 Batch:320 Log-loss0.048157770186662674\n",
      "Epoch:6 Batch:400 Log-loss0.041071586310863495\n",
      "Epoch:6 Batch:480 Log-loss0.048039715737104416\n",
      "Epoch:6 Batch:560 Log-loss0.06152864545583725\n",
      "Epoch average log loss:0.045414053588839515\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.04370172354155502, best_val_loss:0.04370172354155502, best_auc:0.9842401904518284\n",
      "Epoch:7 Batch:0 Log-loss0.08000490814447403\n",
      "Epoch:7 Batch:80 Log-loss0.032432105392217636\n",
      "Epoch:7 Batch:160 Log-loss0.037247005850076675\n",
      "Epoch:7 Batch:240 Log-loss0.04121160879731178\n",
      "Epoch:7 Batch:320 Log-loss0.02823188155889511\n",
      "Epoch:7 Batch:400 Log-loss0.044714752584695816\n",
      "Epoch:7 Batch:480 Log-loss0.0362713448703289\n",
      "Epoch:7 Batch:560 Log-loss0.03567473217844963\n",
      "Epoch average log loss:0.04422570667562208\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.042770775646477306, best_val_loss:0.042770775646477306, best_auc:0.9863475901890899\n",
      "Epoch:8 Batch:0 Log-loss0.041321635246276855\n",
      "Epoch:8 Batch:80 Log-loss0.0473940335214138\n",
      "Epoch:8 Batch:160 Log-loss0.0432409904897213\n",
      "Epoch:8 Batch:240 Log-loss0.02834255062043667\n",
      "Epoch:8 Batch:320 Log-loss0.06384647637605667\n",
      "Epoch:8 Batch:400 Log-loss0.05821917578577995\n",
      "Epoch:8 Batch:480 Log-loss0.043230313807725906\n",
      "Epoch:8 Batch:560 Log-loss0.05409863963723183\n",
      "Epoch average log loss:0.043754592444747685\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.042421863490850635, best_val_loss:0.042421863490850635, best_auc:0.9863476678583624\n",
      "Epoch:9 Batch:0 Log-loss0.033915627747774124\n",
      "Epoch:9 Batch:80 Log-loss0.032376836985349655\n",
      "Epoch:9 Batch:160 Log-loss0.02701421082019806\n",
      "Epoch:9 Batch:240 Log-loss0.03711366280913353\n",
      "Epoch:9 Batch:320 Log-loss0.05754483863711357\n",
      "Epoch:9 Batch:400 Log-loss0.04717816784977913\n",
      "Epoch:9 Batch:480 Log-loss0.044930215924978256\n",
      "Epoch:9 Batch:560 Log-loss0.040020253509283066\n",
      "Epoch average log loss:0.04251104575648371\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.04186073761445113, best_val_loss:0.04186073761445113, best_auc:0.9871809354493738\n",
      "Epoch:10 Batch:0 Log-loss0.037352148443460464\n",
      "Epoch:10 Batch:80 Log-loss0.05172116681933403\n",
      "Epoch:10 Batch:160 Log-loss0.036038488149642944\n",
      "Epoch:10 Batch:240 Log-loss0.03825180605053902\n",
      "Epoch:10 Batch:320 Log-loss0.0424891822040081\n",
      "Epoch:10 Batch:400 Log-loss0.04354037344455719\n",
      "Epoch:10 Batch:480 Log-loss0.03737623617053032\n",
      "Epoch:10 Batch:560 Log-loss0.03440140560269356\n",
      "Epoch average log loss:0.04237352089091603\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.04123436847809497, best_val_loss:0.04123436847809497, best_auc:0.9877461562421144\n",
      "Epoch:11 Batch:0 Log-loss0.05668654665350914\n",
      "Epoch:11 Batch:80 Log-loss0.026080070063471794\n",
      "Epoch:11 Batch:160 Log-loss0.051517222076654434\n",
      "Epoch:11 Batch:240 Log-loss0.040196165442466736\n",
      "Epoch:11 Batch:320 Log-loss0.046359315514564514\n",
      "Epoch:11 Batch:400 Log-loss0.03194884583353996\n",
      "Epoch:11 Batch:480 Log-loss0.04486441984772682\n",
      "Epoch:11 Batch:560 Log-loss0.032197460532188416\n",
      "Epoch average log loss:0.041835029616153666\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch11, val_loss:0.040990592165298205, best_val_loss:0.040990592165298205, best_auc:0.9881383444432031\n",
      "Epoch:12 Batch:0 Log-loss0.04331113025546074\n",
      "Epoch:12 Batch:80 Log-loss0.04544078931212425\n",
      "Epoch:12 Batch:160 Log-loss0.03762124478816986\n",
      "Epoch:12 Batch:240 Log-loss0.02956393174827099\n",
      "Epoch:12 Batch:320 Log-loss0.03450698032975197\n",
      "Epoch:12 Batch:400 Log-loss0.04610147699713707\n",
      "Epoch:12 Batch:480 Log-loss0.04548020660877228\n",
      "Epoch:12 Batch:560 Log-loss0.03797968104481697\n",
      "Epoch average log loss:0.041492501965590886\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.040903084397610597, best_val_loss:0.040903084397610597, best_auc:0.9889094009041215\n",
      "Epoch:13 Batch:0 Log-loss0.04646982625126839\n",
      "Epoch:13 Batch:80 Log-loss0.04410000517964363\n",
      "Epoch:13 Batch:160 Log-loss0.04394088685512543\n",
      "Epoch:13 Batch:240 Log-loss0.045927342027425766\n",
      "Epoch:13 Batch:320 Log-loss0.032945241779088974\n",
      "Epoch:13 Batch:400 Log-loss0.03736409544944763\n",
      "Epoch:13 Batch:480 Log-loss0.04502715542912483\n",
      "Epoch:13 Batch:560 Log-loss0.0316111221909523\n",
      "Epoch average log loss:0.04109807218025838\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch13, val_loss:0.04019749684520977, best_val_loss:0.04019749684520977, best_auc:0.9891504214667376\n",
      "Epoch:14 Batch:0 Log-loss0.03617684915661812\n",
      "Epoch:14 Batch:80 Log-loss0.035155776888132095\n",
      "Epoch:14 Batch:160 Log-loss0.0315578393638134\n",
      "Epoch:14 Batch:240 Log-loss0.051219891756772995\n",
      "Epoch:14 Batch:320 Log-loss0.057541623711586\n",
      "Epoch:14 Batch:400 Log-loss0.04350334033370018\n",
      "Epoch:14 Batch:480 Log-loss0.03393327072262764\n",
      "Epoch:14 Batch:560 Log-loss0.055478405207395554\n",
      "Epoch average log loss:0.04046662018102195\n",
      "In Epoch14, val_loss:0.04021494737439858, best_val_loss:0.04019749684520977, best_auc:0.9891504214667376\n",
      "Epoch:15 Batch:0 Log-loss0.04117966070771217\n",
      "Epoch:15 Batch:80 Log-loss0.05358746275305748\n",
      "Epoch:15 Batch:160 Log-loss0.040504857897758484\n",
      "Epoch:15 Batch:240 Log-loss0.043265607208013535\n",
      "Epoch:15 Batch:320 Log-loss0.034390438348054886\n",
      "Epoch:15 Batch:400 Log-loss0.04454394057393074\n",
      "Epoch:15 Batch:480 Log-loss0.03479601815342903\n",
      "Epoch:15 Batch:560 Log-loss0.0642038881778717\n",
      "Epoch average log loss:0.04032108976971358\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch15, val_loss:0.03981916391784723, best_val_loss:0.03981916391784723, best_auc:0.9897247756156066\n",
      "Epoch:16 Batch:0 Log-loss0.030551955103874207\n",
      "Epoch:16 Batch:80 Log-loss0.026205172762274742\n",
      "Epoch:16 Batch:160 Log-loss0.04601241275668144\n",
      "Epoch:16 Batch:240 Log-loss0.024560557678341866\n",
      "Epoch:16 Batch:320 Log-loss0.03407248482108116\n",
      "Epoch:16 Batch:400 Log-loss0.0511835478246212\n",
      "Epoch:16 Batch:480 Log-loss0.03537903353571892\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:16 Batch:560 Log-loss0.04825394228100777\n",
      "Epoch average log loss:0.04019081565950598\n",
      "In Epoch16, val_loss:0.04001604260462636, best_val_loss:0.03981916391784723, best_auc:0.9897247756156066\n",
      "Epoch:17 Batch:0 Log-loss0.030554376542568207\n",
      "Epoch:17 Batch:80 Log-loss0.040036942809820175\n",
      "Epoch:17 Batch:160 Log-loss0.03191262483596802\n",
      "Epoch:17 Batch:240 Log-loss0.04790053144097328\n",
      "Epoch:17 Batch:320 Log-loss0.05502623692154884\n",
      "Epoch:17 Batch:400 Log-loss0.0524011105298996\n",
      "Epoch:17 Batch:480 Log-loss0.044097576290369034\n",
      "Epoch:17 Batch:560 Log-loss0.04616418480873108\n",
      "Epoch average log loss:0.03987186909985862\n",
      "In Epoch17, val_loss:0.0402473788965836, best_val_loss:0.03981916391784723, best_auc:0.9897247756156066\n",
      "Epoch:18 Batch:0 Log-loss0.039020586758852005\n",
      "Epoch:18 Batch:80 Log-loss0.035747867077589035\n",
      "Epoch:18 Batch:160 Log-loss0.04417452588677406\n",
      "Epoch:18 Batch:240 Log-loss0.04101055860519409\n",
      "Epoch:18 Batch:320 Log-loss0.03627045452594757\n",
      "Epoch:18 Batch:400 Log-loss0.03318973258137703\n",
      "Epoch:18 Batch:480 Log-loss0.03747634217143059\n",
      "Epoch:18 Batch:560 Log-loss0.030711239203810692\n",
      "Epoch average log loss:0.03955491353491587\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch18, val_loss:0.0397303831135156, best_val_loss:0.0397303831135156, best_auc:0.9899050188758215\n",
      "Epoch:19 Batch:0 Log-loss0.057395219802856445\n",
      "Epoch:19 Batch:80 Log-loss0.044174689799547195\n",
      "Epoch:19 Batch:160 Log-loss0.037151921540498734\n",
      "Epoch:19 Batch:240 Log-loss0.029545361176133156\n",
      "Epoch:19 Batch:320 Log-loss0.02130276896059513\n",
      "Epoch:19 Batch:400 Log-loss0.05157218500971794\n",
      "Epoch:19 Batch:480 Log-loss0.0401759035885334\n",
      "Epoch:19 Batch:560 Log-loss0.03377915546298027\n",
      "Epoch average log loss:0.039685980696231124\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch19, val_loss:0.03972738141380575, best_val_loss:0.03972738141380575, best_auc:0.9900204845900444\n",
      "Epoch:20 Batch:0 Log-loss0.03276200219988823\n",
      "Epoch:20 Batch:80 Log-loss0.027364159002900124\n",
      "Epoch:20 Batch:160 Log-loss0.052782341837882996\n",
      "Epoch:20 Batch:240 Log-loss0.03976357355713844\n",
      "Epoch:20 Batch:320 Log-loss0.04130973294377327\n",
      "Epoch:20 Batch:400 Log-loss0.028296759352087975\n",
      "Epoch:20 Batch:480 Log-loss0.03632228448987007\n",
      "Epoch:20 Batch:560 Log-loss0.050703730434179306\n",
      "Epoch average log loss:0.0391057062195614\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch20, val_loss:0.03962084754619742, best_val_loss:0.03962084754619742, best_auc:0.9899154330059465\n",
      "Epoch:21 Batch:0 Log-loss0.029224982485175133\n",
      "Epoch:21 Batch:80 Log-loss0.04694565013051033\n",
      "Epoch:21 Batch:160 Log-loss0.03439593315124512\n",
      "Epoch:21 Batch:240 Log-loss0.029188642278313637\n",
      "Epoch:21 Batch:320 Log-loss0.030371783301234245\n",
      "Epoch:21 Batch:400 Log-loss0.025474274531006813\n",
      "Epoch:21 Batch:480 Log-loss0.07570493221282959\n",
      "Epoch:21 Batch:560 Log-loss0.03347472473978996\n",
      "Epoch average log loss:0.03921712457002806\n",
      "In Epoch21, val_loss:0.039849902453658693, best_val_loss:0.03962084754619742, best_auc:0.9899154330059465\n",
      "Epoch:22 Batch:0 Log-loss0.032868001610040665\n",
      "Epoch:22 Batch:80 Log-loss0.038992688059806824\n",
      "Epoch:22 Batch:160 Log-loss0.03957522287964821\n",
      "Epoch:22 Batch:240 Log-loss0.05950707197189331\n",
      "Epoch:22 Batch:320 Log-loss0.05313588306307793\n",
      "Epoch:22 Batch:400 Log-loss0.037952374666929245\n",
      "Epoch:22 Batch:480 Log-loss0.02245163917541504\n",
      "Epoch:22 Batch:560 Log-loss0.05228220671415329\n",
      "Epoch average log loss:0.039125044607291264\n",
      "In Epoch22, val_loss:0.03983297376151098, best_val_loss:0.03962084754619742, best_auc:0.9899154330059465\n",
      "Epoch:23 Batch:0 Log-loss0.0415993295609951\n",
      "Epoch:23 Batch:80 Log-loss0.04188942909240723\n",
      "Epoch:23 Batch:160 Log-loss0.03958193585276604\n",
      "Epoch:23 Batch:240 Log-loss0.05643022432923317\n",
      "Epoch:23 Batch:320 Log-loss0.03816458582878113\n",
      "Epoch:23 Batch:400 Log-loss0.04645783081650734\n",
      "Epoch:23 Batch:480 Log-loss0.03365526348352432\n",
      "Epoch:23 Batch:560 Log-loss0.02511320821940899\n",
      "Epoch average log loss:0.03891988236989294\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch23, val_loss:0.039360947302810124, best_val_loss:0.039360947302810124, best_auc:0.9901595843286891\n",
      "Epoch:24 Batch:0 Log-loss0.036629337817430496\n",
      "Epoch:24 Batch:80 Log-loss0.045898571610450745\n",
      "Epoch:24 Batch:160 Log-loss0.0327591709792614\n",
      "Epoch:24 Batch:240 Log-loss0.031660158187150955\n",
      "Epoch:24 Batch:320 Log-loss0.029714560136198997\n",
      "Epoch:24 Batch:400 Log-loss0.032970674335956573\n",
      "Epoch:24 Batch:480 Log-loss0.039848990738391876\n",
      "Epoch:24 Batch:560 Log-loss0.03863950073719025\n",
      "Epoch average log loss:0.038693609185117696\n",
      "In Epoch24, val_loss:0.03956584492316265, best_val_loss:0.039360947302810124, best_auc:0.9901595843286891\n",
      "Epoch:25 Batch:0 Log-loss0.0466376394033432\n",
      "Epoch:25 Batch:80 Log-loss0.04606236144900322\n",
      "Epoch:25 Batch:160 Log-loss0.03999432548880577\n",
      "Epoch:25 Batch:240 Log-loss0.048292454332113266\n",
      "Epoch:25 Batch:320 Log-loss0.041237764060497284\n",
      "Epoch:25 Batch:400 Log-loss0.03125461935997009\n",
      "Epoch:25 Batch:480 Log-loss0.02008323185145855\n",
      "Epoch:25 Batch:560 Log-loss0.028084997087717056\n",
      "Epoch average log loss:0.038642454097446585\n",
      "In Epoch25, val_loss:0.039672325996167636, best_val_loss:0.039360947302810124, best_auc:0.9901595843286891\n",
      "Epoch:26 Batch:0 Log-loss0.04869027063250542\n",
      "Epoch:26 Batch:80 Log-loss0.0404578261077404\n",
      "Epoch:26 Batch:160 Log-loss0.03690177574753761\n",
      "Epoch:26 Batch:240 Log-loss0.05001673102378845\n",
      "Epoch:26 Batch:320 Log-loss0.05791310966014862\n",
      "Epoch:26 Batch:400 Log-loss0.044969748705625534\n",
      "Epoch:26 Batch:480 Log-loss0.047693926841020584\n",
      "Epoch:26 Batch:560 Log-loss0.04012880474328995\n",
      "Epoch average log loss:0.03852035573218018\n",
      "In Epoch26, val_loss:0.03945045986035422, best_val_loss:0.039360947302810124, best_auc:0.9901595843286891\n",
      "Epoch:27 Batch:0 Log-loss0.041447337716817856\n",
      "Epoch:27 Batch:80 Log-loss0.05095571279525757\n",
      "Epoch:27 Batch:160 Log-loss0.03906654939055443\n",
      "Epoch:27 Batch:240 Log-loss0.03328346088528633\n",
      "Epoch:27 Batch:320 Log-loss0.026575550436973572\n",
      "Epoch:27 Batch:400 Log-loss0.02987060882151127\n",
      "Epoch:27 Batch:480 Log-loss0.049197006970644\n",
      "Epoch:27 Batch:560 Log-loss0.0432962030172348\n",
      "Epoch average log loss:0.03853023181935506\n",
      "In Epoch27, val_loss:0.03943056648835309, best_val_loss:0.039360947302810124, best_auc:0.9901595843286891\n",
      "Epoch:28 Batch:0 Log-loss0.021263299509882927\n",
      "Epoch:28 Batch:80 Log-loss0.03557891026139259\n",
      "Epoch:28 Batch:160 Log-loss0.03461156040430069\n",
      "Epoch:28 Batch:240 Log-loss0.04683737829327583\n",
      "Epoch:28 Batch:320 Log-loss0.028816156089305878\n",
      "Epoch:28 Batch:400 Log-loss0.05492144450545311\n",
      "Epoch:28 Batch:480 Log-loss0.0336797721683979\n",
      "Epoch:28 Batch:560 Log-loss0.05023270100355148\n",
      "Epoch average log loss:0.038382820222925926\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch28, val_loss:0.039346948406361285, best_val_loss:0.039346948406361285, best_auc:0.9902301936587047\n",
      "Epoch:29 Batch:0 Log-loss0.037102412432432175\n",
      "Epoch:29 Batch:80 Log-loss0.031240852549672127\n",
      "Epoch:29 Batch:160 Log-loss0.053505364805459976\n",
      "Epoch:29 Batch:240 Log-loss0.03374054655432701\n",
      "Epoch:29 Batch:320 Log-loss0.045943260192871094\n",
      "Epoch:29 Batch:400 Log-loss0.04851953312754631\n",
      "Epoch:29 Batch:480 Log-loss0.05076823756098747\n",
      "Epoch:29 Batch:560 Log-loss0.03820497915148735\n",
      "Epoch average log loss:0.03833161386927324\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch29, val_loss:0.03922579811684825, best_val_loss:0.03922579811684825, best_auc:0.9902905248538839\n",
      "Epoch:30 Batch:0 Log-loss0.05680811032652855\n",
      "Epoch:30 Batch:80 Log-loss0.034635819494724274\n",
      "Epoch:30 Batch:160 Log-loss0.03096146136522293\n",
      "Epoch:30 Batch:240 Log-loss0.043430689722299576\n",
      "Epoch:30 Batch:320 Log-loss0.038097042590379715\n",
      "Epoch:30 Batch:400 Log-loss0.04138702154159546\n",
      "Epoch:30 Batch:480 Log-loss0.058958157896995544\n",
      "Epoch:30 Batch:560 Log-loss0.03568137064576149\n",
      "Epoch average log loss:0.038146275567955204\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch30, val_loss:0.03920176149522164, best_val_loss:0.03920176149522164, best_auc:0.9902525662097995\n",
      "Epoch:31 Batch:0 Log-loss0.05593644082546234\n",
      "Epoch:31 Batch:80 Log-loss0.031532615423202515\n",
      "Epoch:31 Batch:160 Log-loss0.04760415479540825\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:31 Batch:240 Log-loss0.033379945904016495\n",
      "Epoch:31 Batch:320 Log-loss0.036133430898189545\n",
      "Epoch:31 Batch:400 Log-loss0.03633055090904236\n",
      "Epoch:31 Batch:480 Log-loss0.029332110658288002\n",
      "Epoch:31 Batch:560 Log-loss0.04237540066242218\n",
      "Epoch average log loss:0.038091836912956624\n",
      "In Epoch31, val_loss:0.03933979842168355, best_val_loss:0.03920176149522164, best_auc:0.9902525662097995\n",
      "Epoch:32 Batch:0 Log-loss0.04068051278591156\n",
      "Epoch:32 Batch:80 Log-loss0.029142724350094795\n",
      "Epoch:32 Batch:160 Log-loss0.03255143389105797\n",
      "Epoch:32 Batch:240 Log-loss0.03811360150575638\n",
      "Epoch:32 Batch:320 Log-loss0.04262396693229675\n",
      "Epoch:32 Batch:400 Log-loss0.04448455199599266\n",
      "Epoch:32 Batch:480 Log-loss0.03675564005970955\n",
      "Epoch:32 Batch:560 Log-loss0.04667548090219498\n",
      "Epoch average log loss:0.03785014832392335\n",
      "In Epoch32, val_loss:0.039418808711276454, best_val_loss:0.03920176149522164, best_auc:0.9902525662097995\n",
      "Epoch:33 Batch:0 Log-loss0.03888056427240372\n",
      "Epoch:33 Batch:80 Log-loss0.038205403834581375\n",
      "Epoch:33 Batch:160 Log-loss0.025301359593868256\n",
      "Epoch:33 Batch:240 Log-loss0.034990403801202774\n",
      "Epoch:33 Batch:320 Log-loss0.03690105304121971\n",
      "Epoch:33 Batch:400 Log-loss0.05515192821621895\n",
      "Epoch:33 Batch:480 Log-loss0.044289667159318924\n",
      "Epoch:33 Batch:560 Log-loss0.05550548806786537\n",
      "Epoch average log loss:0.037994398648983664\n",
      "In Epoch33, val_loss:0.039331051505183436, best_val_loss:0.03920176149522164, best_auc:0.9902525662097995\n",
      "Epoch:34 Batch:0 Log-loss0.027740487828850746\n",
      "Epoch:34 Batch:80 Log-loss0.04051259532570839\n",
      "Epoch:34 Batch:160 Log-loss0.04576845094561577\n",
      "Epoch:34 Batch:240 Log-loss0.04444688558578491\n",
      "Epoch:34 Batch:320 Log-loss0.0318322628736496\n",
      "Epoch:34 Batch:400 Log-loss0.04274772107601166\n",
      "Epoch:34 Batch:480 Log-loss0.03552406653761864\n",
      "Epoch:34 Batch:560 Log-loss0.03820803388953209\n",
      "Epoch average log loss:0.038008018845825324\n",
      "In Epoch34, val_loss:0.03925109522287082, best_val_loss:0.03920176149522164, best_auc:0.9902525662097995\n",
      "Epoch:35 Batch:0 Log-loss0.04345189407467842\n",
      "Epoch:35 Batch:80 Log-loss0.020754603669047356\n",
      "Epoch:35 Batch:160 Log-loss0.04265333712100983\n",
      "Epoch:35 Batch:240 Log-loss0.040438782423734665\n",
      "Epoch:35 Batch:320 Log-loss0.043050479143857956\n",
      "Epoch:35 Batch:400 Log-loss0.03482632711529732\n",
      "Epoch:35 Batch:480 Log-loss0.02676207572221756\n",
      "Epoch:35 Batch:560 Log-loss0.04559783637523651\n",
      "Epoch average log loss:0.03783992655142876\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch35, val_loss:0.03909190979720897, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:36 Batch:0 Log-loss0.035077232867479324\n",
      "Epoch:36 Batch:80 Log-loss0.02506803534924984\n",
      "Epoch:36 Batch:160 Log-loss0.0270870178937912\n",
      "Epoch:36 Batch:240 Log-loss0.0497576929628849\n",
      "Epoch:36 Batch:320 Log-loss0.03824177011847496\n",
      "Epoch:36 Batch:400 Log-loss0.03795119374990463\n",
      "Epoch:36 Batch:480 Log-loss0.04652334377169609\n",
      "Epoch:36 Batch:560 Log-loss0.026407968252897263\n",
      "Epoch average log loss:0.03773237844330392\n",
      "In Epoch36, val_loss:0.039168759421618994, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:37 Batch:0 Log-loss0.01876979134976864\n",
      "Epoch:37 Batch:80 Log-loss0.024361489340662956\n",
      "Epoch:37 Batch:160 Log-loss0.03706727176904678\n",
      "Epoch:37 Batch:240 Log-loss0.04835648834705353\n",
      "Epoch:37 Batch:320 Log-loss0.042525727301836014\n",
      "Epoch:37 Batch:400 Log-loss0.048140641301870346\n",
      "Epoch:37 Batch:480 Log-loss0.04270373284816742\n",
      "Epoch:37 Batch:560 Log-loss0.03030714951455593\n",
      "Epoch average log loss:0.037723517441190776\n",
      "In Epoch37, val_loss:0.0393396560416445, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:38 Batch:0 Log-loss0.03528239578008652\n",
      "Epoch:38 Batch:80 Log-loss0.03209848329424858\n",
      "Epoch:38 Batch:160 Log-loss0.038390107452869415\n",
      "Epoch:38 Batch:240 Log-loss0.04145326465368271\n",
      "Epoch:38 Batch:320 Log-loss0.05522271618247032\n",
      "Epoch:38 Batch:400 Log-loss0.04718798026442528\n",
      "Epoch:38 Batch:480 Log-loss0.057902637869119644\n",
      "Epoch:38 Batch:560 Log-loss0.05026255175471306\n",
      "Epoch average log loss:0.037694422508190786\n",
      "In Epoch38, val_loss:0.03925050392581474, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:39 Batch:0 Log-loss0.043762460350990295\n",
      "Epoch:39 Batch:80 Log-loss0.030844056978821754\n",
      "Epoch:39 Batch:160 Log-loss0.03355883061885834\n",
      "Epoch:39 Batch:240 Log-loss0.03749329224228859\n",
      "Epoch:39 Batch:320 Log-loss0.02600746788084507\n",
      "Epoch:39 Batch:400 Log-loss0.042691703885793686\n",
      "Epoch:39 Batch:480 Log-loss0.037132419645786285\n",
      "Epoch:39 Batch:560 Log-loss0.035591572523117065\n",
      "Epoch average log loss:0.0375026458375422\n",
      "In Epoch39, val_loss:0.039168213975251207, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:40 Batch:0 Log-loss0.02694934792816639\n",
      "Epoch:40 Batch:80 Log-loss0.02550438977777958\n",
      "Epoch:40 Batch:160 Log-loss0.036239080131053925\n",
      "Epoch:40 Batch:240 Log-loss0.038483113050460815\n",
      "Epoch:40 Batch:320 Log-loss0.05131113529205322\n",
      "Epoch:40 Batch:400 Log-loss0.02480342425405979\n",
      "Epoch:40 Batch:480 Log-loss0.023087678477168083\n",
      "Epoch:40 Batch:560 Log-loss0.021478217095136642\n",
      "Epoch average log loss:0.037479433340818755\n",
      "In Epoch40, val_loss:0.0392650267094795, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:41 Batch:0 Log-loss0.04597810283303261\n",
      "Epoch:41 Batch:80 Log-loss0.033796634525060654\n",
      "Epoch:41 Batch:160 Log-loss0.051545608788728714\n",
      "Epoch:41 Batch:240 Log-loss0.03528379276394844\n",
      "Epoch:41 Batch:320 Log-loss0.044070035219192505\n",
      "Epoch:41 Batch:400 Log-loss0.03288054093718529\n",
      "Epoch:41 Batch:480 Log-loss0.023864643648266792\n",
      "Epoch:41 Batch:560 Log-loss0.03450201824307442\n",
      "Epoch average log loss:0.037548157599355494\n",
      "In Epoch41, val_loss:0.039145258077624444, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:42 Batch:0 Log-loss0.02483450062572956\n",
      "Epoch:42 Batch:80 Log-loss0.03436203673481941\n",
      "Epoch:42 Batch:160 Log-loss0.04074421152472496\n",
      "Epoch:42 Batch:240 Log-loss0.04961821436882019\n",
      "Epoch:42 Batch:320 Log-loss0.04241020977497101\n",
      "Epoch:42 Batch:400 Log-loss0.04657353088259697\n",
      "Epoch:42 Batch:480 Log-loss0.03884356841444969\n",
      "Epoch:42 Batch:560 Log-loss0.03797724097967148\n",
      "Epoch average log loss:0.037701487587764856\n",
      "In Epoch42, val_loss:0.03913294696711841, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:43 Batch:0 Log-loss0.025699861347675323\n",
      "Epoch:43 Batch:80 Log-loss0.02781265787780285\n",
      "Epoch:43 Batch:160 Log-loss0.05157412588596344\n",
      "Epoch:43 Batch:240 Log-loss0.04673925042152405\n",
      "Epoch:43 Batch:320 Log-loss0.054043982177972794\n",
      "Epoch:43 Batch:400 Log-loss0.034657347947359085\n",
      "Epoch:43 Batch:480 Log-loss0.02850390039384365\n",
      "Epoch:43 Batch:560 Log-loss0.03726945444941521\n",
      "Epoch average log loss:0.0375913208782939\n",
      "In Epoch43, val_loss:0.03913225612463948, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:44 Batch:0 Log-loss0.05122232064604759\n",
      "Epoch:44 Batch:80 Log-loss0.02127266116440296\n",
      "Epoch:44 Batch:160 Log-loss0.04940663278102875\n",
      "Epoch:44 Batch:240 Log-loss0.02804102748632431\n",
      "Epoch:44 Batch:320 Log-loss0.03837544471025467\n",
      "Epoch:44 Batch:400 Log-loss0.034195609390735626\n",
      "Epoch:44 Batch:480 Log-loss0.033045731484889984\n",
      "Epoch:44 Batch:560 Log-loss0.03216913342475891\n",
      "Epoch average log loss:0.03739785723321672\n",
      "In Epoch44, val_loss:0.03930070204928879, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:45 Batch:0 Log-loss0.033981893211603165\n",
      "Epoch:45 Batch:80 Log-loss0.03419753536581993\n",
      "Epoch:45 Batch:160 Log-loss0.04676952585577965\n",
      "Epoch:45 Batch:240 Log-loss0.03806556388735771\n",
      "Epoch:45 Batch:320 Log-loss0.023072516545653343\n",
      "Epoch:45 Batch:400 Log-loss0.042466118931770325\n",
      "Epoch:45 Batch:480 Log-loss0.03944319859147072\n",
      "Epoch:45 Batch:560 Log-loss0.029181336984038353\n",
      "Epoch average log loss:0.0375700542836317\n",
      "In Epoch45, val_loss:0.0393109995517262, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:46 Batch:0 Log-loss0.03838652744889259\n",
      "Epoch:46 Batch:80 Log-loss0.05162309110164642\n",
      "Epoch:46 Batch:160 Log-loss0.045851144939661026\n",
      "Epoch:46 Batch:240 Log-loss0.0369761697947979\n",
      "Epoch:46 Batch:320 Log-loss0.03209755942225456\n",
      "Epoch:46 Batch:400 Log-loss0.040081147104501724\n",
      "Epoch:46 Batch:480 Log-loss0.044804614037275314\n",
      "Epoch:46 Batch:560 Log-loss0.03136324882507324\n",
      "Epoch average log loss:0.037592621706426146\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In Epoch46, val_loss:0.03925254646954041, best_val_loss:0.03909190979720897, best_auc:0.9903393737793028\n",
      "Epoch:47 Batch:0 Log-loss0.042367175221443176\n",
      "Epoch:47 Batch:80 Log-loss0.026822848245501518\n",
      "Epoch:47 Batch:160 Log-loss0.048959504812955856\n",
      "Epoch:47 Batch:240 Log-loss0.04415258765220642\n",
      "Epoch:47 Batch:320 Log-loss0.03561440482735634\n",
      "Epoch:47 Batch:400 Log-loss0.028074493631720543\n",
      "Epoch:47 Batch:480 Log-loss0.04009208455681801\n",
      "Epoch:47 Batch:560 Log-loss0.031041815876960754\n",
      "Epoch average log loss:0.0375291524820828\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_641.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 2\n",
      "Epoch:1 Batch:0 Log-loss0.6674628853797913\n",
      "Epoch:1 Batch:80 Log-loss0.13349492847919464\n",
      "Epoch:1 Batch:160 Log-loss0.085947185754776\n",
      "Epoch:1 Batch:240 Log-loss0.0798916444182396\n",
      "Epoch:1 Batch:320 Log-loss0.0926297977566719\n",
      "Epoch:1 Batch:400 Log-loss0.06787348538637161\n",
      "Epoch:1 Batch:480 Log-loss0.09275113791227341\n",
      "Epoch:1 Batch:560 Log-loss0.058310747146606445\n",
      "Epoch average log loss:0.10413421556087477\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.05714214608237786, best_val_loss:0.05714214608237786, best_auc:0.968008672445535\n",
      "Epoch:2 Batch:0 Log-loss0.06191709265112877\n",
      "Epoch:2 Batch:80 Log-loss0.06678467243909836\n",
      "Epoch:2 Batch:160 Log-loss0.07715121656656265\n",
      "Epoch:2 Batch:240 Log-loss0.06504728645086288\n",
      "Epoch:2 Batch:320 Log-loss0.053942497819662094\n",
      "Epoch:2 Batch:400 Log-loss0.05083617568016052\n",
      "Epoch:2 Batch:480 Log-loss0.06371616572141647\n",
      "Epoch:2 Batch:560 Log-loss0.06698186695575714\n",
      "Epoch average log loss:0.05753376933945609\n",
      "In Epoch2, val_loss:0.058826764220781674, best_val_loss:0.05714214608237786, best_auc:0.968008672445535\n",
      "Epoch:3 Batch:0 Log-loss0.05738280713558197\n",
      "Epoch:3 Batch:80 Log-loss0.03707490488886833\n",
      "Epoch:3 Batch:160 Log-loss0.04746672138571739\n",
      "Epoch:3 Batch:240 Log-loss0.057601362466812134\n",
      "Epoch:3 Batch:320 Log-loss0.0408872552216053\n",
      "Epoch:3 Batch:400 Log-loss0.049332860857248306\n",
      "Epoch:3 Batch:480 Log-loss0.06163378432393074\n",
      "Epoch:3 Batch:560 Log-loss0.05091668292880058\n",
      "Epoch average log loss:0.054834373062476516\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.05152615863333205, best_val_loss:0.05152615863333205, best_auc:0.9772164024911859\n",
      "Epoch:4 Batch:0 Log-loss0.05114249885082245\n",
      "Epoch:4 Batch:80 Log-loss0.05420183762907982\n",
      "Epoch:4 Batch:160 Log-loss0.05542115494608879\n",
      "Epoch:4 Batch:240 Log-loss0.06378933042287827\n",
      "Epoch:4 Batch:320 Log-loss0.06656341999769211\n",
      "Epoch:4 Batch:400 Log-loss0.05495089665055275\n",
      "Epoch:4 Batch:480 Log-loss0.05155276879668236\n",
      "Epoch:4 Batch:560 Log-loss0.06327976286411285\n",
      "Epoch average log loss:0.052878146039854204\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.047659953303373415, best_val_loss:0.047659953303373415, best_auc:0.9788863832872855\n",
      "Epoch:5 Batch:0 Log-loss0.06020604446530342\n",
      "Epoch:5 Batch:80 Log-loss0.04994918033480644\n",
      "Epoch:5 Batch:160 Log-loss0.04425058141350746\n",
      "Epoch:5 Batch:240 Log-loss0.049613598734140396\n",
      "Epoch:5 Batch:320 Log-loss0.05514732003211975\n",
      "Epoch:5 Batch:400 Log-loss0.04260633513331413\n",
      "Epoch:5 Batch:480 Log-loss0.05094057321548462\n",
      "Epoch:5 Batch:560 Log-loss0.08339058607816696\n",
      "Epoch average log loss:0.05053080711846373\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.04687948904047245, best_val_loss:0.04687948904047245, best_auc:0.9794893201796199\n",
      "Epoch:6 Batch:0 Log-loss0.03783368691802025\n",
      "Epoch:6 Batch:80 Log-loss0.048043135553598404\n",
      "Epoch:6 Batch:160 Log-loss0.029887497425079346\n",
      "Epoch:6 Batch:240 Log-loss0.058682817965745926\n",
      "Epoch:6 Batch:320 Log-loss0.053233515471220016\n",
      "Epoch:6 Batch:400 Log-loss0.059521641582250595\n",
      "Epoch:6 Batch:480 Log-loss0.043279264122247696\n",
      "Epoch:6 Batch:560 Log-loss0.07432930916547775\n",
      "Epoch average log loss:0.04970811972660678\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.04622632325249165, best_val_loss:0.04622632325249165, best_auc:0.9806905817899656\n",
      "Epoch:7 Batch:0 Log-loss0.03589232638478279\n",
      "Epoch:7 Batch:80 Log-loss0.05499287322163582\n",
      "Epoch:7 Batch:160 Log-loss0.05598917976021767\n",
      "Epoch:7 Batch:240 Log-loss0.04734833538532257\n",
      "Epoch:7 Batch:320 Log-loss0.05610433220863342\n",
      "Epoch:7 Batch:400 Log-loss0.04382457956671715\n",
      "Epoch:7 Batch:480 Log-loss0.03935958445072174\n",
      "Epoch:7 Batch:560 Log-loss0.04865361005067825\n",
      "Epoch average log loss:0.04927943533486021\n",
      "In Epoch7, val_loss:0.04644194517680503, best_val_loss:0.04622632325249165, best_auc:0.9806905817899656\n",
      "Epoch:8 Batch:0 Log-loss0.04295099154114723\n",
      "Epoch:8 Batch:80 Log-loss0.0360599122941494\n",
      "Epoch:8 Batch:160 Log-loss0.048834096640348434\n",
      "Epoch:8 Batch:240 Log-loss0.04988127574324608\n",
      "Epoch:8 Batch:320 Log-loss0.04501594975590706\n",
      "Epoch:8 Batch:400 Log-loss0.046731024980545044\n",
      "Epoch:8 Batch:480 Log-loss0.04561471566557884\n",
      "Epoch:8 Batch:560 Log-loss0.054260607808828354\n",
      "Epoch average log loss:0.049242081569640765\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.04538907128782401, best_val_loss:0.04538907128782401, best_auc:0.9819503221157239\n",
      "Epoch:9 Batch:0 Log-loss0.041922688484191895\n",
      "Epoch:9 Batch:80 Log-loss0.05834601819515228\n",
      "Epoch:9 Batch:160 Log-loss0.06209326907992363\n",
      "Epoch:9 Batch:240 Log-loss0.04808398708701134\n",
      "Epoch:9 Batch:320 Log-loss0.07001245766878128\n",
      "Epoch:9 Batch:400 Log-loss0.056761424988508224\n",
      "Epoch:9 Batch:480 Log-loss0.03379306569695473\n",
      "Epoch:9 Batch:560 Log-loss0.046617548912763596\n",
      "Epoch average log loss:0.04853013912548444\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.044769498368474846, best_val_loss:0.044769498368474846, best_auc:0.9822709116269491\n",
      "Epoch:10 Batch:0 Log-loss0.0733063817024231\n",
      "Epoch:10 Batch:80 Log-loss0.05257158353924751\n",
      "Epoch:10 Batch:160 Log-loss0.04369648918509483\n",
      "Epoch:10 Batch:240 Log-loss0.04521780088543892\n",
      "Epoch:10 Batch:320 Log-loss0.045379798859357834\n",
      "Epoch:10 Batch:400 Log-loss0.058526914566755295\n",
      "Epoch:10 Batch:480 Log-loss0.05026014521718025\n",
      "Epoch:10 Batch:560 Log-loss0.045616887509822845\n",
      "Epoch average log loss:0.047659950676773276\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.04411275173800397, best_val_loss:0.04411275173800397, best_auc:0.9838429390806631\n",
      "Epoch:11 Batch:0 Log-loss0.0586872361600399\n",
      "Epoch:11 Batch:80 Log-loss0.040305573493242264\n",
      "Epoch:11 Batch:160 Log-loss0.04314669594168663\n",
      "Epoch:11 Batch:240 Log-loss0.044235724955797195\n",
      "Epoch:11 Batch:320 Log-loss0.06295311450958252\n",
      "Epoch:11 Batch:400 Log-loss0.05247485265135765\n",
      "Epoch:11 Batch:480 Log-loss0.04986407235264778\n",
      "Epoch:11 Batch:560 Log-loss0.04683021083474159\n",
      "Epoch average log loss:0.046905684527674955\n",
      "In Epoch11, val_loss:0.0443130362903727, best_val_loss:0.04411275173800397, best_auc:0.9838429390806631\n",
      "Epoch:12 Batch:0 Log-loss0.03272705897688866\n",
      "Epoch:12 Batch:80 Log-loss0.048168864101171494\n",
      "Epoch:12 Batch:160 Log-loss0.027172937989234924\n",
      "Epoch:12 Batch:240 Log-loss0.05300930514931679\n",
      "Epoch:12 Batch:320 Log-loss0.0446135550737381\n",
      "Epoch:12 Batch:400 Log-loss0.05089464411139488\n",
      "Epoch:12 Batch:480 Log-loss0.039628177881240845\n",
      "Epoch:12 Batch:560 Log-loss0.04014073684811592\n",
      "Epoch average log loss:0.046184854452232166\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.04302802243240981, best_val_loss:0.04302802243240981, best_auc:0.9862666836362987\n",
      "Epoch:13 Batch:0 Log-loss0.06262096017599106\n",
      "Epoch:13 Batch:80 Log-loss0.04533347859978676\n",
      "Epoch:13 Batch:160 Log-loss0.035048868507146835\n",
      "Epoch:13 Batch:240 Log-loss0.0386260561645031\n",
      "Epoch:13 Batch:320 Log-loss0.03772922232747078\n",
      "Epoch:13 Batch:400 Log-loss0.06391829997301102\n",
      "Epoch:13 Batch:480 Log-loss0.056756842881441116\n",
      "Epoch:13 Batch:560 Log-loss0.03147265315055847\n",
      "Epoch average log loss:0.04576340030679213\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch13, val_loss:0.0423679772131392, best_val_loss:0.0423679772131392, best_auc:0.9867719148782609\n",
      "Epoch:14 Batch:0 Log-loss0.044908564537763596\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:14 Batch:80 Log-loss0.04302411153912544\n",
      "Epoch:14 Batch:160 Log-loss0.0454287976026535\n",
      "Epoch:14 Batch:240 Log-loss0.036822326481342316\n",
      "Epoch:14 Batch:320 Log-loss0.04764799773693085\n",
      "Epoch:14 Batch:400 Log-loss0.041338857263326645\n",
      "Epoch:14 Batch:480 Log-loss0.0558943934738636\n",
      "Epoch:14 Batch:560 Log-loss0.04525012522935867\n",
      "Epoch average log loss:0.04535711080700691\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch14, val_loss:0.04180933231121664, best_val_loss:0.04180933231121664, best_auc:0.9873879664320498\n",
      "Epoch:15 Batch:0 Log-loss0.0514286570250988\n",
      "Epoch:15 Batch:80 Log-loss0.05611097440123558\n",
      "Epoch:15 Batch:160 Log-loss0.03017943538725376\n",
      "Epoch:15 Batch:240 Log-loss0.05530748888850212\n",
      "Epoch:15 Batch:320 Log-loss0.03709632530808449\n",
      "Epoch:15 Batch:400 Log-loss0.03576720133423805\n",
      "Epoch:15 Batch:480 Log-loss0.036006804555654526\n",
      "Epoch:15 Batch:560 Log-loss0.04189033806324005\n",
      "Epoch average log loss:0.04461571832653135\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch15, val_loss:0.04137818074882676, best_val_loss:0.04137818074882676, best_auc:0.9879994369336274\n",
      "Epoch:16 Batch:0 Log-loss0.04167916253209114\n",
      "Epoch:16 Batch:80 Log-loss0.04791467264294624\n",
      "Epoch:16 Batch:160 Log-loss0.03264525160193443\n",
      "Epoch:16 Batch:240 Log-loss0.03867903724312782\n",
      "Epoch:16 Batch:320 Log-loss0.04976176843047142\n",
      "Epoch:16 Batch:400 Log-loss0.057544928044080734\n",
      "Epoch:16 Batch:480 Log-loss0.033082980662584305\n",
      "Epoch:16 Batch:560 Log-loss0.03045179508626461\n",
      "Epoch average log loss:0.044554705156146415\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch16, val_loss:0.041074401149057575, best_val_loss:0.041074401149057575, best_auc:0.9882801413917218\n",
      "Epoch:17 Batch:0 Log-loss0.03167031705379486\n",
      "Epoch:17 Batch:80 Log-loss0.04771893098950386\n",
      "Epoch:17 Batch:160 Log-loss0.039355818182229996\n",
      "Epoch:17 Batch:240 Log-loss0.05368274450302124\n",
      "Epoch:17 Batch:320 Log-loss0.02868175320327282\n",
      "Epoch:17 Batch:400 Log-loss0.04254194721579552\n",
      "Epoch:17 Batch:480 Log-loss0.039243850857019424\n",
      "Epoch:17 Batch:560 Log-loss0.041116829961538315\n",
      "Epoch average log loss:0.04407350214397801\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch17, val_loss:0.04054841111012516, best_val_loss:0.04054841111012516, best_auc:0.9888099030741784\n",
      "Epoch:18 Batch:0 Log-loss0.05640360713005066\n",
      "Epoch:18 Batch:80 Log-loss0.056571681052446365\n",
      "Epoch:18 Batch:160 Log-loss0.04470519721508026\n",
      "Epoch:18 Batch:240 Log-loss0.05637267604470253\n",
      "Epoch:18 Batch:320 Log-loss0.044797588139772415\n",
      "Epoch:18 Batch:400 Log-loss0.04113227501511574\n",
      "Epoch:18 Batch:480 Log-loss0.06429500877857208\n",
      "Epoch:18 Batch:560 Log-loss0.038043588399887085\n",
      "Epoch average log loss:0.04352265161329082\n",
      "In Epoch18, val_loss:0.04087850384123989, best_val_loss:0.04054841111012516, best_auc:0.9888099030741784\n",
      "Epoch:19 Batch:0 Log-loss0.035400811582803726\n",
      "Epoch:19 Batch:80 Log-loss0.0388978086411953\n",
      "Epoch:19 Batch:160 Log-loss0.02211771346628666\n",
      "Epoch:19 Batch:240 Log-loss0.042946651577949524\n",
      "Epoch:19 Batch:320 Log-loss0.040625330060720444\n",
      "Epoch:19 Batch:400 Log-loss0.04464815557003021\n",
      "Epoch:19 Batch:480 Log-loss0.031777456402778625\n",
      "Epoch:19 Batch:560 Log-loss0.04439917206764221\n",
      "Epoch average log loss:0.04338696941932929\n",
      "In Epoch19, val_loss:0.04077899570199439, best_val_loss:0.04054841111012516, best_auc:0.9888099030741784\n",
      "Epoch:20 Batch:0 Log-loss0.045567020773887634\n",
      "Epoch:20 Batch:80 Log-loss0.04689214751124382\n",
      "Epoch:20 Batch:160 Log-loss0.07199416309595108\n",
      "Epoch:20 Batch:240 Log-loss0.03865961357951164\n",
      "Epoch:20 Batch:320 Log-loss0.06511274725198746\n",
      "Epoch:20 Batch:400 Log-loss0.050999581813812256\n",
      "Epoch:20 Batch:480 Log-loss0.035409118980169296\n",
      "Epoch:20 Batch:560 Log-loss0.047268521040678024\n",
      "Epoch average log loss:0.04289483199827373\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch20, val_loss:0.04027355117907865, best_val_loss:0.04027355117907865, best_auc:0.9894316720587796\n",
      "Epoch:21 Batch:0 Log-loss0.03872326388955116\n",
      "Epoch:21 Batch:80 Log-loss0.06331247836351395\n",
      "Epoch:21 Batch:160 Log-loss0.040890567004680634\n",
      "Epoch:21 Batch:240 Log-loss0.04232025146484375\n",
      "Epoch:21 Batch:320 Log-loss0.045347776263952255\n",
      "Epoch:21 Batch:400 Log-loss0.030413374304771423\n",
      "Epoch:21 Batch:480 Log-loss0.04128233715891838\n",
      "Epoch:21 Batch:560 Log-loss0.03640811890363693\n",
      "Epoch average log loss:0.04265589511986556\n",
      "In Epoch21, val_loss:0.04029953478647738, best_val_loss:0.04027355117907865, best_auc:0.9894316720587796\n",
      "Epoch:22 Batch:0 Log-loss0.023715049028396606\n",
      "Epoch:22 Batch:80 Log-loss0.05399076268076897\n",
      "Epoch:22 Batch:160 Log-loss0.052513640373945236\n",
      "Epoch:22 Batch:240 Log-loss0.04052603244781494\n",
      "Epoch:22 Batch:320 Log-loss0.03449692949652672\n",
      "Epoch:22 Batch:400 Log-loss0.040388721972703934\n",
      "Epoch:22 Batch:480 Log-loss0.02723337523639202\n",
      "Epoch:22 Batch:560 Log-loss0.048246704041957855\n",
      "Epoch average log loss:0.04233785351139626\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch22, val_loss:0.040062000787809214, best_val_loss:0.040062000787809214, best_auc:0.9897496778834959\n",
      "Epoch:23 Batch:0 Log-loss0.04646274819970131\n",
      "Epoch:23 Batch:80 Log-loss0.048614028841257095\n",
      "Epoch:23 Batch:160 Log-loss0.04028327763080597\n",
      "Epoch:23 Batch:240 Log-loss0.033883579075336456\n",
      "Epoch:23 Batch:320 Log-loss0.03886597231030464\n",
      "Epoch:23 Batch:400 Log-loss0.03116636537015438\n",
      "Epoch:23 Batch:480 Log-loss0.04175201430916786\n",
      "Epoch:23 Batch:560 Log-loss0.036291494965553284\n",
      "Epoch average log loss:0.042293565959802694\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch23, val_loss:0.03953386500671157, best_val_loss:0.03953386500671157, best_auc:0.9899186666342269\n",
      "Epoch:24 Batch:0 Log-loss0.038239724934101105\n",
      "Epoch:24 Batch:80 Log-loss0.04053857550024986\n",
      "Epoch:24 Batch:160 Log-loss0.028428522869944572\n",
      "Epoch:24 Batch:240 Log-loss0.05031374469399452\n",
      "Epoch:24 Batch:320 Log-loss0.0506085567176342\n",
      "Epoch:24 Batch:400 Log-loss0.05824275314807892\n",
      "Epoch:24 Batch:480 Log-loss0.03721153363585472\n",
      "Epoch:24 Batch:560 Log-loss0.036063771694898605\n",
      "Epoch average log loss:0.042184537785526895\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch24, val_loss:0.039513653055685376, best_val_loss:0.039513653055685376, best_auc:0.9900493025313924\n",
      "Epoch:25 Batch:0 Log-loss0.036967068910598755\n",
      "Epoch:25 Batch:80 Log-loss0.05432334914803505\n",
      "Epoch:25 Batch:160 Log-loss0.03843158856034279\n",
      "Epoch:25 Batch:240 Log-loss0.040818747133016586\n",
      "Epoch:25 Batch:320 Log-loss0.044014643877744675\n",
      "Epoch:25 Batch:400 Log-loss0.04916827008128166\n",
      "Epoch:25 Batch:480 Log-loss0.051160648465156555\n",
      "Epoch:25 Batch:560 Log-loss0.032120488584041595\n",
      "Epoch average log loss:0.042099531127938204\n",
      "In Epoch25, val_loss:0.039611571590377155, best_val_loss:0.039513653055685376, best_auc:0.9900493025313924\n",
      "Epoch:26 Batch:0 Log-loss0.04176619276404381\n",
      "Epoch:26 Batch:80 Log-loss0.048694029450416565\n",
      "Epoch:26 Batch:160 Log-loss0.056253742426633835\n",
      "Epoch:26 Batch:240 Log-loss0.04842514172196388\n",
      "Epoch:26 Batch:320 Log-loss0.05248120054602623\n",
      "Epoch:26 Batch:400 Log-loss0.02473163791000843\n",
      "Epoch:26 Batch:480 Log-loss0.03605896607041359\n",
      "Epoch:26 Batch:560 Log-loss0.03814508020877838\n",
      "Epoch average log loss:0.041632587968238766\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch26, val_loss:0.03940670926727209, best_val_loss:0.03940670926727209, best_auc:0.9901891199196409\n",
      "Epoch:27 Batch:0 Log-loss0.040793441236019135\n",
      "Epoch:27 Batch:80 Log-loss0.044488370418548584\n",
      "Epoch:27 Batch:160 Log-loss0.041373517364263535\n",
      "Epoch:27 Batch:240 Log-loss0.039940450340509415\n",
      "Epoch:27 Batch:320 Log-loss0.04416869953274727\n",
      "Epoch:27 Batch:400 Log-loss0.031095480546355247\n",
      "Epoch:27 Batch:480 Log-loss0.05342353507876396\n",
      "Epoch:27 Batch:560 Log-loss0.03623441979289055\n",
      "Epoch average log loss:0.04154940055284117\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch27, val_loss:0.03937635083854776, best_val_loss:0.03937635083854776, best_auc:0.990283994957364\n",
      "Epoch:28 Batch:0 Log-loss0.04495592787861824\n",
      "Epoch:28 Batch:80 Log-loss0.05469140037894249\n",
      "Epoch:28 Batch:160 Log-loss0.04335939884185791\n",
      "Epoch:28 Batch:240 Log-loss0.04574039950966835\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:28 Batch:320 Log-loss0.03656997159123421\n",
      "Epoch:28 Batch:400 Log-loss0.04381921514868736\n",
      "Epoch:28 Batch:480 Log-loss0.03894702345132828\n",
      "Epoch:28 Batch:560 Log-loss0.04251810535788536\n",
      "Epoch average log loss:0.0414368053366031\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch28, val_loss:0.03920685232542534, best_val_loss:0.03920685232542534, best_auc:0.9903751770621172\n",
      "Epoch:29 Batch:0 Log-loss0.04486597701907158\n",
      "Epoch:29 Batch:80 Log-loss0.04063516482710838\n",
      "Epoch:29 Batch:160 Log-loss0.05469881370663643\n",
      "Epoch:29 Batch:240 Log-loss0.041476618498563766\n",
      "Epoch:29 Batch:320 Log-loss0.044276732951402664\n",
      "Epoch:29 Batch:400 Log-loss0.03186545521020889\n",
      "Epoch:29 Batch:480 Log-loss0.042694564908742905\n",
      "Epoch:29 Batch:560 Log-loss0.035684920847415924\n",
      "Epoch average log loss:0.04128334309040968\n",
      "In Epoch29, val_loss:0.039228665459441744, best_val_loss:0.03920685232542534, best_auc:0.9903751770621172\n",
      "Epoch:30 Batch:0 Log-loss0.055004749447107315\n",
      "Epoch:30 Batch:80 Log-loss0.057288169860839844\n",
      "Epoch:30 Batch:160 Log-loss0.03822014480829239\n",
      "Epoch:30 Batch:240 Log-loss0.027558529749512672\n",
      "Epoch:30 Batch:320 Log-loss0.05554458498954773\n",
      "Epoch:30 Batch:400 Log-loss0.04773963987827301\n",
      "Epoch:30 Batch:480 Log-loss0.034027665853500366\n",
      "Epoch:30 Batch:560 Log-loss0.03167283907532692\n",
      "Epoch average log loss:0.04107347691897303\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch30, val_loss:0.03906538860353163, best_val_loss:0.03906538860353163, best_auc:0.990462085211765\n",
      "Epoch:31 Batch:0 Log-loss0.056337397545576096\n",
      "Epoch:31 Batch:80 Log-loss0.04735056683421135\n",
      "Epoch:31 Batch:160 Log-loss0.04950053617358208\n",
      "Epoch:31 Batch:240 Log-loss0.037847451865673065\n",
      "Epoch:31 Batch:320 Log-loss0.050393879413604736\n",
      "Epoch:31 Batch:400 Log-loss0.03490881621837616\n",
      "Epoch:31 Batch:480 Log-loss0.03614421188831329\n",
      "Epoch:31 Batch:560 Log-loss0.040171291679143906\n",
      "Epoch average log loss:0.04116123653283077\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch31, val_loss:0.038884766541579, best_val_loss:0.038884766541579, best_auc:0.9905569087913092\n",
      "Epoch:32 Batch:0 Log-loss0.030599072575569153\n",
      "Epoch:32 Batch:80 Log-loss0.05048452317714691\n",
      "Epoch:32 Batch:160 Log-loss0.04781155660748482\n",
      "Epoch:32 Batch:240 Log-loss0.043922338634729385\n",
      "Epoch:32 Batch:320 Log-loss0.038807883858680725\n",
      "Epoch:32 Batch:400 Log-loss0.03694789111614227\n",
      "Epoch:32 Batch:480 Log-loss0.03054519183933735\n",
      "Epoch:32 Batch:560 Log-loss0.025749845430254936\n",
      "Epoch average log loss:0.0410825494843136\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch32, val_loss:0.03884779001351305, best_val_loss:0.03884779001351305, best_auc:0.9905498306223238\n",
      "Epoch:33 Batch:0 Log-loss0.03312059864401817\n",
      "Epoch:33 Batch:80 Log-loss0.05830293521285057\n",
      "Epoch:33 Batch:160 Log-loss0.0380365289747715\n",
      "Epoch:33 Batch:240 Log-loss0.029758505523204803\n",
      "Epoch:33 Batch:320 Log-loss0.02749660424888134\n",
      "Epoch:33 Batch:400 Log-loss0.03351760283112526\n",
      "Epoch:33 Batch:480 Log-loss0.030529310926795006\n",
      "Epoch:33 Batch:560 Log-loss0.02973444201052189\n",
      "Epoch average log loss:0.040846776656274286\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch33, val_loss:0.038763977045555426, best_val_loss:0.038763977045555426, best_auc:0.9906085181118982\n",
      "Epoch:34 Batch:0 Log-loss0.054058898240327835\n",
      "Epoch:34 Batch:80 Log-loss0.05162443593144417\n",
      "Epoch:34 Batch:160 Log-loss0.036636579781770706\n",
      "Epoch:34 Batch:240 Log-loss0.029062068089842796\n",
      "Epoch:34 Batch:320 Log-loss0.04750663414597511\n",
      "Epoch:34 Batch:400 Log-loss0.03836996853351593\n",
      "Epoch:34 Batch:480 Log-loss0.04690835252404213\n",
      "Epoch:34 Batch:560 Log-loss0.03746769204735756\n",
      "Epoch average log loss:0.04073582081057663\n",
      "In Epoch34, val_loss:0.03889590803129557, best_val_loss:0.038763977045555426, best_auc:0.9906085181118982\n",
      "Epoch:35 Batch:0 Log-loss0.04743371903896332\n",
      "Epoch:35 Batch:80 Log-loss0.04564325138926506\n",
      "Epoch:35 Batch:160 Log-loss0.03525165095925331\n",
      "Epoch:35 Batch:240 Log-loss0.04595295712351799\n",
      "Epoch:35 Batch:320 Log-loss0.04196086898446083\n",
      "Epoch:35 Batch:400 Log-loss0.02781822718679905\n",
      "Epoch:35 Batch:480 Log-loss0.04434442147612572\n",
      "Epoch:35 Batch:560 Log-loss0.03025965578854084\n",
      "Epoch average log loss:0.04072943697018283\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch35, val_loss:0.038753219311753274, best_val_loss:0.038753219311753274, best_auc:0.9906388598709773\n",
      "Epoch:36 Batch:0 Log-loss0.03853854909539223\n",
      "Epoch:36 Batch:80 Log-loss0.04198562726378441\n",
      "Epoch:36 Batch:160 Log-loss0.04405214264988899\n",
      "Epoch:36 Batch:240 Log-loss0.041654836386442184\n",
      "Epoch:36 Batch:320 Log-loss0.035950109362602234\n",
      "Epoch:36 Batch:400 Log-loss0.041746508330106735\n",
      "Epoch:36 Batch:480 Log-loss0.0359845869243145\n",
      "Epoch:36 Batch:560 Log-loss0.0431765541434288\n",
      "Epoch average log loss:0.04063151445505875\n",
      "In Epoch36, val_loss:0.038756164676235556, best_val_loss:0.038753219311753274, best_auc:0.9906388598709773\n",
      "Epoch:37 Batch:0 Log-loss0.025409957394003868\n",
      "Epoch:37 Batch:80 Log-loss0.04918801411986351\n",
      "Epoch:37 Batch:160 Log-loss0.03163429722189903\n",
      "Epoch:37 Batch:240 Log-loss0.04075906053185463\n",
      "Epoch:37 Batch:320 Log-loss0.04798944294452667\n",
      "Epoch:37 Batch:400 Log-loss0.046976108103990555\n",
      "Epoch:37 Batch:480 Log-loss0.04317820444703102\n",
      "Epoch:37 Batch:560 Log-loss0.04045826941728592\n",
      "Epoch average log loss:0.04055188094664897\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch37, val_loss:0.03866009171772642, best_val_loss:0.03866009171772642, best_auc:0.9907772518370553\n",
      "Epoch:38 Batch:0 Log-loss0.04248513653874397\n",
      "Epoch:38 Batch:80 Log-loss0.05857374891638756\n",
      "Epoch:38 Batch:160 Log-loss0.037353258579969406\n",
      "Epoch:38 Batch:240 Log-loss0.054910991340875626\n",
      "Epoch:38 Batch:320 Log-loss0.05056973174214363\n",
      "Epoch:38 Batch:400 Log-loss0.03409624099731445\n",
      "Epoch:38 Batch:480 Log-loss0.034341104328632355\n",
      "Epoch:38 Batch:560 Log-loss0.03320680186152458\n",
      "Epoch average log loss:0.04042274724093399\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch38, val_loss:0.03860460507981058, best_val_loss:0.03860460507981058, best_auc:0.9907490889936247\n",
      "Epoch:39 Batch:0 Log-loss0.03975170478224754\n",
      "Epoch:39 Batch:80 Log-loss0.037248414009809494\n",
      "Epoch:39 Batch:160 Log-loss0.040130551904439926\n",
      "Epoch:39 Batch:240 Log-loss0.05328240990638733\n",
      "Epoch:39 Batch:320 Log-loss0.03460898622870445\n",
      "Epoch:39 Batch:400 Log-loss0.046523500233888626\n",
      "Epoch:39 Batch:480 Log-loss0.03671562671661377\n",
      "Epoch:39 Batch:560 Log-loss0.03267165645956993\n",
      "Epoch average log loss:0.04044204341646816\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch39, val_loss:0.038502631441594805, best_val_loss:0.038502631441594805, best_auc:0.9907638151250046\n",
      "Epoch:40 Batch:0 Log-loss0.04062015935778618\n",
      "Epoch:40 Batch:80 Log-loss0.04466071352362633\n",
      "Epoch:40 Batch:160 Log-loss0.046893104910850525\n",
      "Epoch:40 Batch:240 Log-loss0.04418782889842987\n",
      "Epoch:40 Batch:320 Log-loss0.04591519758105278\n",
      "Epoch:40 Batch:400 Log-loss0.04929463937878609\n",
      "Epoch:40 Batch:480 Log-loss0.03008786402642727\n",
      "Epoch:40 Batch:560 Log-loss0.04177003726363182\n",
      "Epoch average log loss:0.040506559081508645\n",
      "In Epoch40, val_loss:0.03873462386903688, best_val_loss:0.038502631441594805, best_auc:0.9907638151250046\n",
      "Epoch:41 Batch:0 Log-loss0.03244095668196678\n",
      "Epoch:41 Batch:80 Log-loss0.047082263976335526\n",
      "Epoch:41 Batch:160 Log-loss0.044686246663331985\n",
      "Epoch:41 Batch:240 Log-loss0.046595800668001175\n",
      "Epoch:41 Batch:320 Log-loss0.03752986714243889\n",
      "Epoch:41 Batch:400 Log-loss0.05353499948978424\n",
      "Epoch:41 Batch:480 Log-loss0.019450215622782707\n",
      "Epoch:41 Batch:560 Log-loss0.05756001174449921\n",
      "Epoch average log loss:0.04031155111733824\n",
      "In Epoch41, val_loss:0.0387826502656495, best_val_loss:0.038502631441594805, best_auc:0.9907638151250046\n",
      "Epoch:42 Batch:0 Log-loss0.0352669283747673\n",
      "Epoch:42 Batch:80 Log-loss0.04484259709715843\n",
      "Epoch:42 Batch:160 Log-loss0.04165510833263397\n",
      "Epoch:42 Batch:240 Log-loss0.06798119843006134\n",
      "Epoch:42 Batch:320 Log-loss0.046287234872579575\n",
      "Epoch:42 Batch:400 Log-loss0.03161726891994476\n",
      "Epoch:42 Batch:480 Log-loss0.04552239552140236\n",
      "Epoch:42 Batch:560 Log-loss0.03465527668595314\n",
      "Epoch average log loss:0.04019892348109611\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In Epoch42, val_loss:0.03858395984099385, best_val_loss:0.038502631441594805, best_auc:0.9907638151250046\n",
      "Epoch:43 Batch:0 Log-loss0.04289352893829346\n",
      "Epoch:43 Batch:80 Log-loss0.03482220694422722\n",
      "Epoch:43 Batch:160 Log-loss0.028714537620544434\n",
      "Epoch:43 Batch:240 Log-loss0.0372578464448452\n",
      "Epoch:43 Batch:320 Log-loss0.042857956141233444\n",
      "Epoch:43 Batch:400 Log-loss0.03794160485267639\n",
      "Epoch:43 Batch:480 Log-loss0.037178099155426025\n",
      "Epoch:43 Batch:560 Log-loss0.02656126394867897\n",
      "Epoch average log loss:0.04009668595224087\n",
      "In Epoch43, val_loss:0.038540649454517684, best_val_loss:0.038502631441594805, best_auc:0.9907638151250046\n",
      "Epoch:44 Batch:0 Log-loss0.04633226990699768\n",
      "Epoch:44 Batch:80 Log-loss0.061987098306417465\n",
      "Epoch:44 Batch:160 Log-loss0.04715224727988243\n",
      "Epoch:44 Batch:240 Log-loss0.05227178335189819\n",
      "Epoch:44 Batch:320 Log-loss0.027004098519682884\n",
      "Epoch:44 Batch:400 Log-loss0.029736021533608437\n",
      "Epoch:44 Batch:480 Log-loss0.04029693827033043\n",
      "Epoch:44 Batch:560 Log-loss0.04200386628508568\n",
      "Epoch average log loss:0.03994653679962669\n",
      "In Epoch44, val_loss:0.0385738719304494, best_val_loss:0.038502631441594805, best_auc:0.9907638151250046\n",
      "Epoch:45 Batch:0 Log-loss0.043422043323516846\n",
      "Epoch:45 Batch:80 Log-loss0.041476137936115265\n",
      "Epoch:45 Batch:160 Log-loss0.04168209806084633\n",
      "Epoch:45 Batch:240 Log-loss0.040825773030519485\n",
      "Epoch:45 Batch:320 Log-loss0.029077036306262016\n",
      "Epoch:45 Batch:400 Log-loss0.05413671210408211\n",
      "Epoch:45 Batch:480 Log-loss0.02737526036798954\n",
      "Epoch:45 Batch:560 Log-loss0.03701971843838692\n",
      "Epoch average log loss:0.040135216802757764\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch45, val_loss:0.038408574866939736, best_val_loss:0.038408574866939736, best_auc:0.9908254612334152\n",
      "Epoch:46 Batch:0 Log-loss0.03234419971704483\n",
      "Epoch:46 Batch:80 Log-loss0.03837668523192406\n",
      "Epoch:46 Batch:160 Log-loss0.03900906816124916\n",
      "Epoch:46 Batch:240 Log-loss0.05632900819182396\n",
      "Epoch:46 Batch:320 Log-loss0.03227293863892555\n",
      "Epoch:46 Batch:400 Log-loss0.04162149876356125\n",
      "Epoch:46 Batch:480 Log-loss0.03390781953930855\n",
      "Epoch:46 Batch:560 Log-loss0.036290280520915985\n",
      "Epoch average log loss:0.04027924213026251\n",
      "In Epoch46, val_loss:0.038474212043237165, best_val_loss:0.038408574866939736, best_auc:0.9908254612334152\n",
      "Epoch:47 Batch:0 Log-loss0.04520173743367195\n",
      "Epoch:47 Batch:80 Log-loss0.05066676810383797\n",
      "Epoch:47 Batch:160 Log-loss0.0678756907582283\n",
      "Epoch:47 Batch:240 Log-loss0.04197247326374054\n",
      "Epoch:47 Batch:320 Log-loss0.055225979536771774\n",
      "Epoch:47 Batch:400 Log-loss0.05077444016933441\n",
      "Epoch:47 Batch:480 Log-loss0.04270404949784279\n",
      "Epoch:47 Batch:560 Log-loss0.045640408992767334\n",
      "Epoch average log loss:0.040102765732444826\n",
      "In Epoch47, val_loss:0.03842010183071474, best_val_loss:0.038408574866939736, best_auc:0.9908254612334152\n",
      "Epoch:48 Batch:0 Log-loss0.035916659981012344\n",
      "Epoch:48 Batch:80 Log-loss0.046045321971178055\n",
      "Epoch:48 Batch:160 Log-loss0.029186246916651726\n",
      "Epoch:48 Batch:240 Log-loss0.040287237614393234\n",
      "Epoch:48 Batch:320 Log-loss0.04408605024218559\n",
      "Epoch:48 Batch:400 Log-loss0.03384741395711899\n",
      "Epoch:48 Batch:480 Log-loss0.051045749336481094\n",
      "Epoch:48 Batch:560 Log-loss0.04844367504119873\n",
      "Epoch average log loss:0.04015889416249203\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch48, val_loss:0.038397548713681534, best_val_loss:0.038397548713681534, best_auc:0.990822527051212\n",
      "Epoch:49 Batch:0 Log-loss0.02466319315135479\n",
      "Epoch:49 Batch:80 Log-loss0.03610636666417122\n",
      "Epoch:49 Batch:160 Log-loss0.049409348517656326\n",
      "Epoch:49 Batch:240 Log-loss0.05615459010004997\n",
      "Epoch:49 Batch:320 Log-loss0.04064323008060455\n",
      "Epoch:49 Batch:400 Log-loss0.03436728194355965\n",
      "Epoch:49 Batch:480 Log-loss0.04315038025379181\n",
      "Epoch:49 Batch:560 Log-loss0.03754863142967224\n",
      "Epoch average log loss:0.03994406895778541\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch49, val_loss:0.038354254972110584, best_val_loss:0.038354254972110584, best_auc:0.9908450376559544\n",
      "Epoch:50 Batch:0 Log-loss0.048050787299871445\n",
      "Epoch:50 Batch:80 Log-loss0.04022941738367081\n",
      "Epoch:50 Batch:160 Log-loss0.04924142360687256\n",
      "Epoch:50 Batch:240 Log-loss0.044423479586839676\n",
      "Epoch:50 Batch:320 Log-loss0.05609357729554176\n",
      "Epoch:50 Batch:400 Log-loss0.055956896394491196\n",
      "Epoch:50 Batch:480 Log-loss0.04728851094841957\n",
      "Epoch:50 Batch:560 Log-loss0.03505021333694458\n",
      "Epoch average log loss:0.03991138408891857\n",
      "In Epoch50, val_loss:0.03845577189326979, best_val_loss:0.038354254972110584, best_auc:0.9908450376559544\n",
      "Epoch:51 Batch:0 Log-loss0.0170005913823843\n",
      "Epoch:51 Batch:80 Log-loss0.0266911331564188\n",
      "Epoch:51 Batch:160 Log-loss0.03270600736141205\n",
      "Epoch:51 Batch:240 Log-loss0.04208117723464966\n",
      "Epoch:51 Batch:320 Log-loss0.04928423464298248\n",
      "Epoch:51 Batch:400 Log-loss0.04069044440984726\n",
      "Epoch:51 Batch:480 Log-loss0.03524056449532509\n",
      "Epoch:51 Batch:560 Log-loss0.0444987453520298\n",
      "Epoch average log loss:0.03999818456691823\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch51, val_loss:0.03834938871655096, best_val_loss:0.03834938871655096, best_auc:0.9908549089774583\n",
      "Epoch:52 Batch:0 Log-loss0.03812346234917641\n",
      "Epoch:52 Batch:80 Log-loss0.0493486262857914\n",
      "Epoch:52 Batch:160 Log-loss0.04537871852517128\n",
      "Epoch:52 Batch:240 Log-loss0.03330189362168312\n",
      "Epoch:52 Batch:320 Log-loss0.03209340199828148\n",
      "Epoch:52 Batch:400 Log-loss0.03887934237718582\n",
      "Epoch:52 Batch:480 Log-loss0.03473256155848503\n",
      "Epoch:52 Batch:560 Log-loss0.032319311052560806\n",
      "Epoch average log loss:0.040124357614799266\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch52, val_loss:0.03831481691267099, best_val_loss:0.03831481691267099, best_auc:0.9908755025808195\n",
      "Epoch:53 Batch:0 Log-loss0.038725849241018295\n",
      "Epoch:53 Batch:80 Log-loss0.036365240812301636\n",
      "Epoch:53 Batch:160 Log-loss0.03699887916445732\n",
      "Epoch:53 Batch:240 Log-loss0.04720919951796532\n",
      "Epoch:53 Batch:320 Log-loss0.03628170117735863\n",
      "Epoch:53 Batch:400 Log-loss0.0453486405313015\n",
      "Epoch:53 Batch:480 Log-loss0.046915654093027115\n",
      "Epoch:53 Batch:560 Log-loss0.03795360028743744\n",
      "Epoch average log loss:0.03964081247270639\n",
      "In Epoch53, val_loss:0.03838937018162709, best_val_loss:0.03831481691267099, best_auc:0.9908755025808195\n",
      "Epoch:54 Batch:0 Log-loss0.03716912493109703\n",
      "Epoch:54 Batch:80 Log-loss0.028542326763272285\n",
      "Epoch:54 Batch:160 Log-loss0.0380343534052372\n",
      "Epoch:54 Batch:240 Log-loss0.028798295184969902\n",
      "Epoch:54 Batch:320 Log-loss0.03923039510846138\n",
      "Epoch:54 Batch:400 Log-loss0.03870384022593498\n",
      "Epoch:54 Batch:480 Log-loss0.031124375760555267\n",
      "Epoch:54 Batch:560 Log-loss0.03629594296216965\n",
      "Epoch average log loss:0.03989505729571517\n",
      "In Epoch54, val_loss:0.03835342708964987, best_val_loss:0.03831481691267099, best_auc:0.9908755025808195\n",
      "Epoch:55 Batch:0 Log-loss0.05013803765177727\n",
      "Epoch:55 Batch:80 Log-loss0.05067276954650879\n",
      "Epoch:55 Batch:160 Log-loss0.04453165829181671\n",
      "Epoch:55 Batch:240 Log-loss0.030540362000465393\n",
      "Epoch:55 Batch:320 Log-loss0.02761034108698368\n",
      "Epoch:55 Batch:400 Log-loss0.02949812449514866\n",
      "Epoch:55 Batch:480 Log-loss0.03248900547623634\n",
      "Epoch:55 Batch:560 Log-loss0.03272109478712082\n",
      "Epoch average log loss:0.03986262103343116\n",
      "In Epoch55, val_loss:0.03844391172326154, best_val_loss:0.03831481691267099, best_auc:0.9908755025808195\n",
      "Epoch:56 Batch:0 Log-loss0.029446005821228027\n",
      "Epoch:56 Batch:80 Log-loss0.03305285423994064\n",
      "Epoch:56 Batch:160 Log-loss0.04072464630007744\n",
      "Epoch:56 Batch:240 Log-loss0.03623173013329506\n",
      "Epoch:56 Batch:320 Log-loss0.02778259478509426\n",
      "Epoch:56 Batch:400 Log-loss0.01969893090426922\n",
      "Epoch:56 Batch:480 Log-loss0.039845384657382965\n",
      "Epoch:56 Batch:560 Log-loss0.03934039920568466\n",
      "Epoch average log loss:0.039823170089428976\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch56, val_loss:0.03828459900954433, best_val_loss:0.03828459900954433, best_auc:0.9908638921770435\n",
      "Epoch:57 Batch:0 Log-loss0.03453580662608147\n",
      "Epoch:57 Batch:80 Log-loss0.04143950715661049\n",
      "Epoch:57 Batch:160 Log-loss0.038236718624830246\n",
      "Epoch:57 Batch:240 Log-loss0.0360349602997303\n",
      "Epoch:57 Batch:320 Log-loss0.03371330350637436\n",
      "Epoch:57 Batch:400 Log-loss0.03742639720439911\n",
      "Epoch:57 Batch:480 Log-loss0.029500281438231468\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:57 Batch:560 Log-loss0.038082998245954514\n",
      "Epoch average log loss:0.04002621650496232\n",
      "In Epoch57, val_loss:0.038357738991957685, best_val_loss:0.03828459900954433, best_auc:0.9908638921770435\n",
      "Epoch:58 Batch:0 Log-loss0.054733142256736755\n",
      "Epoch:58 Batch:80 Log-loss0.05223669111728668\n",
      "Epoch:58 Batch:160 Log-loss0.03271069377660751\n",
      "Epoch:58 Batch:240 Log-loss0.04749557003378868\n",
      "Epoch:58 Batch:320 Log-loss0.032637711614370346\n",
      "Epoch:58 Batch:400 Log-loss0.033266108483076096\n",
      "Epoch:58 Batch:480 Log-loss0.03806423768401146\n",
      "Epoch:58 Batch:560 Log-loss0.046329859644174576\n",
      "Epoch average log loss:0.039877284609247\n",
      "In Epoch58, val_loss:0.03837416896985276, best_val_loss:0.03828459900954433, best_auc:0.9908638921770435\n",
      "Epoch:59 Batch:0 Log-loss0.036391668021678925\n",
      "Epoch:59 Batch:80 Log-loss0.0391409732401371\n",
      "Epoch:59 Batch:160 Log-loss0.043105658143758774\n",
      "Epoch:59 Batch:240 Log-loss0.03191642090678215\n",
      "Epoch:59 Batch:320 Log-loss0.030474543571472168\n",
      "Epoch:59 Batch:400 Log-loss0.0457703173160553\n",
      "Epoch:59 Batch:480 Log-loss0.03319797292351723\n",
      "Epoch:59 Batch:560 Log-loss0.03101334162056446\n",
      "Epoch average log loss:0.03993816308544151\n",
      "In Epoch59, val_loss:0.038323874487419966, best_val_loss:0.03828459900954433, best_auc:0.9908638921770435\n",
      "Epoch:60 Batch:0 Log-loss0.03728875517845154\n",
      "Epoch:60 Batch:80 Log-loss0.036585550755262375\n",
      "Epoch:60 Batch:160 Log-loss0.05354909226298332\n",
      "Epoch:60 Batch:240 Log-loss0.040405988693237305\n",
      "Epoch:60 Batch:320 Log-loss0.04280184581875801\n",
      "Epoch:60 Batch:400 Log-loss0.035289961844682693\n",
      "Epoch:60 Batch:480 Log-loss0.03709724545478821\n",
      "Epoch:60 Batch:560 Log-loss0.03663504868745804\n",
      "Epoch average log loss:0.039569377300462553\n",
      "In Epoch60, val_loss:0.03832851609872697, best_val_loss:0.03828459900954433, best_auc:0.9908638921770435\n",
      "Epoch:61 Batch:0 Log-loss0.030437082052230835\n",
      "Epoch:61 Batch:80 Log-loss0.04082083702087402\n",
      "Epoch:61 Batch:160 Log-loss0.03623935580253601\n",
      "Epoch:61 Batch:240 Log-loss0.02324453927576542\n",
      "Epoch:61 Batch:320 Log-loss0.04014882072806358\n",
      "Epoch:61 Batch:400 Log-loss0.031696517020463943\n",
      "Epoch:61 Batch:480 Log-loss0.04332570731639862\n",
      "Epoch:61 Batch:560 Log-loss0.04130036383867264\n",
      "Epoch average log loss:0.03967910939494946\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch61, val_loss:0.0382845167122374, best_val_loss:0.0382845167122374, best_auc:0.9908678435726084\n",
      "Epoch:62 Batch:0 Log-loss0.03674718365073204\n",
      "Epoch:62 Batch:80 Log-loss0.045680392533540726\n",
      "Epoch:62 Batch:160 Log-loss0.034910496324300766\n",
      "Epoch:62 Batch:240 Log-loss0.06034219264984131\n",
      "Epoch:62 Batch:320 Log-loss0.03367512300610542\n",
      "Epoch:62 Batch:400 Log-loss0.030080199241638184\n",
      "Epoch:62 Batch:480 Log-loss0.021111777052283287\n",
      "Epoch:62 Batch:560 Log-loss0.05750536918640137\n",
      "Epoch average log loss:0.03974645281144019\n",
      "In Epoch62, val_loss:0.038355192167474905, best_val_loss:0.0382845167122374, best_auc:0.9908678435726084\n",
      "Epoch:63 Batch:0 Log-loss0.046937476843595505\n",
      "Epoch:63 Batch:80 Log-loss0.03210289403796196\n",
      "Epoch:63 Batch:160 Log-loss0.05404241755604744\n",
      "Epoch:63 Batch:240 Log-loss0.04819900169968605\n",
      "Epoch:63 Batch:320 Log-loss0.028742574155330658\n",
      "Epoch:63 Batch:400 Log-loss0.04660935699939728\n",
      "Epoch:63 Batch:480 Log-loss0.04051799327135086\n",
      "Epoch:63 Batch:560 Log-loss0.06276846677064896\n",
      "Epoch average log loss:0.03982161962693291\n",
      "In Epoch63, val_loss:0.038299983944484944, best_val_loss:0.0382845167122374, best_auc:0.9908678435726084\n",
      "Epoch:64 Batch:0 Log-loss0.028639769181609154\n",
      "Epoch:64 Batch:80 Log-loss0.03046574257314205\n",
      "Epoch:64 Batch:160 Log-loss0.03372468426823616\n",
      "Epoch:64 Batch:240 Log-loss0.04767315089702606\n",
      "Epoch:64 Batch:320 Log-loss0.03808104246854782\n",
      "Epoch:64 Batch:400 Log-loss0.03225655108690262\n",
      "Epoch:64 Batch:480 Log-loss0.04355284944176674\n",
      "Epoch:64 Batch:560 Log-loss0.035347867757081985\n",
      "Epoch average log loss:0.039724644892183796\n",
      "In Epoch64, val_loss:0.03831407569091281, best_val_loss:0.0382845167122374, best_auc:0.9908678435726084\n",
      "Epoch:65 Batch:0 Log-loss0.039809610694646835\n",
      "Epoch:65 Batch:80 Log-loss0.03501826152205467\n",
      "Epoch:65 Batch:160 Log-loss0.039117202162742615\n",
      "Epoch:65 Batch:240 Log-loss0.058469731360673904\n",
      "Epoch:65 Batch:320 Log-loss0.043492287397384644\n",
      "Epoch:65 Batch:400 Log-loss0.03430807217955589\n",
      "Epoch:65 Batch:480 Log-loss0.03461526706814766\n",
      "Epoch:65 Batch:560 Log-loss0.03843122720718384\n",
      "Epoch average log loss:0.03960103899506586\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch65, val_loss:0.038262037034503184, best_val_loss:0.038262037034503184, best_auc:0.9908868824257685\n",
      "Epoch:66 Batch:0 Log-loss0.039395224303007126\n",
      "Epoch:66 Batch:80 Log-loss0.03566495701670647\n",
      "Epoch:66 Batch:160 Log-loss0.03672874718904495\n",
      "Epoch:66 Batch:240 Log-loss0.03374646231532097\n",
      "Epoch:66 Batch:320 Log-loss0.036557216197252274\n",
      "Epoch:66 Batch:400 Log-loss0.03974961116909981\n",
      "Epoch:66 Batch:480 Log-loss0.03811859339475632\n",
      "Epoch:66 Batch:560 Log-loss0.021075258031487465\n",
      "Epoch average log loss:0.03975110898526119\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch66, val_loss:0.03825401620105184, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:67 Batch:0 Log-loss0.04157667234539986\n",
      "Epoch:67 Batch:80 Log-loss0.043486058712005615\n",
      "Epoch:67 Batch:160 Log-loss0.04137241467833519\n",
      "Epoch:67 Batch:240 Log-loss0.027312790974974632\n",
      "Epoch:67 Batch:320 Log-loss0.047757696360349655\n",
      "Epoch:67 Batch:400 Log-loss0.04218070209026337\n",
      "Epoch:67 Batch:480 Log-loss0.03699542209506035\n",
      "Epoch:67 Batch:560 Log-loss0.0376073457300663\n",
      "Epoch average log loss:0.03982504159045805\n",
      "In Epoch67, val_loss:0.038256023136055035, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:68 Batch:0 Log-loss0.04761864244937897\n",
      "Epoch:68 Batch:80 Log-loss0.033787801861763\n",
      "Epoch:68 Batch:160 Log-loss0.0570414662361145\n",
      "Epoch:68 Batch:240 Log-loss0.04403800144791603\n",
      "Epoch:68 Batch:320 Log-loss0.03413476049900055\n",
      "Epoch:68 Batch:400 Log-loss0.037568338215351105\n",
      "Epoch:68 Batch:480 Log-loss0.04133624956011772\n",
      "Epoch:68 Batch:560 Log-loss0.045596376061439514\n",
      "Epoch average log loss:0.03995829263835081\n",
      "In Epoch68, val_loss:0.03827189310628542, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:69 Batch:0 Log-loss0.03868239372968674\n",
      "Epoch:69 Batch:80 Log-loss0.039700377732515335\n",
      "Epoch:69 Batch:160 Log-loss0.0499044768512249\n",
      "Epoch:69 Batch:240 Log-loss0.036002736538648605\n",
      "Epoch:69 Batch:320 Log-loss0.04131050780415535\n",
      "Epoch:69 Batch:400 Log-loss0.029523789882659912\n",
      "Epoch:69 Batch:480 Log-loss0.03450974449515343\n",
      "Epoch:69 Batch:560 Log-loss0.045742280781269073\n",
      "Epoch average log loss:0.03969423891311245\n",
      "In Epoch69, val_loss:0.038260111683563094, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:70 Batch:0 Log-loss0.02132764272391796\n",
      "Epoch:70 Batch:80 Log-loss0.03362442925572395\n",
      "Epoch:70 Batch:160 Log-loss0.05077865347266197\n",
      "Epoch:70 Batch:240 Log-loss0.03648972138762474\n",
      "Epoch:70 Batch:320 Log-loss0.06509125232696533\n",
      "Epoch:70 Batch:400 Log-loss0.0396064929664135\n",
      "Epoch:70 Batch:480 Log-loss0.04667418822646141\n",
      "Epoch:70 Batch:560 Log-loss0.04163548722863197\n",
      "Epoch average log loss:0.03966989533510059\n",
      "In Epoch70, val_loss:0.03830008488469883, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:71 Batch:0 Log-loss0.042409587651491165\n",
      "Epoch:71 Batch:80 Log-loss0.04522085562348366\n",
      "Epoch:71 Batch:160 Log-loss0.04868711158633232\n",
      "Epoch:71 Batch:240 Log-loss0.039958406239748\n",
      "Epoch:71 Batch:320 Log-loss0.029635315760970116\n",
      "Epoch:71 Batch:400 Log-loss0.04448985680937767\n",
      "Epoch:71 Batch:480 Log-loss0.02360643446445465\n",
      "Epoch:71 Batch:560 Log-loss0.038928303867578506\n",
      "Epoch average log loss:0.03972614187680717\n",
      "In Epoch71, val_loss:0.038288614239940955, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:72 Batch:0 Log-loss0.03796300292015076\n",
      "Epoch:72 Batch:80 Log-loss0.040217075496912\n",
      "Epoch:72 Batch:160 Log-loss0.03348783776164055\n",
      "Epoch:72 Batch:240 Log-loss0.03960280865430832\n",
      "Epoch:72 Batch:320 Log-loss0.03472728654742241\n",
      "Epoch:72 Batch:400 Log-loss0.04031243920326233\n",
      "Epoch:72 Batch:480 Log-loss0.043156351894140244\n",
      "Epoch:72 Batch:560 Log-loss0.03508937731385231\n",
      "Epoch average log loss:0.039580755408055016\n",
      "In Epoch72, val_loss:0.038306009611746517, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:73 Batch:0 Log-loss0.037563033401966095\n",
      "Epoch:73 Batch:80 Log-loss0.022351861000061035\n",
      "Epoch:73 Batch:160 Log-loss0.033470846712589264\n",
      "Epoch:73 Batch:240 Log-loss0.04191089794039726\n",
      "Epoch:73 Batch:320 Log-loss0.031777553260326385\n",
      "Epoch:73 Batch:400 Log-loss0.03203774616122246\n",
      "Epoch:73 Batch:480 Log-loss0.055559661239385605\n",
      "Epoch:73 Batch:560 Log-loss0.04367543011903763\n",
      "Epoch average log loss:0.039636465038971176\n",
      "In Epoch73, val_loss:0.03829174023995011, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:74 Batch:0 Log-loss0.03933379799127579\n",
      "Epoch:74 Batch:80 Log-loss0.041197191923856735\n",
      "Epoch:74 Batch:160 Log-loss0.03193904459476471\n",
      "Epoch:74 Batch:240 Log-loss0.03898513689637184\n",
      "Epoch:74 Batch:320 Log-loss0.028872095048427582\n",
      "Epoch:74 Batch:400 Log-loss0.04453450068831444\n",
      "Epoch:74 Batch:480 Log-loss0.05351494252681732\n",
      "Epoch:74 Batch:560 Log-loss0.03863035514950752\n",
      "Epoch average log loss:0.03952039785550109\n",
      "In Epoch74, val_loss:0.03829385247031482, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:75 Batch:0 Log-loss0.027753861621022224\n",
      "Epoch:75 Batch:80 Log-loss0.037525370717048645\n",
      "Epoch:75 Batch:160 Log-loss0.037916239351034164\n",
      "Epoch:75 Batch:240 Log-loss0.029860978946089745\n",
      "Epoch:75 Batch:320 Log-loss0.037465717643499374\n",
      "Epoch:75 Batch:400 Log-loss0.03928009048104286\n",
      "Epoch:75 Batch:480 Log-loss0.03821396455168724\n",
      "Epoch:75 Batch:560 Log-loss0.043628644198179245\n",
      "Epoch average log loss:0.03976437993613737\n",
      "In Epoch75, val_loss:0.038279021926971836, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:76 Batch:0 Log-loss0.04039270803332329\n",
      "Epoch:76 Batch:80 Log-loss0.048846885561943054\n",
      "Epoch:76 Batch:160 Log-loss0.03741889074444771\n",
      "Epoch:76 Batch:240 Log-loss0.050823260098695755\n",
      "Epoch:76 Batch:320 Log-loss0.040948592126369476\n",
      "Epoch:76 Batch:400 Log-loss0.03613412380218506\n",
      "Epoch:76 Batch:480 Log-loss0.04104011878371239\n",
      "Epoch:76 Batch:560 Log-loss0.032616205513477325\n",
      "Epoch average log loss:0.03972431370722396\n",
      "In Epoch76, val_loss:0.03828261432594405, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:77 Batch:0 Log-loss0.046432074159383774\n",
      "Epoch:77 Batch:80 Log-loss0.04068287834525108\n",
      "Epoch:77 Batch:160 Log-loss0.03549571707844734\n",
      "Epoch:77 Batch:240 Log-loss0.025858616456389427\n",
      "Epoch:77 Batch:320 Log-loss0.05993804708123207\n",
      "Epoch:77 Batch:400 Log-loss0.03791633993387222\n",
      "Epoch:77 Batch:480 Log-loss0.026899166405200958\n",
      "Epoch:77 Batch:560 Log-loss0.03887471184134483\n",
      "Epoch average log loss:0.039627330906556124\n",
      "In Epoch77, val_loss:0.03829070292997051, best_val_loss:0.03825401620105184, best_auc:0.9908820199505582\n",
      "Epoch:78 Batch:0 Log-loss0.04740040376782417\n",
      "Epoch:78 Batch:80 Log-loss0.020263483747839928\n",
      "Epoch:78 Batch:160 Log-loss0.040489789098501205\n",
      "Epoch:78 Batch:240 Log-loss0.03770133852958679\n",
      "Epoch:78 Batch:320 Log-loss0.04792563244700432\n",
      "Epoch:78 Batch:400 Log-loss0.0332830436527729\n",
      "Epoch:78 Batch:480 Log-loss0.05438195541501045\n",
      "Epoch:78 Batch:560 Log-loss0.03287345543503761\n",
      "Epoch average log loss:0.03953571086084204\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_642.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 3\n",
      "Epoch:1 Batch:0 Log-loss0.7025661468505859\n",
      "Epoch:1 Batch:80 Log-loss0.13886696100234985\n",
      "Epoch:1 Batch:160 Log-loss0.04273459315299988\n",
      "Epoch:1 Batch:240 Log-loss0.09659724682569504\n",
      "Epoch:1 Batch:320 Log-loss0.05226079747080803\n",
      "Epoch:1 Batch:400 Log-loss0.06841335445642471\n",
      "Epoch:1 Batch:480 Log-loss0.05825486406683922\n",
      "Epoch:1 Batch:560 Log-loss0.06794405728578568\n",
      "Epoch average log loss:0.08657555095518805\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.054499716560085926, best_val_loss:0.054499716560085926, best_auc:0.9668120589159925\n",
      "Epoch:2 Batch:0 Log-loss0.048736006021499634\n",
      "Epoch:2 Batch:80 Log-loss0.06431032717227936\n",
      "Epoch:2 Batch:160 Log-loss0.05309044197201729\n",
      "Epoch:2 Batch:240 Log-loss0.047755707055330276\n",
      "Epoch:2 Batch:320 Log-loss0.04963376745581627\n",
      "Epoch:2 Batch:400 Log-loss0.05388284847140312\n",
      "Epoch:2 Batch:480 Log-loss0.03965809568762779\n",
      "Epoch:2 Batch:560 Log-loss0.05548763647675514\n",
      "Epoch average log loss:0.052864760157120015\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.05091267820132015, best_val_loss:0.05091267820132015, best_auc:0.9735173817351495\n",
      "Epoch:3 Batch:0 Log-loss0.05747039243578911\n",
      "Epoch:3 Batch:80 Log-loss0.059174638241529465\n",
      "Epoch:3 Batch:160 Log-loss0.048981934785842896\n",
      "Epoch:3 Batch:240 Log-loss0.04908977821469307\n",
      "Epoch:3 Batch:320 Log-loss0.05933251604437828\n",
      "Epoch:3 Batch:400 Log-loss0.04172428324818611\n",
      "Epoch:3 Batch:480 Log-loss0.04593749716877937\n",
      "Epoch:3 Batch:560 Log-loss0.05400964617729187\n",
      "Epoch average log loss:0.05002751033420542\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.046582272539942134, best_val_loss:0.046582272539942134, best_auc:0.9773803736657566\n",
      "Epoch:4 Batch:0 Log-loss0.06538286060094833\n",
      "Epoch:4 Batch:80 Log-loss0.044673893600702286\n",
      "Epoch:4 Batch:160 Log-loss0.040520720183849335\n",
      "Epoch:4 Batch:240 Log-loss0.03924522548913956\n",
      "Epoch:4 Batch:320 Log-loss0.0460764579474926\n",
      "Epoch:4 Batch:400 Log-loss0.046759847551584244\n",
      "Epoch:4 Batch:480 Log-loss0.037664465606212616\n",
      "Epoch:4 Batch:560 Log-loss0.03938416391611099\n",
      "Epoch average log loss:0.04738885180226394\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.04530385584285468, best_val_loss:0.04530385584285468, best_auc:0.9803603607468445\n",
      "Epoch:5 Batch:0 Log-loss0.04087410494685173\n",
      "Epoch:5 Batch:80 Log-loss0.06096510961651802\n",
      "Epoch:5 Batch:160 Log-loss0.046019297093153\n",
      "Epoch:5 Batch:240 Log-loss0.03121294267475605\n",
      "Epoch:5 Batch:320 Log-loss0.04322044178843498\n",
      "Epoch:5 Batch:400 Log-loss0.04899213835597038\n",
      "Epoch:5 Batch:480 Log-loss0.050337743014097214\n",
      "Epoch:5 Batch:560 Log-loss0.046093087643384933\n",
      "Epoch average log loss:0.04545355086613979\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.04413613307302605, best_val_loss:0.04413613307302605, best_auc:0.9819436197861663\n",
      "Epoch:6 Batch:0 Log-loss0.05835539475083351\n",
      "Epoch:6 Batch:80 Log-loss0.032655876129865646\n",
      "Epoch:6 Batch:160 Log-loss0.04984517768025398\n",
      "Epoch:6 Batch:240 Log-loss0.04499933496117592\n",
      "Epoch:6 Batch:320 Log-loss0.05956270173192024\n",
      "Epoch:6 Batch:400 Log-loss0.032884374260902405\n",
      "Epoch:6 Batch:480 Log-loss0.04173871874809265\n",
      "Epoch:6 Batch:560 Log-loss0.047067951411008835\n",
      "Epoch average log loss:0.044102847702535135\n",
      "In Epoch6, val_loss:0.04448686296641902, best_val_loss:0.04413613307302605, best_auc:0.9819436197861663\n",
      "Epoch:7 Batch:0 Log-loss0.045841094106435776\n",
      "Epoch:7 Batch:80 Log-loss0.03888612240552902\n",
      "Epoch:7 Batch:160 Log-loss0.04490029439330101\n",
      "Epoch:7 Batch:240 Log-loss0.0409068688750267\n",
      "Epoch:7 Batch:320 Log-loss0.037851590663194656\n",
      "Epoch:7 Batch:400 Log-loss0.028994789347052574\n",
      "Epoch:7 Batch:480 Log-loss0.07178381830453873\n",
      "Epoch:7 Batch:560 Log-loss0.03319898620247841\n",
      "Epoch average log loss:0.043402095149005095\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.04291851693182191, best_val_loss:0.04291851693182191, best_auc:0.9841725771843354\n",
      "Epoch:8 Batch:0 Log-loss0.034047115594148636\n",
      "Epoch:8 Batch:80 Log-loss0.059486884623765945\n",
      "Epoch:8 Batch:160 Log-loss0.05721089616417885\n",
      "Epoch:8 Batch:240 Log-loss0.039188411086797714\n",
      "Epoch:8 Batch:320 Log-loss0.02943357080221176\n",
      "Epoch:8 Batch:400 Log-loss0.05595755577087402\n",
      "Epoch:8 Batch:480 Log-loss0.022103486582636833\n",
      "Epoch:8 Batch:560 Log-loss0.02891804836690426\n",
      "Epoch average log loss:0.04285677559673786\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.042618333163786666, best_val_loss:0.042618333163786666, best_auc:0.9843822466395635\n",
      "Epoch:9 Batch:0 Log-loss0.03040611930191517\n",
      "Epoch:9 Batch:80 Log-loss0.045506518334150314\n",
      "Epoch:9 Batch:160 Log-loss0.03708621859550476\n",
      "Epoch:9 Batch:240 Log-loss0.041095007210969925\n",
      "Epoch:9 Batch:320 Log-loss0.041118040680885315\n",
      "Epoch:9 Batch:400 Log-loss0.038357023149728775\n",
      "Epoch:9 Batch:480 Log-loss0.04492565616965294\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:9 Batch:560 Log-loss0.039844170212745667\n",
      "Epoch average log loss:0.042132799769751726\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.0425120992683235, best_val_loss:0.0425120992683235, best_auc:0.985032869551759\n",
      "Epoch:10 Batch:0 Log-loss0.04476436972618103\n",
      "Epoch:10 Batch:80 Log-loss0.05466615781188011\n",
      "Epoch:10 Batch:160 Log-loss0.050803884863853455\n",
      "Epoch:10 Batch:240 Log-loss0.047501251101493835\n",
      "Epoch:10 Batch:320 Log-loss0.037773046642541885\n",
      "Epoch:10 Batch:400 Log-loss0.039184629917144775\n",
      "Epoch:10 Batch:480 Log-loss0.04436181113123894\n",
      "Epoch:10 Batch:560 Log-loss0.036837633699178696\n",
      "Epoch average log loss:0.04175305679839637\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.04176219172849749, best_val_loss:0.04176219172849749, best_auc:0.985760322660381\n",
      "Epoch:11 Batch:0 Log-loss0.04284526780247688\n",
      "Epoch:11 Batch:80 Log-loss0.037643738090991974\n",
      "Epoch:11 Batch:160 Log-loss0.02908274717628956\n",
      "Epoch:11 Batch:240 Log-loss0.04754204675555229\n",
      "Epoch:11 Batch:320 Log-loss0.04939112067222595\n",
      "Epoch:11 Batch:400 Log-loss0.03371192887425423\n",
      "Epoch:11 Batch:480 Log-loss0.043375466018915176\n",
      "Epoch:11 Batch:560 Log-loss0.033834852278232574\n",
      "Epoch average log loss:0.04112429324138377\n",
      "In Epoch11, val_loss:0.041912726569860614, best_val_loss:0.04176219172849749, best_auc:0.985760322660381\n",
      "Epoch:12 Batch:0 Log-loss0.024884043261408806\n",
      "Epoch:12 Batch:80 Log-loss0.0309237539768219\n",
      "Epoch:12 Batch:160 Log-loss0.030954932793974876\n",
      "Epoch:12 Batch:240 Log-loss0.04892865940928459\n",
      "Epoch:12 Batch:320 Log-loss0.0408148393034935\n",
      "Epoch:12 Batch:400 Log-loss0.06717076897621155\n",
      "Epoch:12 Batch:480 Log-loss0.03075428307056427\n",
      "Epoch:12 Batch:560 Log-loss0.0322173647582531\n",
      "Epoch average log loss:0.04082544813032395\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.041659403775078256, best_val_loss:0.041659403775078256, best_auc:0.9867979194615758\n",
      "Epoch:13 Batch:0 Log-loss0.036158256232738495\n",
      "Epoch:13 Batch:80 Log-loss0.034130584448575974\n",
      "Epoch:13 Batch:160 Log-loss0.03686269000172615\n",
      "Epoch:13 Batch:240 Log-loss0.039073795080184937\n",
      "Epoch:13 Batch:320 Log-loss0.05649438872933388\n",
      "Epoch:13 Batch:400 Log-loss0.0472290925681591\n",
      "Epoch:13 Batch:480 Log-loss0.03630121797323227\n",
      "Epoch:13 Batch:560 Log-loss0.03158460184931755\n",
      "Epoch average log loss:0.040580303401553205\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch13, val_loss:0.041361808799877844, best_val_loss:0.041361808799877844, best_auc:0.9867758868337505\n",
      "Epoch:14 Batch:0 Log-loss0.03228659927845001\n",
      "Epoch:14 Batch:80 Log-loss0.041503068059682846\n",
      "Epoch:14 Batch:160 Log-loss0.027223968878388405\n",
      "Epoch:14 Batch:240 Log-loss0.030194131657481194\n",
      "Epoch:14 Batch:320 Log-loss0.04341960325837135\n",
      "Epoch:14 Batch:400 Log-loss0.028793228790163994\n",
      "Epoch:14 Batch:480 Log-loss0.03762732446193695\n",
      "Epoch:14 Batch:560 Log-loss0.03849809244275093\n",
      "Epoch average log loss:0.04023784174184714\n",
      "In Epoch14, val_loss:0.0420348815948737, best_val_loss:0.041361808799877844, best_auc:0.9867758868337505\n",
      "Epoch:15 Batch:0 Log-loss0.03196331486105919\n",
      "Epoch:15 Batch:80 Log-loss0.029300473630428314\n",
      "Epoch:15 Batch:160 Log-loss0.02717427909374237\n",
      "Epoch:15 Batch:240 Log-loss0.0336838960647583\n",
      "Epoch:15 Batch:320 Log-loss0.038588810712099075\n",
      "Epoch:15 Batch:400 Log-loss0.07193969935178757\n",
      "Epoch:15 Batch:480 Log-loss0.03866741806268692\n",
      "Epoch:15 Batch:560 Log-loss0.04721210524439812\n",
      "Epoch average log loss:0.03997223835904151\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch15, val_loss:0.041225212885794636, best_val_loss:0.041225212885794636, best_auc:0.9871006264901041\n",
      "Epoch:16 Batch:0 Log-loss0.06306109577417374\n",
      "Epoch:16 Batch:80 Log-loss0.04408055916428566\n",
      "Epoch:16 Batch:160 Log-loss0.023611484095454216\n",
      "Epoch:16 Batch:240 Log-loss0.038421910256147385\n",
      "Epoch:16 Batch:320 Log-loss0.02555447816848755\n",
      "Epoch:16 Batch:400 Log-loss0.057042863219976425\n",
      "Epoch:16 Batch:480 Log-loss0.02741667442023754\n",
      "Epoch:16 Batch:560 Log-loss0.033268146216869354\n",
      "Epoch average log loss:0.03950580643889095\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch16, val_loss:0.041120751381074454, best_val_loss:0.041120751381074454, best_auc:0.9873921766626513\n",
      "Epoch:17 Batch:0 Log-loss0.04831891879439354\n",
      "Epoch:17 Batch:80 Log-loss0.04155659303069115\n",
      "Epoch:17 Batch:160 Log-loss0.04382264241576195\n",
      "Epoch:17 Batch:240 Log-loss0.04112276807427406\n",
      "Epoch:17 Batch:320 Log-loss0.03581080958247185\n",
      "Epoch:17 Batch:400 Log-loss0.04642648994922638\n",
      "Epoch:17 Batch:480 Log-loss0.03701026365160942\n",
      "Epoch:17 Batch:560 Log-loss0.04153243824839592\n",
      "Epoch average log loss:0.03938484998819019\n",
      "In Epoch17, val_loss:0.04128922882355674, best_val_loss:0.041120751381074454, best_auc:0.9873921766626513\n",
      "Epoch:18 Batch:0 Log-loss0.036365240812301636\n",
      "Epoch:18 Batch:80 Log-loss0.03249673917889595\n",
      "Epoch:18 Batch:160 Log-loss0.043931324034929276\n",
      "Epoch:18 Batch:240 Log-loss0.0322706438601017\n",
      "Epoch:18 Batch:320 Log-loss0.029746783897280693\n",
      "Epoch:18 Batch:400 Log-loss0.04946950450539589\n",
      "Epoch:18 Batch:480 Log-loss0.05036439374089241\n",
      "Epoch:18 Batch:560 Log-loss0.04573938250541687\n",
      "Epoch average log loss:0.03908885215475623\n",
      "In Epoch18, val_loss:0.04119091267194249, best_val_loss:0.041120751381074454, best_auc:0.9873921766626513\n",
      "Epoch:19 Batch:0 Log-loss0.033488254994153976\n",
      "Epoch:19 Batch:80 Log-loss0.04694831743836403\n",
      "Epoch:19 Batch:160 Log-loss0.032702166587114334\n",
      "Epoch:19 Batch:240 Log-loss0.03278351202607155\n",
      "Epoch:19 Batch:320 Log-loss0.03569015488028526\n",
      "Epoch:19 Batch:400 Log-loss0.02380097098648548\n",
      "Epoch:19 Batch:480 Log-loss0.03464842587709427\n",
      "Epoch:19 Batch:560 Log-loss0.04095062240958214\n",
      "Epoch average log loss:0.038836713607556056\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch19, val_loss:0.041106102817941734, best_val_loss:0.041106102817941734, best_auc:0.9877337934138389\n",
      "Epoch:20 Batch:0 Log-loss0.04218438267707825\n",
      "Epoch:20 Batch:80 Log-loss0.03231807053089142\n",
      "Epoch:20 Batch:160 Log-loss0.04154794290661812\n",
      "Epoch:20 Batch:240 Log-loss0.040158990770578384\n",
      "Epoch:20 Batch:320 Log-loss0.042938295751810074\n",
      "Epoch:20 Batch:400 Log-loss0.04359519109129906\n",
      "Epoch:20 Batch:480 Log-loss0.039133619517087936\n",
      "Epoch:20 Batch:560 Log-loss0.04592885076999664\n",
      "Epoch average log loss:0.038769438533511544\n",
      "In Epoch20, val_loss:0.041212822881722365, best_val_loss:0.041106102817941734, best_auc:0.9877337934138389\n",
      "Epoch:21 Batch:0 Log-loss0.039963241666555405\n",
      "Epoch:21 Batch:80 Log-loss0.040292005985975266\n",
      "Epoch:21 Batch:160 Log-loss0.04203158617019653\n",
      "Epoch:21 Batch:240 Log-loss0.02519661746919155\n",
      "Epoch:21 Batch:320 Log-loss0.02684885822236538\n",
      "Epoch:21 Batch:400 Log-loss0.04425441846251488\n",
      "Epoch:21 Batch:480 Log-loss0.04106795787811279\n",
      "Epoch:21 Batch:560 Log-loss0.050766680389642715\n",
      "Epoch average log loss:0.03872900190097945\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch21, val_loss:0.040868927607020246, best_val_loss:0.040868927607020246, best_auc:0.9879047948043408\n",
      "Epoch:22 Batch:0 Log-loss0.041300512850284576\n",
      "Epoch:22 Batch:80 Log-loss0.040711358189582825\n",
      "Epoch:22 Batch:160 Log-loss0.03269568458199501\n",
      "Epoch:22 Batch:240 Log-loss0.06048952415585518\n",
      "Epoch:22 Batch:320 Log-loss0.037532079964876175\n",
      "Epoch:22 Batch:400 Log-loss0.02695830725133419\n",
      "Epoch:22 Batch:480 Log-loss0.031347889453172684\n",
      "Epoch:22 Batch:560 Log-loss0.03820754587650299\n",
      "Epoch average log loss:0.038467666945819344\n",
      "In Epoch22, val_loss:0.0410074609826918, best_val_loss:0.040868927607020246, best_auc:0.9879047948043408\n",
      "Epoch:23 Batch:0 Log-loss0.041183214634656906\n",
      "Epoch:23 Batch:80 Log-loss0.044963207095861435\n",
      "Epoch:23 Batch:160 Log-loss0.029258044436573982\n",
      "Epoch:23 Batch:240 Log-loss0.04363206401467323\n",
      "Epoch:23 Batch:320 Log-loss0.03932119160890579\n",
      "Epoch:23 Batch:400 Log-loss0.04997287318110466\n",
      "Epoch:23 Batch:480 Log-loss0.036521509289741516\n",
      "Epoch:23 Batch:560 Log-loss0.04935849830508232\n",
      "Epoch average log loss:0.03836334795092366\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch23, val_loss:0.04071669930716485, best_val_loss:0.04071669930716485, best_auc:0.9880413748118803\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:24 Batch:0 Log-loss0.033515941351652145\n",
      "Epoch:24 Batch:80 Log-loss0.0496051125228405\n",
      "Epoch:24 Batch:160 Log-loss0.047146961092948914\n",
      "Epoch:24 Batch:240 Log-loss0.036767419427633286\n",
      "Epoch:24 Batch:320 Log-loss0.036778468638658524\n",
      "Epoch:24 Batch:400 Log-loss0.03485209867358208\n",
      "Epoch:24 Batch:480 Log-loss0.030108099803328514\n",
      "Epoch:24 Batch:560 Log-loss0.04030716419219971\n",
      "Epoch average log loss:0.0380753274208733\n",
      "In Epoch24, val_loss:0.04083952330894639, best_val_loss:0.04071669930716485, best_auc:0.9880413748118803\n",
      "Epoch:25 Batch:0 Log-loss0.03944004327058792\n",
      "Epoch:25 Batch:80 Log-loss0.03436566889286041\n",
      "Epoch:25 Batch:160 Log-loss0.04776051640510559\n",
      "Epoch:25 Batch:240 Log-loss0.030198633670806885\n",
      "Epoch:25 Batch:320 Log-loss0.05548122897744179\n",
      "Epoch:25 Batch:400 Log-loss0.059334516525268555\n",
      "Epoch:25 Batch:480 Log-loss0.046979594975709915\n",
      "Epoch:25 Batch:560 Log-loss0.04621219262480736\n",
      "Epoch average log loss:0.03798107707906248\n",
      "In Epoch25, val_loss:0.040817666684670274, best_val_loss:0.04071669930716485, best_auc:0.9880413748118803\n",
      "Epoch:26 Batch:0 Log-loss0.0440351702272892\n",
      "Epoch:26 Batch:80 Log-loss0.03699576482176781\n",
      "Epoch:26 Batch:160 Log-loss0.029536614194512367\n",
      "Epoch:26 Batch:240 Log-loss0.03032863885164261\n",
      "Epoch:26 Batch:320 Log-loss0.03927946463227272\n",
      "Epoch:26 Batch:400 Log-loss0.04029751196503639\n",
      "Epoch:26 Batch:480 Log-loss0.03120584599673748\n",
      "Epoch:26 Batch:560 Log-loss0.04144730046391487\n",
      "Epoch average log loss:0.038118135649710894\n",
      "In Epoch26, val_loss:0.04073235738509772, best_val_loss:0.04071669930716485, best_auc:0.9880413748118803\n",
      "Epoch:27 Batch:0 Log-loss0.05191117152571678\n",
      "Epoch:27 Batch:80 Log-loss0.03381313011050224\n",
      "Epoch:27 Batch:160 Log-loss0.023664606735110283\n",
      "Epoch:27 Batch:240 Log-loss0.04404868558049202\n",
      "Epoch:27 Batch:320 Log-loss0.035164956003427505\n",
      "Epoch:27 Batch:400 Log-loss0.027525702491402626\n",
      "Epoch:27 Batch:480 Log-loss0.039475198835134506\n",
      "Epoch:27 Batch:560 Log-loss0.03461230918765068\n",
      "Epoch average log loss:0.03789678359337683\n",
      "In Epoch27, val_loss:0.04073509697487447, best_val_loss:0.04071669930716485, best_auc:0.9880413748118803\n",
      "Epoch:28 Batch:0 Log-loss0.02755231224000454\n",
      "Epoch:28 Batch:80 Log-loss0.0256823617964983\n",
      "Epoch:28 Batch:160 Log-loss0.04934963583946228\n",
      "Epoch:28 Batch:240 Log-loss0.0460958369076252\n",
      "Epoch:28 Batch:320 Log-loss0.03252604231238365\n",
      "Epoch:28 Batch:400 Log-loss0.04359348490834236\n",
      "Epoch:28 Batch:480 Log-loss0.037342894822359085\n",
      "Epoch:28 Batch:560 Log-loss0.052261222153902054\n",
      "Epoch average log loss:0.037868250756790595\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch28, val_loss:0.040711116136355426, best_val_loss:0.040711116136355426, best_auc:0.9880581602412558\n",
      "Epoch:29 Batch:0 Log-loss0.026399990543723106\n",
      "Epoch:29 Batch:80 Log-loss0.021518096327781677\n",
      "Epoch:29 Batch:160 Log-loss0.032206930220127106\n",
      "Epoch:29 Batch:240 Log-loss0.03786751627922058\n",
      "Epoch:29 Batch:320 Log-loss0.04146299511194229\n",
      "Epoch:29 Batch:400 Log-loss0.017738139256834984\n",
      "Epoch:29 Batch:480 Log-loss0.02716086059808731\n",
      "Epoch:29 Batch:560 Log-loss0.03848279267549515\n",
      "Epoch average log loss:0.037813807262240776\n",
      "In Epoch29, val_loss:0.04101299527946724, best_val_loss:0.040711116136355426, best_auc:0.9880581602412558\n",
      "Epoch:30 Batch:0 Log-loss0.042602453380823135\n",
      "Epoch:30 Batch:80 Log-loss0.05265320837497711\n",
      "Epoch:30 Batch:160 Log-loss0.03456048294901848\n",
      "Epoch:30 Batch:240 Log-loss0.03541259840130806\n",
      "Epoch:30 Batch:320 Log-loss0.038714177906513214\n",
      "Epoch:30 Batch:400 Log-loss0.045074593275785446\n",
      "Epoch:30 Batch:480 Log-loss0.03228180482983589\n",
      "Epoch:30 Batch:560 Log-loss0.04161469265818596\n",
      "Epoch average log loss:0.0377790090162307\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch30, val_loss:0.04049713226431128, best_val_loss:0.04049713226431128, best_auc:0.9881498295792716\n",
      "Epoch:31 Batch:0 Log-loss0.02839360386133194\n",
      "Epoch:31 Batch:80 Log-loss0.03290253505110741\n",
      "Epoch:31 Batch:160 Log-loss0.031130924820899963\n",
      "Epoch:31 Batch:240 Log-loss0.02494913339614868\n",
      "Epoch:31 Batch:320 Log-loss0.04640522226691246\n",
      "Epoch:31 Batch:400 Log-loss0.03908328339457512\n",
      "Epoch:31 Batch:480 Log-loss0.03768030181527138\n",
      "Epoch:31 Batch:560 Log-loss0.030472200363874435\n",
      "Epoch average log loss:0.03762345414475671\n",
      "In Epoch31, val_loss:0.04116486975214914, best_val_loss:0.04049713226431128, best_auc:0.9881498295792716\n",
      "Epoch:32 Batch:0 Log-loss0.04109504073858261\n",
      "Epoch:32 Batch:80 Log-loss0.03098524920642376\n",
      "Epoch:32 Batch:160 Log-loss0.04012639448046684\n",
      "Epoch:32 Batch:240 Log-loss0.06324804574251175\n",
      "Epoch:32 Batch:320 Log-loss0.03060959279537201\n",
      "Epoch:32 Batch:400 Log-loss0.04252948984503746\n",
      "Epoch:32 Batch:480 Log-loss0.044706448912620544\n",
      "Epoch:32 Batch:560 Log-loss0.03060677833855152\n",
      "Epoch average log loss:0.03744300503416785\n",
      "In Epoch32, val_loss:0.04072122549968273, best_val_loss:0.04049713226431128, best_auc:0.9881498295792716\n",
      "Epoch:33 Batch:0 Log-loss0.026044344529509544\n",
      "Epoch:33 Batch:80 Log-loss0.042067285627126694\n",
      "Epoch:33 Batch:160 Log-loss0.033103469759225845\n",
      "Epoch:33 Batch:240 Log-loss0.0408799909055233\n",
      "Epoch:33 Batch:320 Log-loss0.04053608328104019\n",
      "Epoch:33 Batch:400 Log-loss0.030370034277439117\n",
      "Epoch:33 Batch:480 Log-loss0.02031981758773327\n",
      "Epoch:33 Batch:560 Log-loss0.04215182363986969\n",
      "Epoch average log loss:0.037633910728618505\n",
      "In Epoch33, val_loss:0.040661782591169794, best_val_loss:0.04049713226431128, best_auc:0.9881498295792716\n",
      "Epoch:34 Batch:0 Log-loss0.030530380085110664\n",
      "Epoch:34 Batch:80 Log-loss0.035132888704538345\n",
      "Epoch:34 Batch:160 Log-loss0.04642767086625099\n",
      "Epoch:34 Batch:240 Log-loss0.03997963294386864\n",
      "Epoch:34 Batch:320 Log-loss0.03660799562931061\n",
      "Epoch:34 Batch:400 Log-loss0.03841731324791908\n",
      "Epoch:34 Batch:480 Log-loss0.033199626952409744\n",
      "Epoch:34 Batch:560 Log-loss0.0317549854516983\n",
      "Epoch average log loss:0.03742453447276992\n",
      "In Epoch34, val_loss:0.04083963737290867, best_val_loss:0.04049713226431128, best_auc:0.9881498295792716\n",
      "Epoch:35 Batch:0 Log-loss0.040117114782333374\n",
      "Epoch:35 Batch:80 Log-loss0.05393627658486366\n",
      "Epoch:35 Batch:160 Log-loss0.042406320571899414\n",
      "Epoch:35 Batch:240 Log-loss0.030802100896835327\n",
      "Epoch:35 Batch:320 Log-loss0.02494143694639206\n",
      "Epoch:35 Batch:400 Log-loss0.03731435537338257\n",
      "Epoch:35 Batch:480 Log-loss0.04309879615902901\n",
      "Epoch:35 Batch:560 Log-loss0.038423389196395874\n",
      "Epoch average log loss:0.03742945708800107\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch35, val_loss:0.040408434608041964, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:36 Batch:0 Log-loss0.044640932232141495\n",
      "Epoch:36 Batch:80 Log-loss0.04867636039853096\n",
      "Epoch:36 Batch:160 Log-loss0.036374375224113464\n",
      "Epoch:36 Batch:240 Log-loss0.03101043403148651\n",
      "Epoch:36 Batch:320 Log-loss0.03480492904782295\n",
      "Epoch:36 Batch:400 Log-loss0.02437763661146164\n",
      "Epoch:36 Batch:480 Log-loss0.045107245445251465\n",
      "Epoch:36 Batch:560 Log-loss0.04907507076859474\n",
      "Epoch average log loss:0.037419874320871065\n",
      "In Epoch36, val_loss:0.04061342118858375, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:37 Batch:0 Log-loss0.0359097383916378\n",
      "Epoch:37 Batch:80 Log-loss0.031218020245432854\n",
      "Epoch:37 Batch:160 Log-loss0.036139119416475296\n",
      "Epoch:37 Batch:240 Log-loss0.03182648494839668\n",
      "Epoch:37 Batch:320 Log-loss0.05115559697151184\n",
      "Epoch:37 Batch:400 Log-loss0.03129522129893303\n",
      "Epoch:37 Batch:480 Log-loss0.05122798681259155\n",
      "Epoch:37 Batch:560 Log-loss0.03469708561897278\n",
      "Epoch average log loss:0.03730923623578357\n",
      "In Epoch37, val_loss:0.040518081900666074, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:38 Batch:0 Log-loss0.030520426109433174\n",
      "Epoch:38 Batch:80 Log-loss0.04682235047221184\n",
      "Epoch:38 Batch:160 Log-loss0.03040393628180027\n",
      "Epoch:38 Batch:240 Log-loss0.032237060368061066\n",
      "Epoch:38 Batch:320 Log-loss0.04582981392741203\n",
      "Epoch:38 Batch:400 Log-loss0.04409119114279747\n",
      "Epoch:38 Batch:480 Log-loss0.040465716272592545\n",
      "Epoch:38 Batch:560 Log-loss0.050166066735982895\n",
      "Epoch average log loss:0.03729359436713691\n",
      "In Epoch38, val_loss:0.040655582566609415, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:39 Batch:0 Log-loss0.023833995684981346\n",
      "Epoch:39 Batch:80 Log-loss0.021289929747581482\n",
      "Epoch:39 Batch:160 Log-loss0.025725841522216797\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:39 Batch:240 Log-loss0.05300792679190636\n",
      "Epoch:39 Batch:320 Log-loss0.05596470460295677\n",
      "Epoch:39 Batch:400 Log-loss0.035542842000722885\n",
      "Epoch:39 Batch:480 Log-loss0.04786532744765282\n",
      "Epoch:39 Batch:560 Log-loss0.04455173388123512\n",
      "Epoch average log loss:0.037376002039361214\n",
      "In Epoch39, val_loss:0.04056011578414893, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:40 Batch:0 Log-loss0.03857572376728058\n",
      "Epoch:40 Batch:80 Log-loss0.02615060657262802\n",
      "Epoch:40 Batch:160 Log-loss0.04137798771262169\n",
      "Epoch:40 Batch:240 Log-loss0.032091740518808365\n",
      "Epoch:40 Batch:320 Log-loss0.030469827353954315\n",
      "Epoch:40 Batch:400 Log-loss0.03673144057393074\n",
      "Epoch:40 Batch:480 Log-loss0.034315325319767\n",
      "Epoch:40 Batch:560 Log-loss0.03398074209690094\n",
      "Epoch average log loss:0.037060772055493935\n",
      "In Epoch40, val_loss:0.04067558193905147, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:41 Batch:0 Log-loss0.039796385914087296\n",
      "Epoch:41 Batch:80 Log-loss0.029372096061706543\n",
      "Epoch:41 Batch:160 Log-loss0.0421624630689621\n",
      "Epoch:41 Batch:240 Log-loss0.05640512704849243\n",
      "Epoch:41 Batch:320 Log-loss0.0456317663192749\n",
      "Epoch:41 Batch:400 Log-loss0.026678666472434998\n",
      "Epoch:41 Batch:480 Log-loss0.0380256325006485\n",
      "Epoch:41 Batch:560 Log-loss0.04382437840104103\n",
      "Epoch average log loss:0.03714740692604599\n",
      "In Epoch41, val_loss:0.04043202047367444, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:42 Batch:0 Log-loss0.043190840631723404\n",
      "Epoch:42 Batch:80 Log-loss0.027341710403561592\n",
      "Epoch:42 Batch:160 Log-loss0.03643542155623436\n",
      "Epoch:42 Batch:240 Log-loss0.0347498394548893\n",
      "Epoch:42 Batch:320 Log-loss0.03683769330382347\n",
      "Epoch:42 Batch:400 Log-loss0.02024926245212555\n",
      "Epoch:42 Batch:480 Log-loss0.03433546423912048\n",
      "Epoch:42 Batch:560 Log-loss0.03741346299648285\n",
      "Epoch average log loss:0.03714505009286638\n",
      "In Epoch42, val_loss:0.04062069057844347, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:43 Batch:0 Log-loss0.03583244979381561\n",
      "Epoch:43 Batch:80 Log-loss0.028308896347880363\n",
      "Epoch:43 Batch:160 Log-loss0.041097138077020645\n",
      "Epoch:43 Batch:240 Log-loss0.0366545207798481\n",
      "Epoch:43 Batch:320 Log-loss0.04388417676091194\n",
      "Epoch:43 Batch:400 Log-loss0.03183192014694214\n",
      "Epoch:43 Batch:480 Log-loss0.04176170751452446\n",
      "Epoch:43 Batch:560 Log-loss0.04855145514011383\n",
      "Epoch average log loss:0.03732147152269525\n",
      "In Epoch43, val_loss:0.04048836644440821, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:44 Batch:0 Log-loss0.0407634899020195\n",
      "Epoch:44 Batch:80 Log-loss0.026619456708431244\n",
      "Epoch:44 Batch:160 Log-loss0.04443873465061188\n",
      "Epoch:44 Batch:240 Log-loss0.032576784491539\n",
      "Epoch:44 Batch:320 Log-loss0.035649631172418594\n",
      "Epoch:44 Batch:400 Log-loss0.04032418504357338\n",
      "Epoch:44 Batch:480 Log-loss0.03763408958911896\n",
      "Epoch:44 Batch:560 Log-loss0.03209266439080238\n",
      "Epoch average log loss:0.037052385859923175\n",
      "In Epoch44, val_loss:0.04057453727144788, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:45 Batch:0 Log-loss0.03663984313607216\n",
      "Epoch:45 Batch:80 Log-loss0.037090595811605453\n",
      "Epoch:45 Batch:160 Log-loss0.031819432973861694\n",
      "Epoch:45 Batch:240 Log-loss0.03316847234964371\n",
      "Epoch:45 Batch:320 Log-loss0.033604491502046585\n",
      "Epoch:45 Batch:400 Log-loss0.037111155688762665\n",
      "Epoch:45 Batch:480 Log-loss0.039899297058582306\n",
      "Epoch:45 Batch:560 Log-loss0.03881652280688286\n",
      "Epoch average log loss:0.0370046596175858\n",
      "In Epoch45, val_loss:0.04068030447691109, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:46 Batch:0 Log-loss0.024858295917510986\n",
      "Epoch:46 Batch:80 Log-loss0.031488534063100815\n",
      "Epoch:46 Batch:160 Log-loss0.04306485876441002\n",
      "Epoch:46 Batch:240 Log-loss0.030499452725052834\n",
      "Epoch:46 Batch:320 Log-loss0.03495970740914345\n",
      "Epoch:46 Batch:400 Log-loss0.06321088969707489\n",
      "Epoch:46 Batch:480 Log-loss0.03625372052192688\n",
      "Epoch:46 Batch:560 Log-loss0.0313195139169693\n",
      "Epoch average log loss:0.03702024196474148\n",
      "In Epoch46, val_loss:0.040566488152779565, best_val_loss:0.040408434608041964, best_auc:0.9882946492066981\n",
      "Epoch:47 Batch:0 Log-loss0.029436543583869934\n",
      "Epoch:47 Batch:80 Log-loss0.03871844708919525\n",
      "Epoch:47 Batch:160 Log-loss0.037351738661527634\n",
      "Epoch:47 Batch:240 Log-loss0.029115552082657814\n",
      "Epoch:47 Batch:320 Log-loss0.040159355849027634\n",
      "Epoch:47 Batch:400 Log-loss0.04526493325829506\n",
      "Epoch:47 Batch:480 Log-loss0.03937894478440285\n",
      "Epoch:47 Batch:560 Log-loss0.04283814877271652\n",
      "Epoch average log loss:0.03704548286872783\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_643.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 4\n",
      "Epoch:1 Batch:0 Log-loss0.6804816722869873\n",
      "Epoch:1 Batch:80 Log-loss0.12520700693130493\n",
      "Epoch:1 Batch:160 Log-loss0.07724425941705704\n",
      "Epoch:1 Batch:240 Log-loss0.0602874755859375\n",
      "Epoch:1 Batch:320 Log-loss0.07075507193803787\n",
      "Epoch:1 Batch:400 Log-loss0.04799054563045502\n",
      "Epoch:1 Batch:480 Log-loss0.05049240589141846\n",
      "Epoch:1 Batch:560 Log-loss0.05577377602458\n",
      "Epoch average log loss:0.08863915434506323\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.050939955242743196, best_val_loss:0.050939955242743196, best_auc:0.9733093620627535\n",
      "Epoch:2 Batch:0 Log-loss0.048877522349357605\n",
      "Epoch:2 Batch:80 Log-loss0.0678967759013176\n",
      "Epoch:2 Batch:160 Log-loss0.0501580685377121\n",
      "Epoch:2 Batch:240 Log-loss0.0593075156211853\n",
      "Epoch:2 Batch:320 Log-loss0.043545227497816086\n",
      "Epoch:2 Batch:400 Log-loss0.03638254851102829\n",
      "Epoch:2 Batch:480 Log-loss0.05081813037395477\n",
      "Epoch:2 Batch:560 Log-loss0.06518390029668808\n",
      "Epoch average log loss:0.05352757970270302\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.04840631531271725, best_val_loss:0.04840631531271725, best_auc:0.9767821853144795\n",
      "Epoch:3 Batch:0 Log-loss0.05849049612879753\n",
      "Epoch:3 Batch:80 Log-loss0.0417415052652359\n",
      "Epoch:3 Batch:160 Log-loss0.04089043661952019\n",
      "Epoch:3 Batch:240 Log-loss0.04541537165641785\n",
      "Epoch:3 Batch:320 Log-loss0.03942639380693436\n",
      "Epoch:3 Batch:400 Log-loss0.03573617339134216\n",
      "Epoch:3 Batch:480 Log-loss0.05375060439109802\n",
      "Epoch:3 Batch:560 Log-loss0.05210522934794426\n",
      "Epoch average log loss:0.05048281163243311\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.046013878862067426, best_val_loss:0.046013878862067426, best_auc:0.980144752008952\n",
      "Epoch:4 Batch:0 Log-loss0.06012580171227455\n",
      "Epoch:4 Batch:80 Log-loss0.05093587934970856\n",
      "Epoch:4 Batch:160 Log-loss0.048089295625686646\n",
      "Epoch:4 Batch:240 Log-loss0.054519910365343094\n",
      "Epoch:4 Batch:320 Log-loss0.0675511434674263\n",
      "Epoch:4 Batch:400 Log-loss0.051031630486249924\n",
      "Epoch:4 Batch:480 Log-loss0.04195740446448326\n",
      "Epoch:4 Batch:560 Log-loss0.05556301400065422\n",
      "Epoch average log loss:0.04835019731295428\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.04318372438778651, best_val_loss:0.04318372438778651, best_auc:0.9840538097563146\n",
      "Epoch:5 Batch:0 Log-loss0.06142130121588707\n",
      "Epoch:5 Batch:80 Log-loss0.047673165798187256\n",
      "Epoch:5 Batch:160 Log-loss0.03532307595014572\n",
      "Epoch:5 Batch:240 Log-loss0.04970159754157066\n",
      "Epoch:5 Batch:320 Log-loss0.025680139660835266\n",
      "Epoch:5 Batch:400 Log-loss0.03275827690958977\n",
      "Epoch:5 Batch:480 Log-loss0.04688389226794243\n",
      "Epoch:5 Batch:560 Log-loss0.049993373453617096\n",
      "Epoch average log loss:0.04615569647867233\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.041911071074072916, best_val_loss:0.041911071074072916, best_auc:0.9859647712809477\n",
      "Epoch:6 Batch:0 Log-loss0.04953153431415558\n",
      "Epoch:6 Batch:80 Log-loss0.05621820688247681\n",
      "Epoch:6 Batch:160 Log-loss0.04281444475054741\n",
      "Epoch:6 Batch:240 Log-loss0.0522228442132473\n",
      "Epoch:6 Batch:320 Log-loss0.05291855335235596\n",
      "Epoch:6 Batch:400 Log-loss0.04422612860798836\n",
      "Epoch:6 Batch:480 Log-loss0.05916876718401909\n",
      "Epoch:6 Batch:560 Log-loss0.050060734152793884\n",
      "Epoch average log loss:0.04493880291203303\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.04129396658311966, best_val_loss:0.04129396658311966, best_auc:0.9865031896975472\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:7 Batch:0 Log-loss0.03617720678448677\n",
      "Epoch:7 Batch:80 Log-loss0.0496673546731472\n",
      "Epoch:7 Batch:160 Log-loss0.06015283986926079\n",
      "Epoch:7 Batch:240 Log-loss0.04327564314007759\n",
      "Epoch:7 Batch:320 Log-loss0.04168299213051796\n",
      "Epoch:7 Batch:400 Log-loss0.04615413025021553\n",
      "Epoch:7 Batch:480 Log-loss0.039308320730924606\n",
      "Epoch:7 Batch:560 Log-loss0.042627155780792236\n",
      "Epoch average log loss:0.04410304794686713\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.04101343677215864, best_val_loss:0.04101343677215864, best_auc:0.9874742174329275\n",
      "Epoch:8 Batch:0 Log-loss0.04478418454527855\n",
      "Epoch:8 Batch:80 Log-loss0.04880278930068016\n",
      "Epoch:8 Batch:160 Log-loss0.05242016911506653\n",
      "Epoch:8 Batch:240 Log-loss0.05851307138800621\n",
      "Epoch:8 Batch:320 Log-loss0.03206633776426315\n",
      "Epoch:8 Batch:400 Log-loss0.053046245127916336\n",
      "Epoch:8 Batch:480 Log-loss0.028051307424902916\n",
      "Epoch:8 Batch:560 Log-loss0.03844509646296501\n",
      "Epoch average log loss:0.04357143050458814\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.040259101236986276, best_val_loss:0.040259101236986276, best_auc:0.9883728861289819\n",
      "Epoch:9 Batch:0 Log-loss0.046163782477378845\n",
      "Epoch:9 Batch:80 Log-loss0.04122104123234749\n",
      "Epoch:9 Batch:160 Log-loss0.05187872424721718\n",
      "Epoch:9 Batch:240 Log-loss0.04382261633872986\n",
      "Epoch:9 Batch:320 Log-loss0.03514586761593819\n",
      "Epoch:9 Batch:400 Log-loss0.04061531648039818\n",
      "Epoch:9 Batch:480 Log-loss0.03455130383372307\n",
      "Epoch:9 Batch:560 Log-loss0.051676761358976364\n",
      "Epoch average log loss:0.042595734906249814\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.03991676611997461, best_val_loss:0.03991676611997461, best_auc:0.9890937463276387\n",
      "Epoch:10 Batch:0 Log-loss0.0405195951461792\n",
      "Epoch:10 Batch:80 Log-loss0.04227524623274803\n",
      "Epoch:10 Batch:160 Log-loss0.041295696049928665\n",
      "Epoch:10 Batch:240 Log-loss0.040716834366321564\n",
      "Epoch:10 Batch:320 Log-loss0.0338776670396328\n",
      "Epoch:10 Batch:400 Log-loss0.04109649360179901\n",
      "Epoch:10 Batch:480 Log-loss0.049349840730428696\n",
      "Epoch:10 Batch:560 Log-loss0.048094723373651505\n",
      "Epoch average log loss:0.04195386349622692\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.03935665656942517, best_val_loss:0.03935665656942517, best_auc:0.989674554462503\n",
      "Epoch:11 Batch:0 Log-loss0.052368540316820145\n",
      "Epoch:11 Batch:80 Log-loss0.038278136402368546\n",
      "Epoch:11 Batch:160 Log-loss0.03592811897397041\n",
      "Epoch:11 Batch:240 Log-loss0.0451868437230587\n",
      "Epoch:11 Batch:320 Log-loss0.04158655181527138\n",
      "Epoch:11 Batch:400 Log-loss0.034909937530756\n",
      "Epoch:11 Batch:480 Log-loss0.043372947722673416\n",
      "Epoch:11 Batch:560 Log-loss0.032337021082639694\n",
      "Epoch average log loss:0.04150753509851971\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch11, val_loss:0.03889287235070598, best_val_loss:0.03889287235070598, best_auc:0.9898877924419125\n",
      "Epoch:12 Batch:0 Log-loss0.04254286363720894\n",
      "Epoch:12 Batch:80 Log-loss0.05452745035290718\n",
      "Epoch:12 Batch:160 Log-loss0.02695683389902115\n",
      "Epoch:12 Batch:240 Log-loss0.04163089022040367\n",
      "Epoch:12 Batch:320 Log-loss0.04139954596757889\n",
      "Epoch:12 Batch:400 Log-loss0.03363613039255142\n",
      "Epoch:12 Batch:480 Log-loss0.045685771852731705\n",
      "Epoch:12 Batch:560 Log-loss0.04052118584513664\n",
      "Epoch average log loss:0.04113359393418899\n",
      "In Epoch12, val_loss:0.03928653400311515, best_val_loss:0.03889287235070598, best_auc:0.9898877924419125\n",
      "Epoch:13 Batch:0 Log-loss0.03149067610502243\n",
      "Epoch:13 Batch:80 Log-loss0.044514287263154984\n",
      "Epoch:13 Batch:160 Log-loss0.03728778287768364\n",
      "Epoch:13 Batch:240 Log-loss0.039423540234565735\n",
      "Epoch:13 Batch:320 Log-loss0.03091331757605076\n",
      "Epoch:13 Batch:400 Log-loss0.04421958699822426\n",
      "Epoch:13 Batch:480 Log-loss0.029588608071208\n",
      "Epoch:13 Batch:560 Log-loss0.038046013563871384\n",
      "Epoch average log loss:0.0408619323113401\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch13, val_loss:0.03868019609652335, best_val_loss:0.03868019609652335, best_auc:0.9900549615153141\n",
      "Epoch:14 Batch:0 Log-loss0.04485970735549927\n",
      "Epoch:14 Batch:80 Log-loss0.03472205623984337\n",
      "Epoch:14 Batch:160 Log-loss0.04468913748860359\n",
      "Epoch:14 Batch:240 Log-loss0.031300101429224014\n",
      "Epoch:14 Batch:320 Log-loss0.0313129685819149\n",
      "Epoch:14 Batch:400 Log-loss0.04197302833199501\n",
      "Epoch:14 Batch:480 Log-loss0.04556789994239807\n",
      "Epoch:14 Batch:560 Log-loss0.023926321417093277\n",
      "Epoch average log loss:0.040496179977032754\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch14, val_loss:0.03857553803513197, best_val_loss:0.03857553803513197, best_auc:0.9902156732119064\n",
      "Epoch:15 Batch:0 Log-loss0.03298618644475937\n",
      "Epoch:15 Batch:80 Log-loss0.03552216663956642\n",
      "Epoch:15 Batch:160 Log-loss0.031134655699133873\n",
      "Epoch:15 Batch:240 Log-loss0.06445112824440002\n",
      "Epoch:15 Batch:320 Log-loss0.044583115726709366\n",
      "Epoch:15 Batch:400 Log-loss0.04731331393122673\n",
      "Epoch:15 Batch:480 Log-loss0.043337199836969376\n",
      "Epoch:15 Batch:560 Log-loss0.06095530837774277\n",
      "Epoch average log loss:0.04023009896710781\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch15, val_loss:0.038318317684546896, best_val_loss:0.038318317684546896, best_auc:0.9902978299915514\n",
      "Epoch:16 Batch:0 Log-loss0.03309077396988869\n",
      "Epoch:16 Batch:80 Log-loss0.045312996953725815\n",
      "Epoch:16 Batch:160 Log-loss0.034615155309438705\n",
      "Epoch:16 Batch:240 Log-loss0.040102917701005936\n",
      "Epoch:16 Batch:320 Log-loss0.05296987295150757\n",
      "Epoch:16 Batch:400 Log-loss0.04495527222752571\n",
      "Epoch:16 Batch:480 Log-loss0.03717024251818657\n",
      "Epoch:16 Batch:560 Log-loss0.03743945434689522\n",
      "Epoch average log loss:0.039920041134714014\n",
      "In Epoch16, val_loss:0.038358173257702244, best_val_loss:0.038318317684546896, best_auc:0.9902978299915514\n",
      "Epoch:17 Batch:0 Log-loss0.03666628524661064\n",
      "Epoch:17 Batch:80 Log-loss0.04852709174156189\n",
      "Epoch:17 Batch:160 Log-loss0.031952790915966034\n",
      "Epoch:17 Batch:240 Log-loss0.03177785873413086\n",
      "Epoch:17 Batch:320 Log-loss0.03700394928455353\n",
      "Epoch:17 Batch:400 Log-loss0.04426557943224907\n",
      "Epoch:17 Batch:480 Log-loss0.04606197774410248\n",
      "Epoch:17 Batch:560 Log-loss0.02755112200975418\n",
      "Epoch average log loss:0.03975719268699842\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch17, val_loss:0.03826100820871036, best_val_loss:0.03826100820871036, best_auc:0.9902907032078891\n",
      "Epoch:18 Batch:0 Log-loss0.03873520344495773\n",
      "Epoch:18 Batch:80 Log-loss0.04026132449507713\n",
      "Epoch:18 Batch:160 Log-loss0.03264545649290085\n",
      "Epoch:18 Batch:240 Log-loss0.02613580785691738\n",
      "Epoch:18 Batch:320 Log-loss0.03637288138270378\n",
      "Epoch:18 Batch:400 Log-loss0.0651928186416626\n",
      "Epoch:18 Batch:480 Log-loss0.01794159784913063\n",
      "Epoch:18 Batch:560 Log-loss0.040412161499261856\n",
      "Epoch average log loss:0.03943243315443397\n",
      "In Epoch18, val_loss:0.03827000960202411, best_val_loss:0.03826100820871036, best_auc:0.9902907032078891\n",
      "Epoch:19 Batch:0 Log-loss0.02996443398296833\n",
      "Epoch:19 Batch:80 Log-loss0.05007984861731529\n",
      "Epoch:19 Batch:160 Log-loss0.030218936502933502\n",
      "Epoch:19 Batch:240 Log-loss0.046538904309272766\n",
      "Epoch:19 Batch:320 Log-loss0.030047940090298653\n",
      "Epoch:19 Batch:400 Log-loss0.0474226288497448\n",
      "Epoch:19 Batch:480 Log-loss0.05823622643947601\n",
      "Epoch:19 Batch:560 Log-loss0.03787747025489807\n",
      "Epoch average log loss:0.03939298279250839\n",
      "In Epoch19, val_loss:0.03829801211642649, best_val_loss:0.03826100820871036, best_auc:0.9902907032078891\n",
      "Epoch:20 Batch:0 Log-loss0.0442253015935421\n",
      "Epoch:20 Batch:80 Log-loss0.039696354418992996\n",
      "Epoch:20 Batch:160 Log-loss0.03986935317516327\n",
      "Epoch:20 Batch:240 Log-loss0.018896043300628662\n",
      "Epoch:20 Batch:320 Log-loss0.03862440958619118\n",
      "Epoch:20 Batch:400 Log-loss0.049208078533411026\n",
      "Epoch:20 Batch:480 Log-loss0.04911394044756889\n",
      "Epoch:20 Batch:560 Log-loss0.029782423749566078\n",
      "Epoch average log loss:0.039044686396872356\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch20, val_loss:0.0381959077538959, best_val_loss:0.0381959077538959, best_auc:0.9903443259157444\n",
      "Epoch:21 Batch:0 Log-loss0.035828012973070145\n",
      "Epoch:21 Batch:80 Log-loss0.034596484154462814\n",
      "Epoch:21 Batch:160 Log-loss0.03084980510175228\n",
      "Epoch:21 Batch:240 Log-loss0.04507467523217201\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:21 Batch:320 Log-loss0.040978554636240005\n",
      "Epoch:21 Batch:400 Log-loss0.03788148984313011\n",
      "Epoch:21 Batch:480 Log-loss0.032204631716012955\n",
      "Epoch:21 Batch:560 Log-loss0.03286150097846985\n",
      "Epoch average log loss:0.03903895652121199\n",
      "In Epoch21, val_loss:0.03843523730654874, best_val_loss:0.0381959077538959, best_auc:0.9903443259157444\n",
      "Epoch:22 Batch:0 Log-loss0.03506571426987648\n",
      "Epoch:22 Batch:80 Log-loss0.05018548667430878\n",
      "Epoch:22 Batch:160 Log-loss0.03942006826400757\n",
      "Epoch:22 Batch:240 Log-loss0.04464910551905632\n",
      "Epoch:22 Batch:320 Log-loss0.04036559537053108\n",
      "Epoch:22 Batch:400 Log-loss0.04386816918849945\n",
      "Epoch:22 Batch:480 Log-loss0.046606164425611496\n",
      "Epoch:22 Batch:560 Log-loss0.036583662033081055\n",
      "Epoch average log loss:0.038848533517947155\n",
      "In Epoch22, val_loss:0.0381985090815911, best_val_loss:0.0381959077538959, best_auc:0.9903443259157444\n",
      "Epoch:23 Batch:0 Log-loss0.03290756791830063\n",
      "Epoch:23 Batch:80 Log-loss0.044387731701135635\n",
      "Epoch:23 Batch:160 Log-loss0.03148932754993439\n",
      "Epoch:23 Batch:240 Log-loss0.03940049931406975\n",
      "Epoch:23 Batch:320 Log-loss0.034526437520980835\n",
      "Epoch:23 Batch:400 Log-loss0.03351211175322533\n",
      "Epoch:23 Batch:480 Log-loss0.037341710180044174\n",
      "Epoch:23 Batch:560 Log-loss0.03822723776102066\n",
      "Epoch average log loss:0.038594845572619565\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch23, val_loss:0.038171371938058786, best_val_loss:0.038171371938058786, best_auc:0.9904879162366497\n",
      "Epoch:24 Batch:0 Log-loss0.03062458522617817\n",
      "Epoch:24 Batch:80 Log-loss0.043996065855026245\n",
      "Epoch:24 Batch:160 Log-loss0.03597090020775795\n",
      "Epoch:24 Batch:240 Log-loss0.051819656044244766\n",
      "Epoch:24 Batch:320 Log-loss0.03196202591061592\n",
      "Epoch:24 Batch:400 Log-loss0.026772335171699524\n",
      "Epoch:24 Batch:480 Log-loss0.046422798186540604\n",
      "Epoch:24 Batch:560 Log-loss0.02494269795715809\n",
      "Epoch average log loss:0.03847076131163963\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch24, val_loss:0.03801985018343491, best_val_loss:0.03801985018343491, best_auc:0.9905124518781275\n",
      "Epoch:25 Batch:0 Log-loss0.036879729479551315\n",
      "Epoch:25 Batch:80 Log-loss0.0435166209936142\n",
      "Epoch:25 Batch:160 Log-loss0.03416714072227478\n",
      "Epoch:25 Batch:240 Log-loss0.050451263785362244\n",
      "Epoch:25 Batch:320 Log-loss0.04145849868655205\n",
      "Epoch:25 Batch:400 Log-loss0.02697102166712284\n",
      "Epoch:25 Batch:480 Log-loss0.047467079013586044\n",
      "Epoch:25 Batch:560 Log-loss0.03657590597867966\n",
      "Epoch average log loss:0.03833630877114566\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch25, val_loss:0.037909952906159915, best_val_loss:0.037909952906159915, best_auc:0.9904679560138203\n",
      "Epoch:26 Batch:0 Log-loss0.04721730574965477\n",
      "Epoch:26 Batch:80 Log-loss0.042399030178785324\n",
      "Epoch:26 Batch:160 Log-loss0.033943548798561096\n",
      "Epoch:26 Batch:240 Log-loss0.03984139487147331\n",
      "Epoch:26 Batch:320 Log-loss0.0324646532535553\n",
      "Epoch:26 Batch:400 Log-loss0.04648458957672119\n",
      "Epoch:26 Batch:480 Log-loss0.043980080634355545\n",
      "Epoch:26 Batch:560 Log-loss0.05428469553589821\n",
      "Epoch average log loss:0.038467504993812844\n",
      "In Epoch26, val_loss:0.03799655665112061, best_val_loss:0.037909952906159915, best_auc:0.9904679560138203\n",
      "Epoch:27 Batch:0 Log-loss0.06306075304746628\n",
      "Epoch:27 Batch:80 Log-loss0.03434831276535988\n",
      "Epoch:27 Batch:160 Log-loss0.04563520476222038\n",
      "Epoch:27 Batch:240 Log-loss0.04762279987335205\n",
      "Epoch:27 Batch:320 Log-loss0.030616505071520805\n",
      "Epoch:27 Batch:400 Log-loss0.0387919582426548\n",
      "Epoch:27 Batch:480 Log-loss0.03247683495283127\n",
      "Epoch:27 Batch:560 Log-loss0.03872736915946007\n",
      "Epoch average log loss:0.038347867796463625\n",
      "In Epoch27, val_loss:0.038104468542563504, best_val_loss:0.037909952906159915, best_auc:0.9904679560138203\n",
      "Epoch:28 Batch:0 Log-loss0.03484375774860382\n",
      "Epoch:28 Batch:80 Log-loss0.033091504126787186\n",
      "Epoch:28 Batch:160 Log-loss0.052746113389730453\n",
      "Epoch:28 Batch:240 Log-loss0.032101552933454514\n",
      "Epoch:28 Batch:320 Log-loss0.03795703127980232\n",
      "Epoch:28 Batch:400 Log-loss0.04727304354310036\n",
      "Epoch:28 Batch:480 Log-loss0.04792996868491173\n",
      "Epoch:28 Batch:560 Log-loss0.04189262539148331\n",
      "Epoch average log loss:0.038332603662274777\n",
      "In Epoch28, val_loss:0.03796206230918574, best_val_loss:0.037909952906159915, best_auc:0.9904679560138203\n",
      "Epoch:29 Batch:0 Log-loss0.053018778562545776\n",
      "Epoch:29 Batch:80 Log-loss0.024911681190133095\n",
      "Epoch:29 Batch:160 Log-loss0.03888998180627823\n",
      "Epoch:29 Batch:240 Log-loss0.0625610277056694\n",
      "Epoch:29 Batch:320 Log-loss0.03503510728478432\n",
      "Epoch:29 Batch:400 Log-loss0.04363305866718292\n",
      "Epoch:29 Batch:480 Log-loss0.039658546447753906\n",
      "Epoch:29 Batch:560 Log-loss0.039787083864212036\n",
      "Epoch average log loss:0.03804981478356889\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch29, val_loss:0.037880706715234046, best_val_loss:0.037880706715234046, best_auc:0.9905255750397087\n",
      "Epoch:30 Batch:0 Log-loss0.038409121334552765\n",
      "Epoch:30 Batch:80 Log-loss0.03876429796218872\n",
      "Epoch:30 Batch:160 Log-loss0.03289758786559105\n",
      "Epoch:30 Batch:240 Log-loss0.027890155091881752\n",
      "Epoch:30 Batch:320 Log-loss0.027185631915926933\n",
      "Epoch:30 Batch:400 Log-loss0.04115039482712746\n",
      "Epoch:30 Batch:480 Log-loss0.02960752137005329\n",
      "Epoch:30 Batch:560 Log-loss0.025724226608872414\n",
      "Epoch average log loss:0.0379217278345355\n",
      "In Epoch30, val_loss:0.038035960440604796, best_val_loss:0.037880706715234046, best_auc:0.9905255750397087\n",
      "Epoch:31 Batch:0 Log-loss0.05366453155875206\n",
      "Epoch:31 Batch:80 Log-loss0.03457460179924965\n",
      "Epoch:31 Batch:160 Log-loss0.030228517949581146\n",
      "Epoch:31 Batch:240 Log-loss0.0358646959066391\n",
      "Epoch:31 Batch:320 Log-loss0.030998753383755684\n",
      "Epoch:31 Batch:400 Log-loss0.031715016812086105\n",
      "Epoch:31 Batch:480 Log-loss0.04008614644408226\n",
      "Epoch:31 Batch:560 Log-loss0.03368573263287544\n",
      "Epoch average log loss:0.037838412049625604\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch31, val_loss:0.03786688153083575, best_val_loss:0.03786688153083575, best_auc:0.9905507090601725\n",
      "Epoch:32 Batch:0 Log-loss0.030966036021709442\n",
      "Epoch:32 Batch:80 Log-loss0.043383706361055374\n",
      "Epoch:32 Batch:160 Log-loss0.03964516893029213\n",
      "Epoch:32 Batch:240 Log-loss0.029874833300709724\n",
      "Epoch:32 Batch:320 Log-loss0.04015430808067322\n",
      "Epoch:32 Batch:400 Log-loss0.03521643951535225\n",
      "Epoch:32 Batch:480 Log-loss0.04489387571811676\n",
      "Epoch:32 Batch:560 Log-loss0.04900787025690079\n",
      "Epoch average log loss:0.03790086671776537\n",
      "In Epoch32, val_loss:0.037939993309977056, best_val_loss:0.03786688153083575, best_auc:0.9905507090601725\n",
      "Epoch:33 Batch:0 Log-loss0.042934197932481766\n",
      "Epoch:33 Batch:80 Log-loss0.04283101484179497\n",
      "Epoch:33 Batch:160 Log-loss0.03525952622294426\n",
      "Epoch:33 Batch:240 Log-loss0.03014683723449707\n",
      "Epoch:33 Batch:320 Log-loss0.05041716620326042\n",
      "Epoch:33 Batch:400 Log-loss0.03760099038481712\n",
      "Epoch:33 Batch:480 Log-loss0.050702545791864395\n",
      "Epoch:33 Batch:560 Log-loss0.04271431267261505\n",
      "Epoch average log loss:0.03783397042924272\n",
      "In Epoch33, val_loss:0.03788753578921026, best_val_loss:0.03786688153083575, best_auc:0.9905507090601725\n",
      "Epoch:34 Batch:0 Log-loss0.03622237965464592\n",
      "Epoch:34 Batch:80 Log-loss0.046267371624708176\n",
      "Epoch:34 Batch:160 Log-loss0.03978605940937996\n",
      "Epoch:34 Batch:240 Log-loss0.029145313426852226\n",
      "Epoch:34 Batch:320 Log-loss0.03056538850069046\n",
      "Epoch:34 Batch:400 Log-loss0.024382710456848145\n",
      "Epoch:34 Batch:480 Log-loss0.04393908381462097\n",
      "Epoch:34 Batch:560 Log-loss0.034378066658973694\n",
      "Epoch average log loss:0.03781391964626631\n",
      "In Epoch34, val_loss:0.03803093401577017, best_val_loss:0.03786688153083575, best_auc:0.9905507090601725\n",
      "Epoch:35 Batch:0 Log-loss0.042976487427949905\n",
      "Epoch:35 Batch:80 Log-loss0.03294948861002922\n",
      "Epoch:35 Batch:160 Log-loss0.06467997282743454\n",
      "Epoch:35 Batch:240 Log-loss0.048560112714767456\n",
      "Epoch:35 Batch:320 Log-loss0.033560801297426224\n",
      "Epoch:35 Batch:400 Log-loss0.031015358865261078\n",
      "Epoch:35 Batch:480 Log-loss0.03199170529842377\n",
      "Epoch:35 Batch:560 Log-loss0.052333734929561615\n",
      "Epoch average log loss:0.037660911403197266\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch35, val_loss:0.03777586024194641, best_val_loss:0.03777586024194641, best_auc:0.9906090771063606\n",
      "Epoch:36 Batch:0 Log-loss0.05448612943291664\n",
      "Epoch:36 Batch:80 Log-loss0.027693763375282288\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:36 Batch:160 Log-loss0.03890731930732727\n",
      "Epoch:36 Batch:240 Log-loss0.03552400320768356\n",
      "Epoch:36 Batch:320 Log-loss0.03394286334514618\n",
      "Epoch:36 Batch:400 Log-loss0.022011829540133476\n",
      "Epoch:36 Batch:480 Log-loss0.045127809047698975\n",
      "Epoch:36 Batch:560 Log-loss0.02589309588074684\n",
      "Epoch average log loss:0.03747035656955892\n",
      "In Epoch36, val_loss:0.03781317906858436, best_val_loss:0.03777586024194641, best_auc:0.9906090771063606\n",
      "Epoch:37 Batch:0 Log-loss0.04708505794405937\n",
      "Epoch:37 Batch:80 Log-loss0.030802428722381592\n",
      "Epoch:37 Batch:160 Log-loss0.0350959375500679\n",
      "Epoch:37 Batch:240 Log-loss0.044335246086120605\n",
      "Epoch:37 Batch:320 Log-loss0.03193534538149834\n",
      "Epoch:37 Batch:400 Log-loss0.020494287833571434\n",
      "Epoch:37 Batch:480 Log-loss0.041236624121665955\n",
      "Epoch:37 Batch:560 Log-loss0.03416894003748894\n",
      "Epoch average log loss:0.03766748538639929\n",
      "In Epoch37, val_loss:0.03782403675826664, best_val_loss:0.03777586024194641, best_auc:0.9906090771063606\n",
      "Epoch:38 Batch:0 Log-loss0.03709869459271431\n",
      "Epoch:38 Batch:80 Log-loss0.044100578874349594\n",
      "Epoch:38 Batch:160 Log-loss0.0420520044863224\n",
      "Epoch:38 Batch:240 Log-loss0.033758606761693954\n",
      "Epoch:38 Batch:320 Log-loss0.03775967285037041\n",
      "Epoch:38 Batch:400 Log-loss0.039751771837472916\n",
      "Epoch:38 Batch:480 Log-loss0.033424261957407\n",
      "Epoch:38 Batch:560 Log-loss0.03359859064221382\n",
      "Epoch average log loss:0.03763208876230887\n",
      "In Epoch38, val_loss:0.037845207939986, best_val_loss:0.03777586024194641, best_auc:0.9906090771063606\n",
      "Epoch:39 Batch:0 Log-loss0.043851882219314575\n",
      "Epoch:39 Batch:80 Log-loss0.026371533051133156\n",
      "Epoch:39 Batch:160 Log-loss0.023819847032427788\n",
      "Epoch:39 Batch:240 Log-loss0.02308291755616665\n",
      "Epoch:39 Batch:320 Log-loss0.023865079507231712\n",
      "Epoch:39 Batch:400 Log-loss0.03999816253781319\n",
      "Epoch:39 Batch:480 Log-loss0.03652970492839813\n",
      "Epoch:39 Batch:560 Log-loss0.0455889031291008\n",
      "Epoch average log loss:0.03745195627478617\n",
      "In Epoch39, val_loss:0.037838015225659105, best_val_loss:0.03777586024194641, best_auc:0.9906090771063606\n",
      "Epoch:40 Batch:0 Log-loss0.0427175872027874\n",
      "Epoch:40 Batch:80 Log-loss0.0362759530544281\n",
      "Epoch:40 Batch:160 Log-loss0.023329095914959908\n",
      "Epoch:40 Batch:240 Log-loss0.032268062233924866\n",
      "Epoch:40 Batch:320 Log-loss0.04038459062576294\n",
      "Epoch:40 Batch:400 Log-loss0.030541492626070976\n",
      "Epoch:40 Batch:480 Log-loss0.03322182968258858\n",
      "Epoch:40 Batch:560 Log-loss0.029326824471354485\n",
      "Epoch average log loss:0.03747291405618723\n",
      "In Epoch40, val_loss:0.037863961540324474, best_val_loss:0.03777586024194641, best_auc:0.9906090771063606\n",
      "Epoch:41 Batch:0 Log-loss0.03569381684064865\n",
      "Epoch:41 Batch:80 Log-loss0.025998495519161224\n",
      "Epoch:41 Batch:160 Log-loss0.03048824518918991\n",
      "Epoch:41 Batch:240 Log-loss0.041165634989738464\n",
      "Epoch:41 Batch:320 Log-loss0.03437440097332001\n",
      "Epoch:41 Batch:400 Log-loss0.05308939889073372\n",
      "Epoch:41 Batch:480 Log-loss0.04174849018454552\n",
      "Epoch:41 Batch:560 Log-loss0.040281590074300766\n",
      "Epoch average log loss:0.03762705421208271\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch41, val_loss:0.03776315267642132, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:42 Batch:0 Log-loss0.05011625215411186\n",
      "Epoch:42 Batch:80 Log-loss0.056112900376319885\n",
      "Epoch:42 Batch:160 Log-loss0.030483180657029152\n",
      "Epoch:42 Batch:240 Log-loss0.038691651076078415\n",
      "Epoch:42 Batch:320 Log-loss0.035662513226270676\n",
      "Epoch:42 Batch:400 Log-loss0.034463778138160706\n",
      "Epoch:42 Batch:480 Log-loss0.05597522482275963\n",
      "Epoch:42 Batch:560 Log-loss0.027199389412999153\n",
      "Epoch average log loss:0.03736393251788935\n",
      "In Epoch42, val_loss:0.03781477496941008, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:43 Batch:0 Log-loss0.046786848455667496\n",
      "Epoch:43 Batch:80 Log-loss0.037415046244859695\n",
      "Epoch:43 Batch:160 Log-loss0.04722675681114197\n",
      "Epoch:43 Batch:240 Log-loss0.02769605815410614\n",
      "Epoch:43 Batch:320 Log-loss0.03755958005785942\n",
      "Epoch:43 Batch:400 Log-loss0.04448467865586281\n",
      "Epoch:43 Batch:480 Log-loss0.051959145814180374\n",
      "Epoch:43 Batch:560 Log-loss0.03111584484577179\n",
      "Epoch average log loss:0.037376015540212394\n",
      "In Epoch43, val_loss:0.037932777535348454, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:44 Batch:0 Log-loss0.04662773385643959\n",
      "Epoch:44 Batch:80 Log-loss0.025297166779637337\n",
      "Epoch:44 Batch:160 Log-loss0.03658270463347435\n",
      "Epoch:44 Batch:240 Log-loss0.03487439081072807\n",
      "Epoch:44 Batch:320 Log-loss0.02867329865694046\n",
      "Epoch:44 Batch:400 Log-loss0.038529783487319946\n",
      "Epoch:44 Batch:480 Log-loss0.03956487029790878\n",
      "Epoch:44 Batch:560 Log-loss0.04329293966293335\n",
      "Epoch average log loss:0.03739602613994586\n",
      "In Epoch44, val_loss:0.03784020270149025, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:45 Batch:0 Log-loss0.03699148818850517\n",
      "Epoch:45 Batch:80 Log-loss0.03171432390809059\n",
      "Epoch:45 Batch:160 Log-loss0.04311637580394745\n",
      "Epoch:45 Batch:240 Log-loss0.03207968920469284\n",
      "Epoch:45 Batch:320 Log-loss0.03989214077591896\n",
      "Epoch:45 Batch:400 Log-loss0.04537639021873474\n",
      "Epoch:45 Batch:480 Log-loss0.030735643580555916\n",
      "Epoch:45 Batch:560 Log-loss0.030817382037639618\n",
      "Epoch average log loss:0.037281167806525316\n",
      "In Epoch45, val_loss:0.03782637691424532, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:46 Batch:0 Log-loss0.048982199281454086\n",
      "Epoch:46 Batch:80 Log-loss0.03261459991335869\n",
      "Epoch:46 Batch:160 Log-loss0.05022413656115532\n",
      "Epoch:46 Batch:240 Log-loss0.04281602427363396\n",
      "Epoch:46 Batch:320 Log-loss0.03956741839647293\n",
      "Epoch:46 Batch:400 Log-loss0.03801266849040985\n",
      "Epoch:46 Batch:480 Log-loss0.03426453843712807\n",
      "Epoch:46 Batch:560 Log-loss0.04123114049434662\n",
      "Epoch average log loss:0.037326430888580425\n",
      "In Epoch46, val_loss:0.03786537720953681, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:47 Batch:0 Log-loss0.040838293731212616\n",
      "Epoch:47 Batch:80 Log-loss0.03284693881869316\n",
      "Epoch:47 Batch:160 Log-loss0.04088335111737251\n",
      "Epoch:47 Batch:240 Log-loss0.04042770713567734\n",
      "Epoch:47 Batch:320 Log-loss0.03929952159523964\n",
      "Epoch:47 Batch:400 Log-loss0.02898515574634075\n",
      "Epoch:47 Batch:480 Log-loss0.03246176987886429\n",
      "Epoch:47 Batch:560 Log-loss0.03357522934675217\n",
      "Epoch average log loss:0.03729877458951835\n",
      "In Epoch47, val_loss:0.037797581575049546, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:48 Batch:0 Log-loss0.038456957787275314\n",
      "Epoch:48 Batch:80 Log-loss0.03982916846871376\n",
      "Epoch:48 Batch:160 Log-loss0.055900707840919495\n",
      "Epoch:48 Batch:240 Log-loss0.04470409080386162\n",
      "Epoch:48 Batch:320 Log-loss0.033918749541044235\n",
      "Epoch:48 Batch:400 Log-loss0.05797654390335083\n",
      "Epoch:48 Batch:480 Log-loss0.02472836524248123\n",
      "Epoch:48 Batch:560 Log-loss0.02720268815755844\n",
      "Epoch average log loss:0.03728048140848322\n",
      "In Epoch48, val_loss:0.03784620248004137, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:49 Batch:0 Log-loss0.04222196340560913\n",
      "Epoch:49 Batch:80 Log-loss0.04849056527018547\n",
      "Epoch:49 Batch:160 Log-loss0.03780694678425789\n",
      "Epoch:49 Batch:240 Log-loss0.032312579452991486\n",
      "Epoch:49 Batch:320 Log-loss0.04024505615234375\n",
      "Epoch:49 Batch:400 Log-loss0.04026434198021889\n",
      "Epoch:49 Batch:480 Log-loss0.031940579414367676\n",
      "Epoch:49 Batch:560 Log-loss0.025321071967482567\n",
      "Epoch average log loss:0.03720881703442761\n",
      "In Epoch49, val_loss:0.03780808318455126, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:50 Batch:0 Log-loss0.05664633587002754\n",
      "Epoch:50 Batch:80 Log-loss0.023274553939700127\n",
      "Epoch:50 Batch:160 Log-loss0.03509598225355148\n",
      "Epoch:50 Batch:240 Log-loss0.04947919026017189\n",
      "Epoch:50 Batch:320 Log-loss0.0305106770247221\n",
      "Epoch:50 Batch:400 Log-loss0.04478234052658081\n",
      "Epoch:50 Batch:480 Log-loss0.046095941215753555\n",
      "Epoch:50 Batch:560 Log-loss0.02921283058822155\n",
      "Epoch average log loss:0.03741024075674691\n",
      "In Epoch50, val_loss:0.037870902715172135, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:51 Batch:0 Log-loss0.032239530235528946\n",
      "Epoch:51 Batch:80 Log-loss0.030426764860749245\n",
      "Epoch:51 Batch:160 Log-loss0.062041010707616806\n",
      "Epoch:51 Batch:240 Log-loss0.02916758321225643\n",
      "Epoch:51 Batch:320 Log-loss0.042549118399620056\n",
      "Epoch:51 Batch:400 Log-loss0.03781760856509209\n",
      "Epoch:51 Batch:480 Log-loss0.03711893409490585\n",
      "Epoch:51 Batch:560 Log-loss0.031935956329107285\n",
      "Epoch average log loss:0.03713841288117692\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In Epoch51, val_loss:0.037810737664454815, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:52 Batch:0 Log-loss0.05458724871277809\n",
      "Epoch:52 Batch:80 Log-loss0.042988378554582596\n",
      "Epoch:52 Batch:160 Log-loss0.029215507209300995\n",
      "Epoch:52 Batch:240 Log-loss0.052889157086610794\n",
      "Epoch:52 Batch:320 Log-loss0.03583129867911339\n",
      "Epoch:52 Batch:400 Log-loss0.03762909024953842\n",
      "Epoch:52 Batch:480 Log-loss0.030087999999523163\n",
      "Epoch:52 Batch:560 Log-loss0.03810743987560272\n",
      "Epoch average log loss:0.03720860525832644\n",
      "In Epoch52, val_loss:0.037854331273436966, best_val_loss:0.03776315267642132, best_auc:0.9905577007216221\n",
      "Epoch:53 Batch:0 Log-loss0.03282177075743675\n",
      "Epoch:53 Batch:80 Log-loss0.04228833690285683\n",
      "Epoch:53 Batch:160 Log-loss0.029869534075260162\n",
      "Epoch:53 Batch:240 Log-loss0.024082409217953682\n",
      "Epoch:53 Batch:320 Log-loss0.04089953750371933\n",
      "Epoch:53 Batch:400 Log-loss0.03663834556937218\n",
      "Epoch:53 Batch:480 Log-loss0.046171512454748154\n",
      "Epoch:53 Batch:560 Log-loss0.030707640573382378\n",
      "Epoch average log loss:0.037278585245699755\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_644.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 5\n",
      "Epoch:1 Batch:0 Log-loss0.6909658908843994\n",
      "Epoch:1 Batch:80 Log-loss0.11347457766532898\n",
      "Epoch:1 Batch:160 Log-loss0.08909059315919876\n",
      "Epoch:1 Batch:240 Log-loss0.08275162428617477\n",
      "Epoch:1 Batch:320 Log-loss0.052547063678503036\n",
      "Epoch:1 Batch:400 Log-loss0.03957435116171837\n",
      "Epoch:1 Batch:480 Log-loss0.05191962048411369\n",
      "Epoch:1 Batch:560 Log-loss0.04153533652424812\n",
      "Epoch average log loss:0.08870059471311313\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.04925690404769358, best_val_loss:0.04925690404769358, best_auc:0.9733967108246452\n",
      "Epoch:2 Batch:0 Log-loss0.0450611375272274\n",
      "Epoch:2 Batch:80 Log-loss0.0424734503030777\n",
      "Epoch:2 Batch:160 Log-loss0.04849684238433838\n",
      "Epoch:2 Batch:240 Log-loss0.0605107806622982\n",
      "Epoch:2 Batch:320 Log-loss0.034460827708244324\n",
      "Epoch:2 Batch:400 Log-loss0.0528901107609272\n",
      "Epoch:2 Batch:480 Log-loss0.058170318603515625\n",
      "Epoch:2 Batch:560 Log-loss0.06411126255989075\n",
      "Epoch average log loss:0.05348346507110234\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.04736293333935654, best_val_loss:0.04736293333935654, best_auc:0.9774577924581411\n",
      "Epoch:3 Batch:0 Log-loss0.06422065943479538\n",
      "Epoch:3 Batch:80 Log-loss0.0610307939350605\n",
      "Epoch:3 Batch:160 Log-loss0.05701683834195137\n",
      "Epoch:3 Batch:240 Log-loss0.0439862497150898\n",
      "Epoch:3 Batch:320 Log-loss0.04640987142920494\n",
      "Epoch:3 Batch:400 Log-loss0.053734395653009415\n",
      "Epoch:3 Batch:480 Log-loss0.048548463732004166\n",
      "Epoch:3 Batch:560 Log-loss0.04797312244772911\n",
      "Epoch average log loss:0.05026727798394859\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.04321639532405844, best_val_loss:0.04321639532405844, best_auc:0.9820309093933952\n",
      "Epoch:4 Batch:0 Log-loss0.050414640456438065\n",
      "Epoch:4 Batch:80 Log-loss0.05118647590279579\n",
      "Epoch:4 Batch:160 Log-loss0.033361904323101044\n",
      "Epoch:4 Batch:240 Log-loss0.07974723726511002\n",
      "Epoch:4 Batch:320 Log-loss0.05795258283615112\n",
      "Epoch:4 Batch:400 Log-loss0.05627306178212166\n",
      "Epoch:4 Batch:480 Log-loss0.04152604565024376\n",
      "Epoch:4 Batch:560 Log-loss0.06607530266046524\n",
      "Epoch average log loss:0.04792324795520731\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.04188038825261239, best_val_loss:0.04188038825261239, best_auc:0.9846554506872018\n",
      "Epoch:5 Batch:0 Log-loss0.035973645746707916\n",
      "Epoch:5 Batch:80 Log-loss0.038066983222961426\n",
      "Epoch:5 Batch:160 Log-loss0.04670937359333038\n",
      "Epoch:5 Batch:240 Log-loss0.03559774160385132\n",
      "Epoch:5 Batch:320 Log-loss0.03893066570162773\n",
      "Epoch:5 Batch:400 Log-loss0.03862549737095833\n",
      "Epoch:5 Batch:480 Log-loss0.0408935546875\n",
      "Epoch:5 Batch:560 Log-loss0.04359813407063484\n",
      "Epoch average log loss:0.04603253030550799\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.04090374172767479, best_val_loss:0.04090374172767479, best_auc:0.9861331477588635\n",
      "Epoch:6 Batch:0 Log-loss0.04129396006464958\n",
      "Epoch:6 Batch:80 Log-loss0.04015837237238884\n",
      "Epoch:6 Batch:160 Log-loss0.039603326469659805\n",
      "Epoch:6 Batch:240 Log-loss0.04181329533457756\n",
      "Epoch:6 Batch:320 Log-loss0.04360322654247284\n",
      "Epoch:6 Batch:400 Log-loss0.05033678933978081\n",
      "Epoch:6 Batch:480 Log-loss0.036745086312294006\n",
      "Epoch:6 Batch:560 Log-loss0.03857935965061188\n",
      "Epoch average log loss:0.04483746245969087\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.03988162828655618, best_val_loss:0.03988162828655618, best_auc:0.9869664369765659\n",
      "Epoch:7 Batch:0 Log-loss0.05451210215687752\n",
      "Epoch:7 Batch:80 Log-loss0.06596138328313828\n",
      "Epoch:7 Batch:160 Log-loss0.04577340558171272\n",
      "Epoch:7 Batch:240 Log-loss0.0695250928401947\n",
      "Epoch:7 Batch:320 Log-loss0.05068141594529152\n",
      "Epoch:7 Batch:400 Log-loss0.05087536573410034\n",
      "Epoch:7 Batch:480 Log-loss0.059582483023405075\n",
      "Epoch:7 Batch:560 Log-loss0.04054327309131622\n",
      "Epoch average log loss:0.04409661928324827\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.03958255788852288, best_val_loss:0.03958255788852288, best_auc:0.9877335644456134\n",
      "Epoch:8 Batch:0 Log-loss0.05273272097110748\n",
      "Epoch:8 Batch:80 Log-loss0.037208400666713715\n",
      "Epoch:8 Batch:160 Log-loss0.03432847931981087\n",
      "Epoch:8 Batch:240 Log-loss0.06246873736381531\n",
      "Epoch:8 Batch:320 Log-loss0.0556645430624485\n",
      "Epoch:8 Batch:400 Log-loss0.03631353750824928\n",
      "Epoch:8 Batch:480 Log-loss0.0397903174161911\n",
      "Epoch:8 Batch:560 Log-loss0.04247766360640526\n",
      "Epoch average log loss:0.04317462416033128\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.03905422625054072, best_val_loss:0.03905422625054072, best_auc:0.9885178634898847\n",
      "Epoch:9 Batch:0 Log-loss0.0473618358373642\n",
      "Epoch:9 Batch:80 Log-loss0.03690820559859276\n",
      "Epoch:9 Batch:160 Log-loss0.03621915727853775\n",
      "Epoch:9 Batch:240 Log-loss0.03857153654098511\n",
      "Epoch:9 Batch:320 Log-loss0.04722920060157776\n",
      "Epoch:9 Batch:400 Log-loss0.03175037354230881\n",
      "Epoch:9 Batch:480 Log-loss0.044417303055524826\n",
      "Epoch:9 Batch:560 Log-loss0.03182258456945419\n",
      "Epoch average log loss:0.04272964182309806\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.03870666017774923, best_val_loss:0.03870666017774923, best_auc:0.9888087690807872\n",
      "Epoch:10 Batch:0 Log-loss0.04283294081687927\n",
      "Epoch:10 Batch:80 Log-loss0.030514603480696678\n",
      "Epoch:10 Batch:160 Log-loss0.02753233164548874\n",
      "Epoch:10 Batch:240 Log-loss0.04958449304103851\n",
      "Epoch:10 Batch:320 Log-loss0.03854243829846382\n",
      "Epoch:10 Batch:400 Log-loss0.030913323163986206\n",
      "Epoch:10 Batch:480 Log-loss0.054072435945272446\n",
      "Epoch:10 Batch:560 Log-loss0.03193598985671997\n",
      "Epoch average log loss:0.04222068471301879\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.038589592499736966, best_val_loss:0.038589592499736966, best_auc:0.9895416109498765\n",
      "Epoch:11 Batch:0 Log-loss0.034038301557302475\n",
      "Epoch:11 Batch:80 Log-loss0.0329098217189312\n",
      "Epoch:11 Batch:160 Log-loss0.043889909982681274\n",
      "Epoch:11 Batch:240 Log-loss0.04132922366261482\n",
      "Epoch:11 Batch:320 Log-loss0.05921731889247894\n",
      "Epoch:11 Batch:400 Log-loss0.04210389778017998\n",
      "Epoch:11 Batch:480 Log-loss0.03557712584733963\n",
      "Epoch:11 Batch:560 Log-loss0.049484364688396454\n",
      "Epoch average log loss:0.04155380026038204\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch11, val_loss:0.038310707143153266, best_val_loss:0.038310707143153266, best_auc:0.9897175513727311\n",
      "Epoch:12 Batch:0 Log-loss0.023788990452885628\n",
      "Epoch:12 Batch:80 Log-loss0.04596644639968872\n",
      "Epoch:12 Batch:160 Log-loss0.05262171849608421\n",
      "Epoch:12 Batch:240 Log-loss0.0431961864233017\n",
      "Epoch:12 Batch:320 Log-loss0.04466850683093071\n",
      "Epoch:12 Batch:400 Log-loss0.03133547678589821\n",
      "Epoch:12 Batch:480 Log-loss0.03283965587615967\n",
      "Epoch:12 Batch:560 Log-loss0.043012212961912155\n",
      "Epoch average log loss:0.04107464376970061\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.03825194682694852, best_val_loss:0.03825194682694852, best_auc:0.9900394789376152\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:13 Batch:0 Log-loss0.032253239303827286\n",
      "Epoch:13 Batch:80 Log-loss0.03070041351020336\n",
      "Epoch:13 Batch:160 Log-loss0.04723528027534485\n",
      "Epoch:13 Batch:240 Log-loss0.034450117498636246\n",
      "Epoch:13 Batch:320 Log-loss0.04955190420150757\n",
      "Epoch:13 Batch:400 Log-loss0.04607824608683586\n",
      "Epoch:13 Batch:480 Log-loss0.04090959578752518\n",
      "Epoch:13 Batch:560 Log-loss0.02864951826632023\n",
      "Epoch average log loss:0.040770111472478936\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch13, val_loss:0.038168949273072544, best_val_loss:0.038168949273072544, best_auc:0.9899251698559312\n",
      "Epoch:14 Batch:0 Log-loss0.025691932067275047\n",
      "Epoch:14 Batch:80 Log-loss0.058148279786109924\n",
      "Epoch:14 Batch:160 Log-loss0.04040117934346199\n",
      "Epoch:14 Batch:240 Log-loss0.02992095984518528\n",
      "Epoch:14 Batch:320 Log-loss0.051131755113601685\n",
      "Epoch:14 Batch:400 Log-loss0.02711285836994648\n",
      "Epoch:14 Batch:480 Log-loss0.05289420112967491\n",
      "Epoch:14 Batch:560 Log-loss0.03132263198494911\n",
      "Epoch average log loss:0.04056281691217529\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch14, val_loss:0.0380550426942321, best_val_loss:0.0380550426942321, best_auc:0.9902016745994081\n",
      "Epoch:15 Batch:0 Log-loss0.04495037719607353\n",
      "Epoch:15 Batch:80 Log-loss0.0466739647090435\n",
      "Epoch:15 Batch:160 Log-loss0.027485882863402367\n",
      "Epoch:15 Batch:240 Log-loss0.03642550855875015\n",
      "Epoch:15 Batch:320 Log-loss0.047489482909440994\n",
      "Epoch:15 Batch:400 Log-loss0.032471250742673874\n",
      "Epoch:15 Batch:480 Log-loss0.03190683200955391\n",
      "Epoch:15 Batch:560 Log-loss0.04222584515810013\n",
      "Epoch average log loss:0.04015051554223256\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch15, val_loss:0.037855288856695986, best_val_loss:0.037855288856695986, best_auc:0.9902989050362759\n",
      "Epoch:16 Batch:0 Log-loss0.030818775296211243\n",
      "Epoch:16 Batch:80 Log-loss0.04728102684020996\n",
      "Epoch:16 Batch:160 Log-loss0.029612554237246513\n",
      "Epoch:16 Batch:240 Log-loss0.03862122818827629\n",
      "Epoch:16 Batch:320 Log-loss0.026935875415802002\n",
      "Epoch:16 Batch:400 Log-loss0.043873775750398636\n",
      "Epoch:16 Batch:480 Log-loss0.04044410213828087\n",
      "Epoch:16 Batch:560 Log-loss0.030984915792942047\n",
      "Epoch average log loss:0.040121395194104736\n",
      "In Epoch16, val_loss:0.038269353946446955, best_val_loss:0.037855288856695986, best_auc:0.9902989050362759\n",
      "Epoch:17 Batch:0 Log-loss0.036678966134786606\n",
      "Epoch:17 Batch:80 Log-loss0.03138235583901405\n",
      "Epoch:17 Batch:160 Log-loss0.03907053917646408\n",
      "Epoch:17 Batch:240 Log-loss0.03630264475941658\n",
      "Epoch:17 Batch:320 Log-loss0.03287914767861366\n",
      "Epoch:17 Batch:400 Log-loss0.04774853587150574\n",
      "Epoch:17 Batch:480 Log-loss0.042642880231142044\n",
      "Epoch:17 Batch:560 Log-loss0.0553593672811985\n",
      "Epoch average log loss:0.03987726478371769\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch17, val_loss:0.037634769888922885, best_val_loss:0.037634769888922885, best_auc:0.9904378178792435\n",
      "Epoch:18 Batch:0 Log-loss0.03527136892080307\n",
      "Epoch:18 Batch:80 Log-loss0.04043952748179436\n",
      "Epoch:18 Batch:160 Log-loss0.04099329560995102\n",
      "Epoch:18 Batch:240 Log-loss0.0437965951859951\n",
      "Epoch:18 Batch:320 Log-loss0.03195341303944588\n",
      "Epoch:18 Batch:400 Log-loss0.06530653685331345\n",
      "Epoch:18 Batch:480 Log-loss0.04576944187283516\n",
      "Epoch:18 Batch:560 Log-loss0.04241370037198067\n",
      "Epoch average log loss:0.03955515819335623\n",
      "In Epoch18, val_loss:0.037855676892101835, best_val_loss:0.037634769888922885, best_auc:0.9904378178792435\n",
      "Epoch:19 Batch:0 Log-loss0.03776048496365547\n",
      "Epoch:19 Batch:80 Log-loss0.05916672945022583\n",
      "Epoch:19 Batch:160 Log-loss0.028265902772545815\n",
      "Epoch:19 Batch:240 Log-loss0.042405251413583755\n",
      "Epoch:19 Batch:320 Log-loss0.024000123143196106\n",
      "Epoch:19 Batch:400 Log-loss0.03715110570192337\n",
      "Epoch:19 Batch:480 Log-loss0.021145658567547798\n",
      "Epoch:19 Batch:560 Log-loss0.03467210754752159\n",
      "Epoch average log loss:0.03931894062219986\n",
      "In Epoch19, val_loss:0.03769902561986605, best_val_loss:0.037634769888922885, best_auc:0.9904378178792435\n",
      "Epoch:20 Batch:0 Log-loss0.04131356254220009\n",
      "Epoch:20 Batch:80 Log-loss0.03213699534535408\n",
      "Epoch:20 Batch:160 Log-loss0.03350958228111267\n",
      "Epoch:20 Batch:240 Log-loss0.037797629833221436\n",
      "Epoch:20 Batch:320 Log-loss0.03901689127087593\n",
      "Epoch:20 Batch:400 Log-loss0.04686594009399414\n",
      "Epoch:20 Batch:480 Log-loss0.03690755367279053\n",
      "Epoch:20 Batch:560 Log-loss0.034812770783901215\n",
      "Epoch average log loss:0.03919355351078723\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch20, val_loss:0.03751798299440615, best_val_loss:0.03751798299440615, best_auc:0.9905276732354028\n",
      "Epoch:21 Batch:0 Log-loss0.035656560212373734\n",
      "Epoch:21 Batch:80 Log-loss0.03831866756081581\n",
      "Epoch:21 Batch:160 Log-loss0.034490279853343964\n",
      "Epoch:21 Batch:240 Log-loss0.03349824994802475\n",
      "Epoch:21 Batch:320 Log-loss0.03087015636265278\n",
      "Epoch:21 Batch:400 Log-loss0.03758993372321129\n",
      "Epoch:21 Batch:480 Log-loss0.04598516225814819\n",
      "Epoch:21 Batch:560 Log-loss0.040142010897397995\n",
      "Epoch average log loss:0.03917163374966809\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch21, val_loss:0.03748210753526843, best_val_loss:0.03748210753526843, best_auc:0.9905468276767015\n",
      "Epoch:22 Batch:0 Log-loss0.048834145069122314\n",
      "Epoch:22 Batch:80 Log-loss0.043041110038757324\n",
      "Epoch:22 Batch:160 Log-loss0.02602563053369522\n",
      "Epoch:22 Batch:240 Log-loss0.0377877913415432\n",
      "Epoch:22 Batch:320 Log-loss0.046275123953819275\n",
      "Epoch:22 Batch:400 Log-loss0.04615719988942146\n",
      "Epoch:22 Batch:480 Log-loss0.04267388582229614\n",
      "Epoch:22 Batch:560 Log-loss0.031797561794519424\n",
      "Epoch average log loss:0.038798755490487176\n",
      "In Epoch22, val_loss:0.03782562689084536, best_val_loss:0.03748210753526843, best_auc:0.9905468276767015\n",
      "Epoch:23 Batch:0 Log-loss0.03967307135462761\n",
      "Epoch:23 Batch:80 Log-loss0.0463283471763134\n",
      "Epoch:23 Batch:160 Log-loss0.03770739212632179\n",
      "Epoch:23 Batch:240 Log-loss0.05310465022921562\n",
      "Epoch:23 Batch:320 Log-loss0.030376626178622246\n",
      "Epoch:23 Batch:400 Log-loss0.04477720335125923\n",
      "Epoch:23 Batch:480 Log-loss0.05454124137759209\n",
      "Epoch:23 Batch:560 Log-loss0.03997877612709999\n",
      "Epoch average log loss:0.038674642589675\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch23, val_loss:0.037370029475516375, best_val_loss:0.037370029475516375, best_auc:0.9905251435438517\n",
      "Epoch:24 Batch:0 Log-loss0.040733736008405685\n",
      "Epoch:24 Batch:80 Log-loss0.04478050768375397\n",
      "Epoch:24 Batch:160 Log-loss0.03811698034405708\n",
      "Epoch:24 Batch:240 Log-loss0.04424625262618065\n",
      "Epoch:24 Batch:320 Log-loss0.03618671000003815\n",
      "Epoch:24 Batch:400 Log-loss0.05059681460261345\n",
      "Epoch:24 Batch:480 Log-loss0.0331563800573349\n",
      "Epoch:24 Batch:560 Log-loss0.043105822056531906\n",
      "Epoch average log loss:0.03863974775054625\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch24, val_loss:0.03733111265470762, best_val_loss:0.03733111265470762, best_auc:0.9905904578762067\n",
      "Epoch:25 Batch:0 Log-loss0.03973986580967903\n",
      "Epoch:25 Batch:80 Log-loss0.04191656783223152\n",
      "Epoch:25 Batch:160 Log-loss0.025374894961714745\n",
      "Epoch:25 Batch:240 Log-loss0.041576992720365524\n",
      "Epoch:25 Batch:320 Log-loss0.040629077702760696\n",
      "Epoch:25 Batch:400 Log-loss0.0367354154586792\n",
      "Epoch:25 Batch:480 Log-loss0.031234972178936005\n",
      "Epoch:25 Batch:560 Log-loss0.044921230524778366\n",
      "Epoch average log loss:0.03867446824775211\n",
      "In Epoch25, val_loss:0.03749375824275288, best_val_loss:0.03733111265470762, best_auc:0.9905904578762067\n",
      "Epoch:26 Batch:0 Log-loss0.02566993050277233\n",
      "Epoch:26 Batch:80 Log-loss0.04035715013742447\n",
      "Epoch:26 Batch:160 Log-loss0.03613752871751785\n",
      "Epoch:26 Batch:240 Log-loss0.04032604396343231\n",
      "Epoch:26 Batch:320 Log-loss0.02945696748793125\n",
      "Epoch:26 Batch:400 Log-loss0.034893911331892014\n",
      "Epoch:26 Batch:480 Log-loss0.028350884094834328\n",
      "Epoch:26 Batch:560 Log-loss0.04575081169605255\n",
      "Epoch average log loss:0.03822734289736088\n",
      "In Epoch26, val_loss:0.03742433716246894, best_val_loss:0.03733111265470762, best_auc:0.9905904578762067\n",
      "Epoch:27 Batch:0 Log-loss0.0372152142226696\n",
      "Epoch:27 Batch:80 Log-loss0.036560866981744766\n",
      "Epoch:27 Batch:160 Log-loss0.045652732253074646\n",
      "Epoch:27 Batch:240 Log-loss0.03365267440676689\n",
      "Epoch:27 Batch:320 Log-loss0.029733531177043915\n",
      "Epoch:27 Batch:400 Log-loss0.030523652210831642\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:27 Batch:480 Log-loss0.02275513857603073\n",
      "Epoch:27 Batch:560 Log-loss0.04262002184987068\n",
      "Epoch average log loss:0.03845697439475251\n",
      "In Epoch27, val_loss:0.03756883108218751, best_val_loss:0.03733111265470762, best_auc:0.9905904578762067\n",
      "Epoch:28 Batch:0 Log-loss0.03924019634723663\n",
      "Epoch:28 Batch:80 Log-loss0.05011703446507454\n",
      "Epoch:28 Batch:160 Log-loss0.03766028210520744\n",
      "Epoch:28 Batch:240 Log-loss0.04296092689037323\n",
      "Epoch:28 Batch:320 Log-loss0.04188014194369316\n",
      "Epoch:28 Batch:400 Log-loss0.028997043147683144\n",
      "Epoch:28 Batch:480 Log-loss0.0510292649269104\n",
      "Epoch:28 Batch:560 Log-loss0.03416885435581207\n",
      "Epoch average log loss:0.03827250435881849\n",
      "In Epoch28, val_loss:0.03758677445342735, best_val_loss:0.03733111265470762, best_auc:0.9905904578762067\n",
      "Epoch:29 Batch:0 Log-loss0.04772840812802315\n",
      "Epoch:29 Batch:80 Log-loss0.02682955376803875\n",
      "Epoch:29 Batch:160 Log-loss0.02664805017411709\n",
      "Epoch:29 Batch:240 Log-loss0.03521977365016937\n",
      "Epoch:29 Batch:320 Log-loss0.040229927748441696\n",
      "Epoch:29 Batch:400 Log-loss0.031615108251571655\n",
      "Epoch:29 Batch:480 Log-loss0.036211129277944565\n",
      "Epoch:29 Batch:560 Log-loss0.056925661861896515\n",
      "Epoch average log loss:0.03805688055830875\n",
      "In Epoch29, val_loss:0.03745031913068316, best_val_loss:0.03733111265470762, best_auc:0.9905904578762067\n",
      "Epoch:30 Batch:0 Log-loss0.023144816979765892\n",
      "Epoch:30 Batch:80 Log-loss0.03126107156276703\n",
      "Epoch:30 Batch:160 Log-loss0.032744552940130234\n",
      "Epoch:30 Batch:240 Log-loss0.059574197977781296\n",
      "Epoch:30 Batch:320 Log-loss0.04057702049612999\n",
      "Epoch:30 Batch:400 Log-loss0.033161863684654236\n",
      "Epoch:30 Batch:480 Log-loss0.04261898994445801\n",
      "Epoch:30 Batch:560 Log-loss0.040335528552532196\n",
      "Epoch average log loss:0.03811043531833483\n",
      "In Epoch30, val_loss:0.0374318734360349, best_val_loss:0.03733111265470762, best_auc:0.9905904578762067\n",
      "Epoch:31 Batch:0 Log-loss0.02910946123301983\n",
      "Epoch:31 Batch:80 Log-loss0.03517846390604973\n",
      "Epoch:31 Batch:160 Log-loss0.03064596652984619\n",
      "Epoch:31 Batch:240 Log-loss0.03712271526455879\n",
      "Epoch:31 Batch:320 Log-loss0.036793697625398636\n",
      "Epoch:31 Batch:400 Log-loss0.04205562546849251\n",
      "Epoch:31 Batch:480 Log-loss0.03413005545735359\n",
      "Epoch:31 Batch:560 Log-loss0.03313083574175835\n",
      "Epoch average log loss:0.03792816386890731\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch31, val_loss:0.037293887688426806, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:32 Batch:0 Log-loss0.03700835630297661\n",
      "Epoch:32 Batch:80 Log-loss0.02528741955757141\n",
      "Epoch:32 Batch:160 Log-loss0.04471563920378685\n",
      "Epoch:32 Batch:240 Log-loss0.03218948096036911\n",
      "Epoch:32 Batch:320 Log-loss0.03468023240566254\n",
      "Epoch:32 Batch:400 Log-loss0.027987251058220863\n",
      "Epoch:32 Batch:480 Log-loss0.033292509615421295\n",
      "Epoch:32 Batch:560 Log-loss0.027879072353243828\n",
      "Epoch average log loss:0.037755442088070724\n",
      "In Epoch32, val_loss:0.03744974997811799, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:33 Batch:0 Log-loss0.0426906943321228\n",
      "Epoch:33 Batch:80 Log-loss0.0399944894015789\n",
      "Epoch:33 Batch:160 Log-loss0.027596093714237213\n",
      "Epoch:33 Batch:240 Log-loss0.03412473574280739\n",
      "Epoch:33 Batch:320 Log-loss0.030360838398337364\n",
      "Epoch:33 Batch:400 Log-loss0.04926173761487007\n",
      "Epoch:33 Batch:480 Log-loss0.03325691819190979\n",
      "Epoch:33 Batch:560 Log-loss0.027308806777000427\n",
      "Epoch average log loss:0.03786059344253902\n",
      "In Epoch33, val_loss:0.03743057205881845, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:34 Batch:0 Log-loss0.02260659635066986\n",
      "Epoch:34 Batch:80 Log-loss0.0473957397043705\n",
      "Epoch:34 Batch:160 Log-loss0.03495695814490318\n",
      "Epoch:34 Batch:240 Log-loss0.04380141198635101\n",
      "Epoch:34 Batch:320 Log-loss0.030956121161580086\n",
      "Epoch:34 Batch:400 Log-loss0.0690394937992096\n",
      "Epoch:34 Batch:480 Log-loss0.027993611991405487\n",
      "Epoch:34 Batch:560 Log-loss0.04658481851220131\n",
      "Epoch average log loss:0.03776203324752195\n",
      "In Epoch34, val_loss:0.03756691390760953, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:35 Batch:0 Log-loss0.026975048705935478\n",
      "Epoch:35 Batch:80 Log-loss0.035917144268751144\n",
      "Epoch:35 Batch:160 Log-loss0.03302455320954323\n",
      "Epoch:35 Batch:240 Log-loss0.049814656376838684\n",
      "Epoch:35 Batch:320 Log-loss0.04377882182598114\n",
      "Epoch:35 Batch:400 Log-loss0.0626586526632309\n",
      "Epoch:35 Batch:480 Log-loss0.04590765759348869\n",
      "Epoch:35 Batch:560 Log-loss0.037905581295490265\n",
      "Epoch average log loss:0.03766642826210175\n",
      "In Epoch35, val_loss:0.03748737245265533, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:36 Batch:0 Log-loss0.03693946823477745\n",
      "Epoch:36 Batch:80 Log-loss0.04346756264567375\n",
      "Epoch:36 Batch:160 Log-loss0.02722167782485485\n",
      "Epoch:36 Batch:240 Log-loss0.05585319176316261\n",
      "Epoch:36 Batch:320 Log-loss0.034080903977155685\n",
      "Epoch:36 Batch:400 Log-loss0.0421769879758358\n",
      "Epoch:36 Batch:480 Log-loss0.04118557646870613\n",
      "Epoch:36 Batch:560 Log-loss0.05253780260682106\n",
      "Epoch average log loss:0.03767565119446122\n",
      "In Epoch36, val_loss:0.037464351943348656, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:37 Batch:0 Log-loss0.03595840558409691\n",
      "Epoch:37 Batch:80 Log-loss0.03799569234251976\n",
      "Epoch:37 Batch:160 Log-loss0.03666822984814644\n",
      "Epoch:37 Batch:240 Log-loss0.02296673320233822\n",
      "Epoch:37 Batch:320 Log-loss0.0569138266146183\n",
      "Epoch:37 Batch:400 Log-loss0.0375787615776062\n",
      "Epoch:37 Batch:480 Log-loss0.03674471005797386\n",
      "Epoch:37 Batch:560 Log-loss0.02523241750895977\n",
      "Epoch average log loss:0.03766195663650121\n",
      "In Epoch37, val_loss:0.037528181013003904, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:38 Batch:0 Log-loss0.03136827424168587\n",
      "Epoch:38 Batch:80 Log-loss0.0279643926769495\n",
      "Epoch:38 Batch:160 Log-loss0.03501736372709274\n",
      "Epoch:38 Batch:240 Log-loss0.0467926561832428\n",
      "Epoch:38 Batch:320 Log-loss0.021045831963419914\n",
      "Epoch:38 Batch:400 Log-loss0.025307709351181984\n",
      "Epoch:38 Batch:480 Log-loss0.03055441938340664\n",
      "Epoch:38 Batch:560 Log-loss0.044335056096315384\n",
      "Epoch average log loss:0.03746495833620429\n",
      "In Epoch38, val_loss:0.03741299975103287, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:39 Batch:0 Log-loss0.0475127138197422\n",
      "Epoch:39 Batch:80 Log-loss0.04014608636498451\n",
      "Epoch:39 Batch:160 Log-loss0.05204283073544502\n",
      "Epoch:39 Batch:240 Log-loss0.03747983276844025\n",
      "Epoch:39 Batch:320 Log-loss0.04272863641381264\n",
      "Epoch:39 Batch:400 Log-loss0.02678062580525875\n",
      "Epoch:39 Batch:480 Log-loss0.037913233041763306\n",
      "Epoch:39 Batch:560 Log-loss0.03392614796757698\n",
      "Epoch average log loss:0.037565841851755974\n",
      "In Epoch39, val_loss:0.037339118966406924, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:40 Batch:0 Log-loss0.02678043395280838\n",
      "Epoch:40 Batch:80 Log-loss0.034324053674936295\n",
      "Epoch:40 Batch:160 Log-loss0.03422143682837486\n",
      "Epoch:40 Batch:240 Log-loss0.04528874158859253\n",
      "Epoch:40 Batch:320 Log-loss0.044542908668518066\n",
      "Epoch:40 Batch:400 Log-loss0.030165115371346474\n",
      "Epoch:40 Batch:480 Log-loss0.04794488847255707\n",
      "Epoch:40 Batch:560 Log-loss0.02917296811938286\n",
      "Epoch average log loss:0.03726185244241995\n",
      "In Epoch40, val_loss:0.03734282788046936, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:41 Batch:0 Log-loss0.041969019919633865\n",
      "Epoch:41 Batch:80 Log-loss0.04712815582752228\n",
      "Epoch:41 Batch:160 Log-loss0.04229976236820221\n",
      "Epoch:41 Batch:240 Log-loss0.033112749457359314\n",
      "Epoch:41 Batch:320 Log-loss0.0332171805202961\n",
      "Epoch:41 Batch:400 Log-loss0.0431615449488163\n",
      "Epoch:41 Batch:480 Log-loss0.0417272113263607\n",
      "Epoch:41 Batch:560 Log-loss0.03448176011443138\n",
      "Epoch average log loss:0.037606031893353375\n",
      "In Epoch41, val_loss:0.03739688379198064, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:42 Batch:0 Log-loss0.042644720524549484\n",
      "Epoch:42 Batch:80 Log-loss0.0345776341855526\n",
      "Epoch:42 Batch:160 Log-loss0.036414820700883865\n",
      "Epoch:42 Batch:240 Log-loss0.04640798270702362\n",
      "Epoch:42 Batch:320 Log-loss0.021598339080810547\n",
      "Epoch:42 Batch:400 Log-loss0.0392276905477047\n",
      "Epoch:42 Batch:480 Log-loss0.03665779531002045\n",
      "Epoch:42 Batch:560 Log-loss0.02751925401389599\n",
      "Epoch average log loss:0.03746427955006116\n",
      "In Epoch42, val_loss:0.037348929061125274, best_val_loss:0.037293887688426806, best_auc:0.9906240011777437\n",
      "Epoch:43 Batch:0 Log-loss0.04388439282774925\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:43 Batch:80 Log-loss0.037961434572935104\n",
      "Epoch:43 Batch:160 Log-loss0.03483762592077255\n",
      "Epoch:43 Batch:240 Log-loss0.03351212292909622\n",
      "Epoch:43 Batch:320 Log-loss0.04093945398926735\n",
      "Epoch:43 Batch:400 Log-loss0.049768637865781784\n",
      "Epoch:43 Batch:480 Log-loss0.040084417909383774\n",
      "Epoch:43 Batch:560 Log-loss0.0402580127120018\n",
      "Epoch average log loss:0.03759565534003611\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_645.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 6\n",
      "Epoch:1 Batch:0 Log-loss0.6927940845489502\n",
      "Epoch:1 Batch:80 Log-loss0.0726037472486496\n",
      "Epoch:1 Batch:160 Log-loss0.04304327443242073\n",
      "Epoch:1 Batch:240 Log-loss0.05778472125530243\n",
      "Epoch:1 Batch:320 Log-loss0.07424319535493851\n",
      "Epoch:1 Batch:400 Log-loss0.06802014261484146\n",
      "Epoch:1 Batch:480 Log-loss0.03602615371346474\n",
      "Epoch:1 Batch:560 Log-loss0.04634514078497887\n",
      "Epoch average log loss:0.08527183783839323\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.04901362420499005, best_val_loss:0.04901362420499005, best_auc:0.9734435753411194\n",
      "Epoch:2 Batch:0 Log-loss0.0701153501868248\n",
      "Epoch:2 Batch:80 Log-loss0.07227625697851181\n",
      "Epoch:2 Batch:160 Log-loss0.07553403079509735\n",
      "Epoch:2 Batch:240 Log-loss0.056049007922410965\n",
      "Epoch:2 Batch:320 Log-loss0.07332294434309006\n",
      "Epoch:2 Batch:400 Log-loss0.06804399192333221\n",
      "Epoch:2 Batch:480 Log-loss0.06263864785432816\n",
      "Epoch:2 Batch:560 Log-loss0.06893185526132584\n",
      "Epoch average log loss:0.05234649583830365\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.046208954627644495, best_val_loss:0.046208954627644495, best_auc:0.9781010027525593\n",
      "Epoch:3 Batch:0 Log-loss0.0695696622133255\n",
      "Epoch:3 Batch:80 Log-loss0.047358136624097824\n",
      "Epoch:3 Batch:160 Log-loss0.04662080109119415\n",
      "Epoch:3 Batch:240 Log-loss0.06323360651731491\n",
      "Epoch:3 Batch:320 Log-loss0.04116709157824516\n",
      "Epoch:3 Batch:400 Log-loss0.054832980036735535\n",
      "Epoch:3 Batch:480 Log-loss0.06582794338464737\n",
      "Epoch:3 Batch:560 Log-loss0.057542864233255386\n",
      "Epoch average log loss:0.04912736267849271\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.04401420342872856, best_val_loss:0.04401420342872856, best_auc:0.9817304582911951\n",
      "Epoch:4 Batch:0 Log-loss0.05932672321796417\n",
      "Epoch:4 Batch:80 Log-loss0.031532030552625656\n",
      "Epoch:4 Batch:160 Log-loss0.05277998372912407\n",
      "Epoch:4 Batch:240 Log-loss0.05040207877755165\n",
      "Epoch:4 Batch:320 Log-loss0.04971468448638916\n",
      "Epoch:4 Batch:400 Log-loss0.0623546838760376\n",
      "Epoch:4 Batch:480 Log-loss0.05779096111655235\n",
      "Epoch:4 Batch:560 Log-loss0.04586689546704292\n",
      "Epoch average log loss:0.04663595514970698\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.04164545905730364, best_val_loss:0.04164545905730364, best_auc:0.9835988214592928\n",
      "Epoch:5 Batch:0 Log-loss0.03694050759077072\n",
      "Epoch:5 Batch:80 Log-loss0.03394032642245293\n",
      "Epoch:5 Batch:160 Log-loss0.03950181603431702\n",
      "Epoch:5 Batch:240 Log-loss0.054559171199798584\n",
      "Epoch:5 Batch:320 Log-loss0.031533483415842056\n",
      "Epoch:5 Batch:400 Log-loss0.04085499048233032\n",
      "Epoch:5 Batch:480 Log-loss0.04053769260644913\n",
      "Epoch:5 Batch:560 Log-loss0.04040436074137688\n",
      "Epoch average log loss:0.04518935944485877\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.041392943118758234, best_val_loss:0.041392943118758234, best_auc:0.9852873563262915\n",
      "Epoch:6 Batch:0 Log-loss0.04864456132054329\n",
      "Epoch:6 Batch:80 Log-loss0.03267962113022804\n",
      "Epoch:6 Batch:160 Log-loss0.07102247327566147\n",
      "Epoch:6 Batch:240 Log-loss0.06170963868498802\n",
      "Epoch:6 Batch:320 Log-loss0.03161019831895828\n",
      "Epoch:6 Batch:400 Log-loss0.05063445493578911\n",
      "Epoch:6 Batch:480 Log-loss0.05789436027407646\n",
      "Epoch:6 Batch:560 Log-loss0.061398617923259735\n",
      "Epoch average log loss:0.04400053130396243\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.0403262767492956, best_val_loss:0.0403262767492956, best_auc:0.9864379171916321\n",
      "Epoch:7 Batch:0 Log-loss0.04474322870373726\n",
      "Epoch:7 Batch:80 Log-loss0.047746140509843826\n",
      "Epoch:7 Batch:160 Log-loss0.04900917783379555\n",
      "Epoch:7 Batch:240 Log-loss0.05054615065455437\n",
      "Epoch:7 Batch:320 Log-loss0.05048440396785736\n",
      "Epoch:7 Batch:400 Log-loss0.041134435683488846\n",
      "Epoch:7 Batch:480 Log-loss0.034124135971069336\n",
      "Epoch:7 Batch:560 Log-loss0.04083462059497833\n",
      "Epoch average log loss:0.04327228317103748\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.03946533911842369, best_val_loss:0.03946533911842369, best_auc:0.9875211353711474\n",
      "Epoch:8 Batch:0 Log-loss0.060965437442064285\n",
      "Epoch:8 Batch:80 Log-loss0.03676832839846611\n",
      "Epoch:8 Batch:160 Log-loss0.03692233934998512\n",
      "Epoch:8 Batch:240 Log-loss0.04402501508593559\n",
      "Epoch:8 Batch:320 Log-loss0.03831769898533821\n",
      "Epoch:8 Batch:400 Log-loss0.05258503928780556\n",
      "Epoch:8 Batch:480 Log-loss0.04390110448002815\n",
      "Epoch:8 Batch:560 Log-loss0.03610297664999962\n",
      "Epoch average log loss:0.04269842121284455\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.0392395733560272, best_val_loss:0.0392395733560272, best_auc:0.987807067404676\n",
      "Epoch:9 Batch:0 Log-loss0.034073151648044586\n",
      "Epoch:9 Batch:80 Log-loss0.05841102823615074\n",
      "Epoch:9 Batch:160 Log-loss0.04065447673201561\n",
      "Epoch:9 Batch:240 Log-loss0.027599507942795753\n",
      "Epoch:9 Batch:320 Log-loss0.04395847022533417\n",
      "Epoch:9 Batch:400 Log-loss0.06388216465711594\n",
      "Epoch:9 Batch:480 Log-loss0.036362696439027786\n",
      "Epoch:9 Batch:560 Log-loss0.035453278571367264\n",
      "Epoch average log loss:0.04198591373195606\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.03882257882399904, best_val_loss:0.03882257882399904, best_auc:0.9885910778129564\n",
      "Epoch:10 Batch:0 Log-loss0.03798128291964531\n",
      "Epoch:10 Batch:80 Log-loss0.037238653749227524\n",
      "Epoch:10 Batch:160 Log-loss0.03860572725534439\n",
      "Epoch:10 Batch:240 Log-loss0.04075969010591507\n",
      "Epoch:10 Batch:320 Log-loss0.02922198735177517\n",
      "Epoch:10 Batch:400 Log-loss0.03860026225447655\n",
      "Epoch:10 Batch:480 Log-loss0.04326706752181053\n",
      "Epoch:10 Batch:560 Log-loss0.050878580659627914\n",
      "Epoch average log loss:0.041656678805260784\n",
      "In Epoch10, val_loss:0.03897048830625751, best_val_loss:0.03882257882399904, best_auc:0.9885910778129564\n",
      "Epoch:11 Batch:0 Log-loss0.02673037350177765\n",
      "Epoch:11 Batch:80 Log-loss0.038186561316251755\n",
      "Epoch:11 Batch:160 Log-loss0.04694211855530739\n",
      "Epoch:11 Batch:240 Log-loss0.031345028430223465\n",
      "Epoch:11 Batch:320 Log-loss0.04822451248764992\n",
      "Epoch:11 Batch:400 Log-loss0.041774723678827286\n",
      "Epoch:11 Batch:480 Log-loss0.03978433087468147\n",
      "Epoch:11 Batch:560 Log-loss0.038955096155405045\n",
      "Epoch average log loss:0.04125804497993418\n",
      "In Epoch11, val_loss:0.038828606760116985, best_val_loss:0.03882257882399904, best_auc:0.9885910778129564\n",
      "Epoch:12 Batch:0 Log-loss0.056804221123456955\n",
      "Epoch:12 Batch:80 Log-loss0.01996740885078907\n",
      "Epoch:12 Batch:160 Log-loss0.029111025854945183\n",
      "Epoch:12 Batch:240 Log-loss0.04654671251773834\n",
      "Epoch:12 Batch:320 Log-loss0.027788281440734863\n",
      "Epoch:12 Batch:400 Log-loss0.052961647510528564\n",
      "Epoch:12 Batch:480 Log-loss0.043199192732572556\n",
      "Epoch:12 Batch:560 Log-loss0.037111081182956696\n",
      "Epoch average log loss:0.040951158754926706\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.038713646150409654, best_val_loss:0.038713646150409654, best_auc:0.9890941867539368\n",
      "Epoch:13 Batch:0 Log-loss0.03219112381339073\n",
      "Epoch:13 Batch:80 Log-loss0.03332368656992912\n",
      "Epoch:13 Batch:160 Log-loss0.04639558494091034\n",
      "Epoch:13 Batch:240 Log-loss0.05447528883814812\n",
      "Epoch:13 Batch:320 Log-loss0.03791206702589989\n",
      "Epoch:13 Batch:400 Log-loss0.04399794712662697\n",
      "Epoch:13 Batch:480 Log-loss0.04363594949245453\n",
      "Epoch:13 Batch:560 Log-loss0.02790125645697117\n",
      "Epoch average log loss:0.04055511446536652\n",
      "In Epoch13, val_loss:0.03902837952863739, best_val_loss:0.038713646150409654, best_auc:0.9890941867539368\n",
      "Epoch:14 Batch:0 Log-loss0.05900833383202553\n",
      "Epoch:14 Batch:80 Log-loss0.040759820491075516\n",
      "Epoch:14 Batch:160 Log-loss0.04885557293891907\n",
      "Epoch:14 Batch:240 Log-loss0.03694501891732216\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:14 Batch:320 Log-loss0.031884316354990005\n",
      "Epoch:14 Batch:400 Log-loss0.029345927760004997\n",
      "Epoch:14 Batch:480 Log-loss0.06262467056512833\n",
      "Epoch:14 Batch:560 Log-loss0.03380608186125755\n",
      "Epoch average log loss:0.04019845632636654\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch14, val_loss:0.03851778102079172, best_val_loss:0.03851778102079172, best_auc:0.9893188794910439\n",
      "Epoch:15 Batch:0 Log-loss0.03941350802779198\n",
      "Epoch:15 Batch:80 Log-loss0.033235590904951096\n",
      "Epoch:15 Batch:160 Log-loss0.047063782811164856\n",
      "Epoch:15 Batch:240 Log-loss0.045274555683135986\n",
      "Epoch:15 Batch:320 Log-loss0.03511225804686546\n",
      "Epoch:15 Batch:400 Log-loss0.03512350097298622\n",
      "Epoch:15 Batch:480 Log-loss0.0358218215405941\n",
      "Epoch:15 Batch:560 Log-loss0.042068224400281906\n",
      "Epoch average log loss:0.03985174111356693\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch15, val_loss:0.038195228366395835, best_val_loss:0.038195228366395835, best_auc:0.9894870909295458\n",
      "Epoch:16 Batch:0 Log-loss0.03569349646568298\n",
      "Epoch:16 Batch:80 Log-loss0.034346066415309906\n",
      "Epoch:16 Batch:160 Log-loss0.03591971471905708\n",
      "Epoch:16 Batch:240 Log-loss0.039347659796476364\n",
      "Epoch:16 Batch:320 Log-loss0.03165755420923233\n",
      "Epoch:16 Batch:400 Log-loss0.04252556338906288\n",
      "Epoch:16 Batch:480 Log-loss0.059642765671014786\n",
      "Epoch:16 Batch:560 Log-loss0.048478830605745316\n",
      "Epoch average log loss:0.03965378364747656\n",
      "In Epoch16, val_loss:0.03844352516224608, best_val_loss:0.038195228366395835, best_auc:0.9894870909295458\n",
      "Epoch:17 Batch:0 Log-loss0.032904788851737976\n",
      "Epoch:17 Batch:80 Log-loss0.050262123346328735\n",
      "Epoch:17 Batch:160 Log-loss0.04731221869587898\n",
      "Epoch:17 Batch:240 Log-loss0.03375399485230446\n",
      "Epoch:17 Batch:320 Log-loss0.025728074833750725\n",
      "Epoch:17 Batch:400 Log-loss0.03367776796221733\n",
      "Epoch:17 Batch:480 Log-loss0.05609523877501488\n",
      "Epoch:17 Batch:560 Log-loss0.03272949904203415\n",
      "Epoch average log loss:0.039508836583367415\n",
      "In Epoch17, val_loss:0.038324663336319564, best_val_loss:0.038195228366395835, best_auc:0.9894870909295458\n",
      "Epoch:18 Batch:0 Log-loss0.0230372566729784\n",
      "Epoch:18 Batch:80 Log-loss0.03890350088477135\n",
      "Epoch:18 Batch:160 Log-loss0.03582458570599556\n",
      "Epoch:18 Batch:240 Log-loss0.042908038944005966\n",
      "Epoch:18 Batch:320 Log-loss0.03178945556282997\n",
      "Epoch:18 Batch:400 Log-loss0.03198922052979469\n",
      "Epoch:18 Batch:480 Log-loss0.03735664486885071\n",
      "Epoch:18 Batch:560 Log-loss0.05733451992273331\n",
      "Epoch average log loss:0.0393096820950242\n",
      "In Epoch18, val_loss:0.03823328726006156, best_val_loss:0.038195228366395835, best_auc:0.9894870909295458\n",
      "Epoch:19 Batch:0 Log-loss0.0501135028898716\n",
      "Epoch:19 Batch:80 Log-loss0.04236838221549988\n",
      "Epoch:19 Batch:160 Log-loss0.03190058097243309\n",
      "Epoch:19 Batch:240 Log-loss0.03735152259469032\n",
      "Epoch:19 Batch:320 Log-loss0.0317196398973465\n",
      "Epoch:19 Batch:400 Log-loss0.039272092282772064\n",
      "Epoch:19 Batch:480 Log-loss0.036083463579416275\n",
      "Epoch:19 Batch:560 Log-loss0.02941063418984413\n",
      "Epoch average log loss:0.039025559801874415\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch19, val_loss:0.0380795167604277, best_val_loss:0.0380795167604277, best_auc:0.989726859394105\n",
      "Epoch:20 Batch:0 Log-loss0.0317743681371212\n",
      "Epoch:20 Batch:80 Log-loss0.03232138603925705\n",
      "Epoch:20 Batch:160 Log-loss0.04277859255671501\n",
      "Epoch:20 Batch:240 Log-loss0.02829214371740818\n",
      "Epoch:20 Batch:320 Log-loss0.05110450088977814\n",
      "Epoch:20 Batch:400 Log-loss0.05417951941490173\n",
      "Epoch:20 Batch:480 Log-loss0.02678452618420124\n",
      "Epoch:20 Batch:560 Log-loss0.027576869353652\n",
      "Epoch average log loss:0.03889837723691016\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch20, val_loss:0.03803237991688545, best_val_loss:0.03803237991688545, best_auc:0.9896879091570194\n",
      "Epoch:21 Batch:0 Log-loss0.038828663527965546\n",
      "Epoch:21 Batch:80 Log-loss0.03253946453332901\n",
      "Epoch:21 Batch:160 Log-loss0.04685916379094124\n",
      "Epoch:21 Batch:240 Log-loss0.029490752145648003\n",
      "Epoch:21 Batch:320 Log-loss0.04307609796524048\n",
      "Epoch:21 Batch:400 Log-loss0.031725943088531494\n",
      "Epoch:21 Batch:480 Log-loss0.02571762539446354\n",
      "Epoch:21 Batch:560 Log-loss0.02668451890349388\n",
      "Epoch average log loss:0.03881106823682785\n",
      "In Epoch21, val_loss:0.03816268727200053, best_val_loss:0.03803237991688545, best_auc:0.9896879091570194\n",
      "Epoch:22 Batch:0 Log-loss0.03680025786161423\n",
      "Epoch:22 Batch:80 Log-loss0.04546086862683296\n",
      "Epoch:22 Batch:160 Log-loss0.0442456491291523\n",
      "Epoch:22 Batch:240 Log-loss0.035078760236501694\n",
      "Epoch:22 Batch:320 Log-loss0.03624284267425537\n",
      "Epoch:22 Batch:400 Log-loss0.05120740458369255\n",
      "Epoch:22 Batch:480 Log-loss0.05021866783499718\n",
      "Epoch:22 Batch:560 Log-loss0.04428579658269882\n",
      "Epoch average log loss:0.038660746066099295\n",
      "In Epoch22, val_loss:0.03823143004916193, best_val_loss:0.03803237991688545, best_auc:0.9896879091570194\n",
      "Epoch:23 Batch:0 Log-loss0.03745035454630852\n",
      "Epoch:23 Batch:80 Log-loss0.02854137308895588\n",
      "Epoch:23 Batch:160 Log-loss0.03783765807747841\n",
      "Epoch:23 Batch:240 Log-loss0.02939191460609436\n",
      "Epoch:23 Batch:320 Log-loss0.027800386771559715\n",
      "Epoch:23 Batch:400 Log-loss0.03449639305472374\n",
      "Epoch:23 Batch:480 Log-loss0.040002454072237015\n",
      "Epoch:23 Batch:560 Log-loss0.03727341443300247\n",
      "Epoch average log loss:0.03860185711777636\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch23, val_loss:0.03799173887323078, best_val_loss:0.03799173887323078, best_auc:0.9897630406160302\n",
      "Epoch:24 Batch:0 Log-loss0.030677618458867073\n",
      "Epoch:24 Batch:80 Log-loss0.03807050362229347\n",
      "Epoch:24 Batch:160 Log-loss0.039115529507398605\n",
      "Epoch:24 Batch:240 Log-loss0.030378760769963264\n",
      "Epoch:24 Batch:320 Log-loss0.04248499870300293\n",
      "Epoch:24 Batch:400 Log-loss0.04905645176768303\n",
      "Epoch:24 Batch:480 Log-loss0.045934032648801804\n",
      "Epoch:24 Batch:560 Log-loss0.04618142917752266\n",
      "Epoch average log loss:0.038367775295461926\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch24, val_loss:0.03789054162807135, best_val_loss:0.03789054162807135, best_auc:0.9899969162801404\n",
      "Epoch:25 Batch:0 Log-loss0.027369989082217216\n",
      "Epoch:25 Batch:80 Log-loss0.029961133375763893\n",
      "Epoch:25 Batch:160 Log-loss0.0309223011136055\n",
      "Epoch:25 Batch:240 Log-loss0.03690411150455475\n",
      "Epoch:25 Batch:320 Log-loss0.03689288720488548\n",
      "Epoch:25 Batch:400 Log-loss0.03700052574276924\n",
      "Epoch:25 Batch:480 Log-loss0.037081290036439896\n",
      "Epoch:25 Batch:560 Log-loss0.03987631946802139\n",
      "Epoch average log loss:0.03816462139865118\n",
      "In Epoch25, val_loss:0.0382568201752323, best_val_loss:0.03789054162807135, best_auc:0.9899969162801404\n",
      "Epoch:26 Batch:0 Log-loss0.05346767231822014\n",
      "Epoch:26 Batch:80 Log-loss0.02967984788119793\n",
      "Epoch:26 Batch:160 Log-loss0.05241578817367554\n",
      "Epoch:26 Batch:240 Log-loss0.030143490061163902\n",
      "Epoch:26 Batch:320 Log-loss0.03374739736318588\n",
      "Epoch:26 Batch:400 Log-loss0.049923792481422424\n",
      "Epoch:26 Batch:480 Log-loss0.027553260326385498\n",
      "Epoch:26 Batch:560 Log-loss0.03416832908987999\n",
      "Epoch average log loss:0.038126985459322374\n",
      "In Epoch26, val_loss:0.038142528524112713, best_val_loss:0.03789054162807135, best_auc:0.9899969162801404\n",
      "Epoch:27 Batch:0 Log-loss0.050495829433202744\n",
      "Epoch:27 Batch:80 Log-loss0.027470670640468597\n",
      "Epoch:27 Batch:160 Log-loss0.026165053248405457\n",
      "Epoch:27 Batch:240 Log-loss0.03259480372071266\n",
      "Epoch:27 Batch:320 Log-loss0.04293355345726013\n",
      "Epoch:27 Batch:400 Log-loss0.03754238784313202\n",
      "Epoch:27 Batch:480 Log-loss0.03666088357567787\n",
      "Epoch:27 Batch:560 Log-loss0.03147894889116287\n",
      "Epoch average log loss:0.03806833000853658\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch27, val_loss:0.037820321437183176, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:28 Batch:0 Log-loss0.042563557624816895\n",
      "Epoch:28 Batch:80 Log-loss0.03708207979798317\n",
      "Epoch:28 Batch:160 Log-loss0.04973258450627327\n",
      "Epoch:28 Batch:240 Log-loss0.030123427510261536\n",
      "Epoch:28 Batch:320 Log-loss0.04907386004924774\n",
      "Epoch:28 Batch:400 Log-loss0.05166717246174812\n",
      "Epoch:28 Batch:480 Log-loss0.03509337082505226\n",
      "Epoch:28 Batch:560 Log-loss0.04093760624527931\n",
      "Epoch average log loss:0.038134439954800266\n",
      "In Epoch28, val_loss:0.037949507263713746, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:29 Batch:0 Log-loss0.04242418333888054\n",
      "Epoch:29 Batch:80 Log-loss0.032052576541900635\n",
      "Epoch:29 Batch:160 Log-loss0.04607522115111351\n",
      "Epoch:29 Batch:240 Log-loss0.03211600333452225\n",
      "Epoch:29 Batch:320 Log-loss0.06305466592311859\n",
      "Epoch:29 Batch:400 Log-loss0.05832577124238014\n",
      "Epoch:29 Batch:480 Log-loss0.02574724145233631\n",
      "Epoch:29 Batch:560 Log-loss0.03929305821657181\n",
      "Epoch average log loss:0.03798205828892865\n",
      "In Epoch29, val_loss:0.03786773756981707, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:30 Batch:0 Log-loss0.02591921202838421\n",
      "Epoch:30 Batch:80 Log-loss0.033962465822696686\n",
      "Epoch:30 Batch:160 Log-loss0.033928029239177704\n",
      "Epoch:30 Batch:240 Log-loss0.035718392580747604\n",
      "Epoch:30 Batch:320 Log-loss0.04978758096694946\n",
      "Epoch:30 Batch:400 Log-loss0.034477606415748596\n",
      "Epoch:30 Batch:480 Log-loss0.027941538020968437\n",
      "Epoch:30 Batch:560 Log-loss0.027523469179868698\n",
      "Epoch average log loss:0.03782896599067109\n",
      "In Epoch30, val_loss:0.03794131476745904, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:31 Batch:0 Log-loss0.03237154707312584\n",
      "Epoch:31 Batch:80 Log-loss0.027460364624857903\n",
      "Epoch:31 Batch:160 Log-loss0.02717469446361065\n",
      "Epoch:31 Batch:240 Log-loss0.0323241762816906\n",
      "Epoch:31 Batch:320 Log-loss0.045785289257764816\n",
      "Epoch:31 Batch:400 Log-loss0.040498245507478714\n",
      "Epoch:31 Batch:480 Log-loss0.038875218480825424\n",
      "Epoch:31 Batch:560 Log-loss0.03251459077000618\n",
      "Epoch average log loss:0.03784822432457336\n",
      "In Epoch31, val_loss:0.038109137141442505, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:32 Batch:0 Log-loss0.038513701409101486\n",
      "Epoch:32 Batch:80 Log-loss0.034087885171175\n",
      "Epoch:32 Batch:160 Log-loss0.03919614478945732\n",
      "Epoch:32 Batch:240 Log-loss0.022301360964775085\n",
      "Epoch:32 Batch:320 Log-loss0.036715954542160034\n",
      "Epoch:32 Batch:400 Log-loss0.038000814616680145\n",
      "Epoch:32 Batch:480 Log-loss0.03072969801723957\n",
      "Epoch:32 Batch:560 Log-loss0.044055938720703125\n",
      "Epoch average log loss:0.037620040757714635\n",
      "In Epoch32, val_loss:0.03790871566203797, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:33 Batch:0 Log-loss0.0355822779238224\n",
      "Epoch:33 Batch:80 Log-loss0.028630495071411133\n",
      "Epoch:33 Batch:160 Log-loss0.0563642643392086\n",
      "Epoch:33 Batch:240 Log-loss0.03640080243349075\n",
      "Epoch:33 Batch:320 Log-loss0.04177823290228844\n",
      "Epoch:33 Batch:400 Log-loss0.04330683872103691\n",
      "Epoch:33 Batch:480 Log-loss0.03328092768788338\n",
      "Epoch:33 Batch:560 Log-loss0.03357747197151184\n",
      "Epoch average log loss:0.0375242296622933\n",
      "In Epoch33, val_loss:0.03806492439326017, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:34 Batch:0 Log-loss0.04165082797408104\n",
      "Epoch:34 Batch:80 Log-loss0.04208889976143837\n",
      "Epoch:34 Batch:160 Log-loss0.0350065641105175\n",
      "Epoch:34 Batch:240 Log-loss0.05245481804013252\n",
      "Epoch:34 Batch:320 Log-loss0.03919055685400963\n",
      "Epoch:34 Batch:400 Log-loss0.04466426745057106\n",
      "Epoch:34 Batch:480 Log-loss0.04216974973678589\n",
      "Epoch:34 Batch:560 Log-loss0.026806745678186417\n",
      "Epoch average log loss:0.037656849220262044\n",
      "In Epoch34, val_loss:0.037996711119224556, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:35 Batch:0 Log-loss0.03563997521996498\n",
      "Epoch:35 Batch:80 Log-loss0.043650779873132706\n",
      "Epoch:35 Batch:160 Log-loss0.036254432052373886\n",
      "Epoch:35 Batch:240 Log-loss0.043078914284706116\n",
      "Epoch:35 Batch:320 Log-loss0.04636875167489052\n",
      "Epoch:35 Batch:400 Log-loss0.03336885944008827\n",
      "Epoch:35 Batch:480 Log-loss0.041321512311697006\n",
      "Epoch:35 Batch:560 Log-loss0.027898920699954033\n",
      "Epoch average log loss:0.03757093364838511\n",
      "In Epoch35, val_loss:0.0379259290001961, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:36 Batch:0 Log-loss0.03855712711811066\n",
      "Epoch:36 Batch:80 Log-loss0.026920558884739876\n",
      "Epoch:36 Batch:160 Log-loss0.04212645813822746\n",
      "Epoch:36 Batch:240 Log-loss0.04448798671364784\n",
      "Epoch:36 Batch:320 Log-loss0.03532024100422859\n",
      "Epoch:36 Batch:400 Log-loss0.037587251514196396\n",
      "Epoch:36 Batch:480 Log-loss0.044340770691633224\n",
      "Epoch:36 Batch:560 Log-loss0.0352197103202343\n",
      "Epoch average log loss:0.0376686339765521\n",
      "In Epoch36, val_loss:0.0378724079662516, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:37 Batch:0 Log-loss0.03941305726766586\n",
      "Epoch:37 Batch:80 Log-loss0.054207876324653625\n",
      "Epoch:37 Batch:160 Log-loss0.03444637730717659\n",
      "Epoch:37 Batch:240 Log-loss0.03429574891924858\n",
      "Epoch:37 Batch:320 Log-loss0.025822967290878296\n",
      "Epoch:37 Batch:400 Log-loss0.054574284702539444\n",
      "Epoch:37 Batch:480 Log-loss0.042721252888441086\n",
      "Epoch:37 Batch:560 Log-loss0.01781155914068222\n",
      "Epoch average log loss:0.03732727568629863\n",
      "In Epoch37, val_loss:0.03795081782363902, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:38 Batch:0 Log-loss0.046780943870544434\n",
      "Epoch:38 Batch:80 Log-loss0.04150822386145592\n",
      "Epoch:38 Batch:160 Log-loss0.04309031739830971\n",
      "Epoch:38 Batch:240 Log-loss0.022499024868011475\n",
      "Epoch:38 Batch:320 Log-loss0.03588370606303215\n",
      "Epoch:38 Batch:400 Log-loss0.03993812948465347\n",
      "Epoch:38 Batch:480 Log-loss0.03498769551515579\n",
      "Epoch:38 Batch:560 Log-loss0.02272813953459263\n",
      "Epoch average log loss:0.037502008430393675\n",
      "In Epoch38, val_loss:0.037911927686988804, best_val_loss:0.037820321437183176, best_auc:0.9899850017057168\n",
      "Epoch:39 Batch:0 Log-loss0.03396042063832283\n",
      "Epoch:39 Batch:80 Log-loss0.03170320391654968\n",
      "Epoch:39 Batch:160 Log-loss0.03190692141652107\n",
      "Epoch:39 Batch:240 Log-loss0.031037235632538795\n",
      "Epoch:39 Batch:320 Log-loss0.03587095066905022\n",
      "Epoch:39 Batch:400 Log-loss0.03927933797240257\n",
      "Epoch:39 Batch:480 Log-loss0.03872719407081604\n",
      "Epoch:39 Batch:560 Log-loss0.032857660204172134\n",
      "Epoch average log loss:0.037329856642255826\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_646.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 7\n",
      "Epoch:1 Batch:0 Log-loss0.6759501099586487\n",
      "Epoch:1 Batch:80 Log-loss0.17744779586791992\n",
      "Epoch:1 Batch:160 Log-loss0.08842679113149643\n",
      "Epoch:1 Batch:240 Log-loss0.06648658961057663\n",
      "Epoch:1 Batch:320 Log-loss0.07408173382282257\n",
      "Epoch:1 Batch:400 Log-loss0.07038727402687073\n",
      "Epoch:1 Batch:480 Log-loss0.07488864660263062\n",
      "Epoch:1 Batch:560 Log-loss0.05442814156413078\n",
      "Epoch average log loss:0.08970887948359763\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.051003651773934566, best_val_loss:0.051003651773934566, best_auc:0.9712022607902474\n",
      "Epoch:2 Batch:0 Log-loss0.05813344940543175\n",
      "Epoch:2 Batch:80 Log-loss0.04096994176506996\n",
      "Epoch:2 Batch:160 Log-loss0.04668135941028595\n",
      "Epoch:2 Batch:240 Log-loss0.03806786611676216\n",
      "Epoch:2 Batch:320 Log-loss0.05059603229165077\n",
      "Epoch:2 Batch:400 Log-loss0.052529171109199524\n",
      "Epoch:2 Batch:480 Log-loss0.07132703065872192\n",
      "Epoch:2 Batch:560 Log-loss0.04518316686153412\n",
      "Epoch average log loss:0.05213568683913244\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.04754479108135381, best_val_loss:0.04754479108135381, best_auc:0.9763672478103427\n",
      "Epoch:3 Batch:0 Log-loss0.04290556535124779\n",
      "Epoch:3 Batch:80 Log-loss0.04081616923213005\n",
      "Epoch:3 Batch:160 Log-loss0.057355642318725586\n",
      "Epoch:3 Batch:240 Log-loss0.05392298102378845\n",
      "Epoch:3 Batch:320 Log-loss0.03890067711472511\n",
      "Epoch:3 Batch:400 Log-loss0.04413892701268196\n",
      "Epoch:3 Batch:480 Log-loss0.039570797234773636\n",
      "Epoch:3 Batch:560 Log-loss0.04377703368663788\n",
      "Epoch average log loss:0.048738475589613826\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.045119199028261274, best_val_loss:0.045119199028261274, best_auc:0.9802461087494247\n",
      "Epoch:4 Batch:0 Log-loss0.03921903297305107\n",
      "Epoch:4 Batch:80 Log-loss0.03725956007838249\n",
      "Epoch:4 Batch:160 Log-loss0.04419875517487526\n",
      "Epoch:4 Batch:240 Log-loss0.03595251217484474\n",
      "Epoch:4 Batch:320 Log-loss0.04778338968753815\n",
      "Epoch:4 Batch:400 Log-loss0.057670462876558304\n",
      "Epoch:4 Batch:480 Log-loss0.05629753693938255\n",
      "Epoch:4 Batch:560 Log-loss0.039483338594436646\n",
      "Epoch average log loss:0.046402782587600604\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.043733682276988345, best_val_loss:0.043733682276988345, best_auc:0.9824646030042133\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:5 Batch:0 Log-loss0.04819232597947121\n",
      "Epoch:5 Batch:80 Log-loss0.046335697174072266\n",
      "Epoch:5 Batch:160 Log-loss0.03540770336985588\n",
      "Epoch:5 Batch:240 Log-loss0.054816752672195435\n",
      "Epoch:5 Batch:320 Log-loss0.05290122330188751\n",
      "Epoch:5 Batch:400 Log-loss0.05128878355026245\n",
      "Epoch:5 Batch:480 Log-loss0.04393814876675606\n",
      "Epoch:5 Batch:560 Log-loss0.0402793250977993\n",
      "Epoch average log loss:0.044765238457226325\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.0425990939515009, best_val_loss:0.0425990939515009, best_auc:0.9835963027268328\n",
      "Epoch:6 Batch:0 Log-loss0.043334852904081345\n",
      "Epoch:6 Batch:80 Log-loss0.04659324884414673\n",
      "Epoch:6 Batch:160 Log-loss0.04048130288720131\n",
      "Epoch:6 Batch:240 Log-loss0.033043477684259415\n",
      "Epoch:6 Batch:320 Log-loss0.055700402706861496\n",
      "Epoch:6 Batch:400 Log-loss0.04544730857014656\n",
      "Epoch:6 Batch:480 Log-loss0.03856174275279045\n",
      "Epoch:6 Batch:560 Log-loss0.03481787443161011\n",
      "Epoch average log loss:0.043945174624345135\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.042202091305193694, best_val_loss:0.042202091305193694, best_auc:0.9843106317457743\n",
      "Epoch:7 Batch:0 Log-loss0.051543042063713074\n",
      "Epoch:7 Batch:80 Log-loss0.060953300446271896\n",
      "Epoch:7 Batch:160 Log-loss0.04732339084148407\n",
      "Epoch:7 Batch:240 Log-loss0.057764261960983276\n",
      "Epoch:7 Batch:320 Log-loss0.023717815056443214\n",
      "Epoch:7 Batch:400 Log-loss0.03516273945569992\n",
      "Epoch:7 Batch:480 Log-loss0.04303370416164398\n",
      "Epoch:7 Batch:560 Log-loss0.031351007521152496\n",
      "Epoch average log loss:0.043267529238281505\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.04157635179657022, best_val_loss:0.04157635179657022, best_auc:0.9854245830898418\n",
      "Epoch:8 Batch:0 Log-loss0.04947705194354057\n",
      "Epoch:8 Batch:80 Log-loss0.05101880803704262\n",
      "Epoch:8 Batch:160 Log-loss0.048350054770708084\n",
      "Epoch:8 Batch:240 Log-loss0.03018815815448761\n",
      "Epoch:8 Batch:320 Log-loss0.040826477110385895\n",
      "Epoch:8 Batch:400 Log-loss0.04577584192156792\n",
      "Epoch:8 Batch:480 Log-loss0.03957519307732582\n",
      "Epoch:8 Batch:560 Log-loss0.044029027223587036\n",
      "Epoch average log loss:0.042606753603156126\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.04145085945779889, best_val_loss:0.04145085945779889, best_auc:0.9858916829128247\n",
      "Epoch:9 Batch:0 Log-loss0.041721854358911514\n",
      "Epoch:9 Batch:80 Log-loss0.03415345400571823\n",
      "Epoch:9 Batch:160 Log-loss0.037905268371105194\n",
      "Epoch:9 Batch:240 Log-loss0.04608927667140961\n",
      "Epoch:9 Batch:320 Log-loss0.0365360789000988\n",
      "Epoch:9 Batch:400 Log-loss0.04680008813738823\n",
      "Epoch:9 Batch:480 Log-loss0.03700898587703705\n",
      "Epoch:9 Batch:560 Log-loss0.052295129746198654\n",
      "Epoch average log loss:0.04189133437549961\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.040837145584466446, best_val_loss:0.040837145584466446, best_auc:0.9863552419983401\n",
      "Epoch:10 Batch:0 Log-loss0.03551487624645233\n",
      "Epoch:10 Batch:80 Log-loss0.04055884853005409\n",
      "Epoch:10 Batch:160 Log-loss0.05574118718504906\n",
      "Epoch:10 Batch:240 Log-loss0.03689048811793327\n",
      "Epoch:10 Batch:320 Log-loss0.05886497721076012\n",
      "Epoch:10 Batch:400 Log-loss0.03696708753705025\n",
      "Epoch:10 Batch:480 Log-loss0.03488166257739067\n",
      "Epoch:10 Batch:560 Log-loss0.049001507461071014\n",
      "Epoch average log loss:0.04123564368845629\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.04069070118169419, best_val_loss:0.04069070118169419, best_auc:0.9869858475549665\n",
      "Epoch:11 Batch:0 Log-loss0.040575090795755386\n",
      "Epoch:11 Batch:80 Log-loss0.03587931767106056\n",
      "Epoch:11 Batch:160 Log-loss0.04388843849301338\n",
      "Epoch:11 Batch:240 Log-loss0.039198096841573715\n",
      "Epoch:11 Batch:320 Log-loss0.030683813616633415\n",
      "Epoch:11 Batch:400 Log-loss0.034808237105607986\n",
      "Epoch:11 Batch:480 Log-loss0.05013794079422951\n",
      "Epoch:11 Batch:560 Log-loss0.04487478360533714\n",
      "Epoch average log loss:0.04100576022174209\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch11, val_loss:0.04056294141628942, best_val_loss:0.04056294141628942, best_auc:0.9872572489629897\n",
      "Epoch:12 Batch:0 Log-loss0.030716190114617348\n",
      "Epoch:12 Batch:80 Log-loss0.033711932599544525\n",
      "Epoch:12 Batch:160 Log-loss0.034250278025865555\n",
      "Epoch:12 Batch:240 Log-loss0.043780338019132614\n",
      "Epoch:12 Batch:320 Log-loss0.030012451112270355\n",
      "Epoch:12 Batch:400 Log-loss0.04120199754834175\n",
      "Epoch:12 Batch:480 Log-loss0.03595530614256859\n",
      "Epoch:12 Batch:560 Log-loss0.0431768000125885\n",
      "Epoch average log loss:0.040644294183169095\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.040411454923132535, best_val_loss:0.040411454923132535, best_auc:0.9874315727431044\n",
      "Epoch:13 Batch:0 Log-loss0.053831521421670914\n",
      "Epoch:13 Batch:80 Log-loss0.0505223385989666\n",
      "Epoch:13 Batch:160 Log-loss0.028958536684513092\n",
      "Epoch:13 Batch:240 Log-loss0.03293760120868683\n",
      "Epoch:13 Batch:320 Log-loss0.052192267030477524\n",
      "Epoch:13 Batch:400 Log-loss0.043387770652770996\n",
      "Epoch:13 Batch:480 Log-loss0.04765729978680611\n",
      "Epoch:13 Batch:560 Log-loss0.03253922611474991\n",
      "Epoch average log loss:0.04036987969385726\n",
      "In Epoch13, val_loss:0.040675841488702784, best_val_loss:0.040411454923132535, best_auc:0.9874315727431044\n",
      "Epoch:14 Batch:0 Log-loss0.038491204380989075\n",
      "Epoch:14 Batch:80 Log-loss0.03169099614024162\n",
      "Epoch:14 Batch:160 Log-loss0.029460681602358818\n",
      "Epoch:14 Batch:240 Log-loss0.049011409282684326\n",
      "Epoch:14 Batch:320 Log-loss0.045437734574079514\n",
      "Epoch:14 Batch:400 Log-loss0.03245794400572777\n",
      "Epoch:14 Batch:480 Log-loss0.034001972526311874\n",
      "Epoch:14 Batch:560 Log-loss0.03455253317952156\n",
      "Epoch average log loss:0.04005755454168788\n",
      "In Epoch14, val_loss:0.04045710354818694, best_val_loss:0.040411454923132535, best_auc:0.9874315727431044\n",
      "Epoch:15 Batch:0 Log-loss0.0501917265355587\n",
      "Epoch:15 Batch:80 Log-loss0.021760275587439537\n",
      "Epoch:15 Batch:160 Log-loss0.057808417826890945\n",
      "Epoch:15 Batch:240 Log-loss0.03580000624060631\n",
      "Epoch:15 Batch:320 Log-loss0.0485335998237133\n",
      "Epoch:15 Batch:400 Log-loss0.03234444558620453\n",
      "Epoch:15 Batch:480 Log-loss0.046691637486219406\n",
      "Epoch:15 Batch:560 Log-loss0.040976010262966156\n",
      "Epoch average log loss:0.039749749856335774\n",
      "In Epoch15, val_loss:0.04045305609571441, best_val_loss:0.040411454923132535, best_auc:0.9874315727431044\n",
      "Epoch:16 Batch:0 Log-loss0.03479643166065216\n",
      "Epoch:16 Batch:80 Log-loss0.029457589611411095\n",
      "Epoch:16 Batch:160 Log-loss0.02991652302443981\n",
      "Epoch:16 Batch:240 Log-loss0.029535988345742226\n",
      "Epoch:16 Batch:320 Log-loss0.04872944951057434\n",
      "Epoch:16 Batch:400 Log-loss0.027622967958450317\n",
      "Epoch:16 Batch:480 Log-loss0.03812631592154503\n",
      "Epoch:16 Batch:560 Log-loss0.04121587797999382\n",
      "Epoch average log loss:0.039467609319503286\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch16, val_loss:0.04008444370997312, best_val_loss:0.04008444370997312, best_auc:0.9876213674362558\n",
      "Epoch:17 Batch:0 Log-loss0.03281615674495697\n",
      "Epoch:17 Batch:80 Log-loss0.040447283536195755\n",
      "Epoch:17 Batch:160 Log-loss0.042149562388658524\n",
      "Epoch:17 Batch:240 Log-loss0.04961399734020233\n",
      "Epoch:17 Batch:320 Log-loss0.03783635050058365\n",
      "Epoch:17 Batch:400 Log-loss0.03376669064164162\n",
      "Epoch:17 Batch:480 Log-loss0.024462297558784485\n",
      "Epoch:17 Batch:560 Log-loss0.039777521044015884\n",
      "Epoch average log loss:0.0392900795476245\n",
      "In Epoch17, val_loss:0.04008540501729454, best_val_loss:0.04008444370997312, best_auc:0.9876213674362558\n",
      "Epoch:18 Batch:0 Log-loss0.04564591124653816\n",
      "Epoch:18 Batch:80 Log-loss0.049741312861442566\n",
      "Epoch:18 Batch:160 Log-loss0.04016975685954094\n",
      "Epoch:18 Batch:240 Log-loss0.05241302028298378\n",
      "Epoch:18 Batch:320 Log-loss0.03854672983288765\n",
      "Epoch:18 Batch:400 Log-loss0.03118354082107544\n",
      "Epoch:18 Batch:480 Log-loss0.0401654988527298\n",
      "Epoch:18 Batch:560 Log-loss0.031893253326416016\n",
      "Epoch average log loss:0.038944887439720334\n",
      "In Epoch18, val_loss:0.040195409195339, best_val_loss:0.04008444370997312, best_auc:0.9876213674362558\n",
      "Epoch:19 Batch:0 Log-loss0.034125033766031265\n",
      "Epoch:19 Batch:80 Log-loss0.046873539686203\n",
      "Epoch:19 Batch:160 Log-loss0.03162873163819313\n",
      "Epoch:19 Batch:240 Log-loss0.036124102771282196\n",
      "Epoch:19 Batch:320 Log-loss0.05704967677593231\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:19 Batch:400 Log-loss0.03831661120057106\n",
      "Epoch:19 Batch:480 Log-loss0.032637886703014374\n",
      "Epoch:19 Batch:560 Log-loss0.034789133816957474\n",
      "Epoch average log loss:0.039025595993734896\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch19, val_loss:0.04006720200186018, best_val_loss:0.04006720200186018, best_auc:0.9880115490992544\n",
      "Epoch:20 Batch:0 Log-loss0.03694509342312813\n",
      "Epoch:20 Batch:80 Log-loss0.03268841654062271\n",
      "Epoch:20 Batch:160 Log-loss0.04011614993214607\n",
      "Epoch:20 Batch:240 Log-loss0.04694761708378792\n",
      "Epoch:20 Batch:320 Log-loss0.03771110624074936\n",
      "Epoch:20 Batch:400 Log-loss0.0586770661175251\n",
      "Epoch:20 Batch:480 Log-loss0.034826524555683136\n",
      "Epoch:20 Batch:560 Log-loss0.03439858555793762\n",
      "Epoch average log loss:0.03901751371366637\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch20, val_loss:0.03997707306546649, best_val_loss:0.03997707306546649, best_auc:0.9880455643533805\n",
      "Epoch:21 Batch:0 Log-loss0.03481486812233925\n",
      "Epoch:21 Batch:80 Log-loss0.04073992371559143\n",
      "Epoch:21 Batch:160 Log-loss0.02894042618572712\n",
      "Epoch:21 Batch:240 Log-loss0.03875547647476196\n",
      "Epoch:21 Batch:320 Log-loss0.05314961448311806\n",
      "Epoch:21 Batch:400 Log-loss0.036631952971220016\n",
      "Epoch:21 Batch:480 Log-loss0.044846344739198685\n",
      "Epoch:21 Batch:560 Log-loss0.025908268988132477\n",
      "Epoch average log loss:0.03841979594435543\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch21, val_loss:0.03996969280096261, best_val_loss:0.03996969280096261, best_auc:0.9880728612757967\n",
      "Epoch:22 Batch:0 Log-loss0.028723545372486115\n",
      "Epoch:22 Batch:80 Log-loss0.03480848670005798\n",
      "Epoch:22 Batch:160 Log-loss0.029135646298527718\n",
      "Epoch:22 Batch:240 Log-loss0.030496127903461456\n",
      "Epoch:22 Batch:320 Log-loss0.05527249351143837\n",
      "Epoch:22 Batch:400 Log-loss0.049305424094200134\n",
      "Epoch:22 Batch:480 Log-loss0.049723345786333084\n",
      "Epoch:22 Batch:560 Log-loss0.03142191097140312\n",
      "Epoch average log loss:0.038464350875334015\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch22, val_loss:0.03985915546031043, best_val_loss:0.03985915546031043, best_auc:0.988084851205245\n",
      "Epoch:23 Batch:0 Log-loss0.03183381259441376\n",
      "Epoch:23 Batch:80 Log-loss0.03249286487698555\n",
      "Epoch:23 Batch:160 Log-loss0.026825370267033577\n",
      "Epoch:23 Batch:240 Log-loss0.03456047549843788\n",
      "Epoch:23 Batch:320 Log-loss0.05149709805846214\n",
      "Epoch:23 Batch:400 Log-loss0.03173712268471718\n",
      "Epoch:23 Batch:480 Log-loss0.04940259829163551\n",
      "Epoch:23 Batch:560 Log-loss0.03566466271877289\n",
      "Epoch average log loss:0.03806385669325079\n",
      "In Epoch23, val_loss:0.0399579755859294, best_val_loss:0.03985915546031043, best_auc:0.988084851205245\n",
      "Epoch:24 Batch:0 Log-loss0.039346542209386826\n",
      "Epoch:24 Batch:80 Log-loss0.04795931279659271\n",
      "Epoch:24 Batch:160 Log-loss0.03572539612650871\n",
      "Epoch:24 Batch:240 Log-loss0.046971991658210754\n",
      "Epoch:24 Batch:320 Log-loss0.027752744033932686\n",
      "Epoch:24 Batch:400 Log-loss0.04948998615145683\n",
      "Epoch:24 Batch:480 Log-loss0.03427843749523163\n",
      "Epoch:24 Batch:560 Log-loss0.04985954612493515\n",
      "Epoch average log loss:0.03808231869147026\n",
      "In Epoch24, val_loss:0.03987047820900289, best_val_loss:0.03985915546031043, best_auc:0.988084851205245\n",
      "Epoch:25 Batch:0 Log-loss0.054493244737386703\n",
      "Epoch:25 Batch:80 Log-loss0.03433459624648094\n",
      "Epoch:25 Batch:160 Log-loss0.0438581258058548\n",
      "Epoch:25 Batch:240 Log-loss0.050882238894701004\n",
      "Epoch:25 Batch:320 Log-loss0.03559335693717003\n",
      "Epoch:25 Batch:400 Log-loss0.03271080181002617\n",
      "Epoch:25 Batch:480 Log-loss0.029436500743031502\n",
      "Epoch:25 Batch:560 Log-loss0.029902970418334007\n",
      "Epoch average log loss:0.038118210388347507\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch25, val_loss:0.03969802070081259, best_val_loss:0.03969802070081259, best_auc:0.9882708219497437\n",
      "Epoch:26 Batch:0 Log-loss0.0434524267911911\n",
      "Epoch:26 Batch:80 Log-loss0.03424379229545593\n",
      "Epoch:26 Batch:160 Log-loss0.044108614325523376\n",
      "Epoch:26 Batch:240 Log-loss0.02947893552482128\n",
      "Epoch:26 Batch:320 Log-loss0.047883953899145126\n",
      "Epoch:26 Batch:400 Log-loss0.03332795947790146\n",
      "Epoch:26 Batch:480 Log-loss0.034452807158231735\n",
      "Epoch:26 Batch:560 Log-loss0.03269960358738899\n",
      "Epoch average log loss:0.03810258448855686\n",
      "In Epoch26, val_loss:0.03979154884065409, best_val_loss:0.03969802070081259, best_auc:0.9882708219497437\n",
      "Epoch:27 Batch:0 Log-loss0.025695517659187317\n",
      "Epoch:27 Batch:80 Log-loss0.03225385770201683\n",
      "Epoch:27 Batch:160 Log-loss0.04377329722046852\n",
      "Epoch:27 Batch:240 Log-loss0.0427875854074955\n",
      "Epoch:27 Batch:320 Log-loss0.03128821775317192\n",
      "Epoch:27 Batch:400 Log-loss0.03908607363700867\n",
      "Epoch:27 Batch:480 Log-loss0.03979849815368652\n",
      "Epoch:27 Batch:560 Log-loss0.023547502234578133\n",
      "Epoch average log loss:0.037969706739698135\n",
      "In Epoch27, val_loss:0.04006778121908618, best_val_loss:0.03969802070081259, best_auc:0.9882708219497437\n",
      "Epoch:28 Batch:0 Log-loss0.026562564074993134\n",
      "Epoch:28 Batch:80 Log-loss0.03958018496632576\n",
      "Epoch:28 Batch:160 Log-loss0.04519592598080635\n",
      "Epoch:28 Batch:240 Log-loss0.030276469886302948\n",
      "Epoch:28 Batch:320 Log-loss0.03831103816628456\n",
      "Epoch:28 Batch:400 Log-loss0.03604623302817345\n",
      "Epoch:28 Batch:480 Log-loss0.04702218994498253\n",
      "Epoch:28 Batch:560 Log-loss0.03939364477992058\n",
      "Epoch average log loss:0.03780478696904278\n",
      "In Epoch28, val_loss:0.03990999060112945, best_val_loss:0.03969802070081259, best_auc:0.9882708219497437\n",
      "Epoch:29 Batch:0 Log-loss0.03694024309515953\n",
      "Epoch:29 Batch:80 Log-loss0.02893284521996975\n",
      "Epoch:29 Batch:160 Log-loss0.0374317392706871\n",
      "Epoch:29 Batch:240 Log-loss0.022543879225850105\n",
      "Epoch:29 Batch:320 Log-loss0.04580897465348244\n",
      "Epoch:29 Batch:400 Log-loss0.04206370934844017\n",
      "Epoch:29 Batch:480 Log-loss0.03809138759970665\n",
      "Epoch:29 Batch:560 Log-loss0.037980109453201294\n",
      "Epoch average log loss:0.03784239355674279\n",
      "In Epoch29, val_loss:0.03976455127329414, best_val_loss:0.03969802070081259, best_auc:0.9882708219497437\n",
      "Epoch:30 Batch:0 Log-loss0.03552478924393654\n",
      "Epoch:30 Batch:80 Log-loss0.02471783012151718\n",
      "Epoch:30 Batch:160 Log-loss0.032607611268758774\n",
      "Epoch:30 Batch:240 Log-loss0.02861209213733673\n",
      "Epoch:30 Batch:320 Log-loss0.04315805062651634\n",
      "Epoch:30 Batch:400 Log-loss0.04498827084898949\n",
      "Epoch:30 Batch:480 Log-loss0.030084028840065002\n",
      "Epoch:30 Batch:560 Log-loss0.047215238213539124\n",
      "Epoch average log loss:0.0375018259709967\n",
      "In Epoch30, val_loss:0.03983005312094392, best_val_loss:0.03969802070081259, best_auc:0.9882708219497437\n",
      "Epoch:31 Batch:0 Log-loss0.040317896753549576\n",
      "Epoch:31 Batch:80 Log-loss0.03909233212471008\n",
      "Epoch:31 Batch:160 Log-loss0.03487880900502205\n",
      "Epoch:31 Batch:240 Log-loss0.030257107689976692\n",
      "Epoch:31 Batch:320 Log-loss0.03351285681128502\n",
      "Epoch:31 Batch:400 Log-loss0.037968896329402924\n",
      "Epoch:31 Batch:480 Log-loss0.03743505850434303\n",
      "Epoch:31 Batch:560 Log-loss0.029078587889671326\n",
      "Epoch average log loss:0.03746293081369783\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch31, val_loss:0.039679786274804695, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:32 Batch:0 Log-loss0.06753794848918915\n",
      "Epoch:32 Batch:80 Log-loss0.03298216313123703\n",
      "Epoch:32 Batch:160 Log-loss0.03301629796624184\n",
      "Epoch:32 Batch:240 Log-loss0.03543683513998985\n",
      "Epoch:32 Batch:320 Log-loss0.041444748640060425\n",
      "Epoch:32 Batch:400 Log-loss0.028585610911250114\n",
      "Epoch:32 Batch:480 Log-loss0.04639727249741554\n",
      "Epoch:32 Batch:560 Log-loss0.04535944014787674\n",
      "Epoch average log loss:0.037488155905157326\n",
      "In Epoch32, val_loss:0.03974641772824252, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:33 Batch:0 Log-loss0.048919882625341415\n",
      "Epoch:33 Batch:80 Log-loss0.02965518645942211\n",
      "Epoch:33 Batch:160 Log-loss0.05121950805187225\n",
      "Epoch:33 Batch:240 Log-loss0.032782088965177536\n",
      "Epoch:33 Batch:320 Log-loss0.0270868930965662\n",
      "Epoch:33 Batch:400 Log-loss0.030563853681087494\n",
      "Epoch:33 Batch:480 Log-loss0.029119955375790596\n",
      "Epoch:33 Batch:560 Log-loss0.04391083866357803\n",
      "Epoch average log loss:0.03749058838002384\n",
      "In Epoch33, val_loss:0.039818715156859676, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:34 Batch:0 Log-loss0.021502474322915077\n",
      "Epoch:34 Batch:80 Log-loss0.042782966047525406\n",
      "Epoch:34 Batch:160 Log-loss0.03672559931874275\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:34 Batch:240 Log-loss0.039723414927721024\n",
      "Epoch:34 Batch:320 Log-loss0.04476425051689148\n",
      "Epoch:34 Batch:400 Log-loss0.029548151418566704\n",
      "Epoch:34 Batch:480 Log-loss0.032862577587366104\n",
      "Epoch:34 Batch:560 Log-loss0.034602243453264236\n",
      "Epoch average log loss:0.037255329251222845\n",
      "In Epoch34, val_loss:0.03994183416685969, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:35 Batch:0 Log-loss0.0524764321744442\n",
      "Epoch:35 Batch:80 Log-loss0.035855259746313095\n",
      "Epoch:35 Batch:160 Log-loss0.03219010308384895\n",
      "Epoch:35 Batch:240 Log-loss0.023379230871796608\n",
      "Epoch:35 Batch:320 Log-loss0.03696596249938011\n",
      "Epoch:35 Batch:400 Log-loss0.027117548510432243\n",
      "Epoch:35 Batch:480 Log-loss0.03594191372394562\n",
      "Epoch:35 Batch:560 Log-loss0.04457966983318329\n",
      "Epoch average log loss:0.037166668667591045\n",
      "In Epoch35, val_loss:0.03995155345286712, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:36 Batch:0 Log-loss0.033967748284339905\n",
      "Epoch:36 Batch:80 Log-loss0.042463283985853195\n",
      "Epoch:36 Batch:160 Log-loss0.03489692881703377\n",
      "Epoch:36 Batch:240 Log-loss0.037167761474847794\n",
      "Epoch:36 Batch:320 Log-loss0.03517281636595726\n",
      "Epoch:36 Batch:400 Log-loss0.04660572111606598\n",
      "Epoch:36 Batch:480 Log-loss0.03232679143548012\n",
      "Epoch:36 Batch:560 Log-loss0.04822191223502159\n",
      "Epoch average log loss:0.03741804759650092\n",
      "In Epoch36, val_loss:0.039818250481654195, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:37 Batch:0 Log-loss0.06873020529747009\n",
      "Epoch:37 Batch:80 Log-loss0.027552103623747826\n",
      "Epoch:37 Batch:160 Log-loss0.027690134942531586\n",
      "Epoch:37 Batch:240 Log-loss0.03067191131412983\n",
      "Epoch:37 Batch:320 Log-loss0.03408404812216759\n",
      "Epoch:37 Batch:400 Log-loss0.028128452599048615\n",
      "Epoch:37 Batch:480 Log-loss0.05047541484236717\n",
      "Epoch:37 Batch:560 Log-loss0.022434627637267113\n",
      "Epoch average log loss:0.0373590559166457\n",
      "In Epoch37, val_loss:0.03973292096832207, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:38 Batch:0 Log-loss0.041544124484062195\n",
      "Epoch:38 Batch:80 Log-loss0.03186434507369995\n",
      "Epoch:38 Batch:160 Log-loss0.04008231684565544\n",
      "Epoch:38 Batch:240 Log-loss0.02594628930091858\n",
      "Epoch:38 Batch:320 Log-loss0.036772049963474274\n",
      "Epoch:38 Batch:400 Log-loss0.04682695493102074\n",
      "Epoch:38 Batch:480 Log-loss0.04418924078345299\n",
      "Epoch:38 Batch:560 Log-loss0.04426531493663788\n",
      "Epoch average log loss:0.037217656230287895\n",
      "In Epoch38, val_loss:0.03976892609078338, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:39 Batch:0 Log-loss0.026215331628918648\n",
      "Epoch:39 Batch:80 Log-loss0.034353096038103104\n",
      "Epoch:39 Batch:160 Log-loss0.04922330752015114\n",
      "Epoch:39 Batch:240 Log-loss0.029170634225010872\n",
      "Epoch:39 Batch:320 Log-loss0.043309178203344345\n",
      "Epoch:39 Batch:400 Log-loss0.031070314347743988\n",
      "Epoch:39 Batch:480 Log-loss0.03385506197810173\n",
      "Epoch:39 Batch:560 Log-loss0.033193789422512054\n",
      "Epoch average log loss:0.03742903997522912\n",
      "In Epoch39, val_loss:0.03971828764334586, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:40 Batch:0 Log-loss0.03348748758435249\n",
      "Epoch:40 Batch:80 Log-loss0.03683379292488098\n",
      "Epoch:40 Batch:160 Log-loss0.015654750168323517\n",
      "Epoch:40 Batch:240 Log-loss0.025938449427485466\n",
      "Epoch:40 Batch:320 Log-loss0.03662684187293053\n",
      "Epoch:40 Batch:400 Log-loss0.028572572395205498\n",
      "Epoch:40 Batch:480 Log-loss0.03873590752482414\n",
      "Epoch:40 Batch:560 Log-loss0.03015575185418129\n",
      "Epoch average log loss:0.03715227232314646\n",
      "In Epoch40, val_loss:0.03974929997138615, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:41 Batch:0 Log-loss0.04302601143717766\n",
      "Epoch:41 Batch:80 Log-loss0.048574745655059814\n",
      "Epoch:41 Batch:160 Log-loss0.030017437413334846\n",
      "Epoch:41 Batch:240 Log-loss0.04252859577536583\n",
      "Epoch:41 Batch:320 Log-loss0.03438721224665642\n",
      "Epoch:41 Batch:400 Log-loss0.02956731803715229\n",
      "Epoch:41 Batch:480 Log-loss0.03325489163398743\n",
      "Epoch:41 Batch:560 Log-loss0.04162978753447533\n",
      "Epoch average log loss:0.03692203135495739\n",
      "In Epoch41, val_loss:0.039800702148453916, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:42 Batch:0 Log-loss0.04387080296874046\n",
      "Epoch:42 Batch:80 Log-loss0.033776186406612396\n",
      "Epoch:42 Batch:160 Log-loss0.04038259759545326\n",
      "Epoch:42 Batch:240 Log-loss0.03797962889075279\n",
      "Epoch:42 Batch:320 Log-loss0.045839887112379074\n",
      "Epoch:42 Batch:400 Log-loss0.04250712692737579\n",
      "Epoch:42 Batch:480 Log-loss0.040068741887807846\n",
      "Epoch:42 Batch:560 Log-loss0.031743403524160385\n",
      "Epoch average log loss:0.036941566933611676\n",
      "In Epoch42, val_loss:0.03985109455742978, best_val_loss:0.039679786274804695, best_auc:0.9883647863109658\n",
      "Epoch:43 Batch:0 Log-loss0.03666197136044502\n",
      "Epoch:43 Batch:80 Log-loss0.03823472559452057\n",
      "Epoch:43 Batch:160 Log-loss0.03153496980667114\n",
      "Epoch:43 Batch:240 Log-loss0.03142831102013588\n",
      "Epoch:43 Batch:320 Log-loss0.035728901624679565\n",
      "Epoch:43 Batch:400 Log-loss0.036126844584941864\n",
      "Epoch:43 Batch:480 Log-loss0.03876423090696335\n",
      "Epoch:43 Batch:560 Log-loss0.021912718191742897\n",
      "Epoch average log loss:0.03705122439922499\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_647.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 8\n",
      "Epoch:1 Batch:0 Log-loss0.6955711841583252\n",
      "Epoch:1 Batch:80 Log-loss0.14567680656909943\n",
      "Epoch:1 Batch:160 Log-loss0.054937493056058884\n",
      "Epoch:1 Batch:240 Log-loss0.0707339197397232\n",
      "Epoch:1 Batch:320 Log-loss0.07430868595838547\n",
      "Epoch:1 Batch:400 Log-loss0.060859352350234985\n",
      "Epoch:1 Batch:480 Log-loss0.07331468909978867\n",
      "Epoch:1 Batch:560 Log-loss0.05053504556417465\n",
      "Epoch average log loss:0.08817678818346135\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.048944909618802906, best_val_loss:0.048944909618802906, best_auc:0.9749769802439404\n",
      "Epoch:2 Batch:0 Log-loss0.046087417751550674\n",
      "Epoch:2 Batch:80 Log-loss0.04842565581202507\n",
      "Epoch:2 Batch:160 Log-loss0.059955764561891556\n",
      "Epoch:2 Batch:240 Log-loss0.05267603322863579\n",
      "Epoch:2 Batch:320 Log-loss0.05708073079586029\n",
      "Epoch:2 Batch:400 Log-loss0.05010464787483215\n",
      "Epoch:2 Batch:480 Log-loss0.027244215831160545\n",
      "Epoch:2 Batch:560 Log-loss0.05575468763709068\n",
      "Epoch average log loss:0.05307486849570913\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.046002701370328476, best_val_loss:0.046002701370328476, best_auc:0.9778178129850158\n",
      "Epoch:3 Batch:0 Log-loss0.035642240196466446\n",
      "Epoch:3 Batch:80 Log-loss0.04351290687918663\n",
      "Epoch:3 Batch:160 Log-loss0.044346097856760025\n",
      "Epoch:3 Batch:240 Log-loss0.05191650614142418\n",
      "Epoch:3 Batch:320 Log-loss0.05360223352909088\n",
      "Epoch:3 Batch:400 Log-loss0.04136073961853981\n",
      "Epoch:3 Batch:480 Log-loss0.04713471606373787\n",
      "Epoch:3 Batch:560 Log-loss0.02610035613179207\n",
      "Epoch average log loss:0.049992712001715386\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.04450914898759919, best_val_loss:0.04450914898759919, best_auc:0.9805634053458075\n",
      "Epoch:4 Batch:0 Log-loss0.04104205593466759\n",
      "Epoch:4 Batch:80 Log-loss0.044253747910261154\n",
      "Epoch:4 Batch:160 Log-loss0.029731256887316704\n",
      "Epoch:4 Batch:240 Log-loss0.04108789935708046\n",
      "Epoch:4 Batch:320 Log-loss0.042184893041849136\n",
      "Epoch:4 Batch:400 Log-loss0.059324804693460464\n",
      "Epoch:4 Batch:480 Log-loss0.04121169075369835\n",
      "Epoch:4 Batch:560 Log-loss0.040521640330553055\n",
      "Epoch average log loss:0.04816411657125822\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.04329157056497099, best_val_loss:0.04329157056497099, best_auc:0.9827106736029675\n",
      "Epoch:5 Batch:0 Log-loss0.059403348714113235\n",
      "Epoch:5 Batch:80 Log-loss0.047285545617341995\n",
      "Epoch:5 Batch:160 Log-loss0.04061711207032204\n",
      "Epoch:5 Batch:240 Log-loss0.052728597074747086\n",
      "Epoch:5 Batch:320 Log-loss0.04538508877158165\n",
      "Epoch:5 Batch:400 Log-loss0.0315374992787838\n",
      "Epoch:5 Batch:480 Log-loss0.0441100150346756\n",
      "Epoch:5 Batch:560 Log-loss0.062104467302560806\n",
      "Epoch average log loss:0.046352025011687406\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.04188258155686933, best_val_loss:0.04188258155686933, best_auc:0.9849775100302244\n",
      "Epoch:6 Batch:0 Log-loss0.04427951201796532\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:6 Batch:80 Log-loss0.03814166039228439\n",
      "Epoch:6 Batch:160 Log-loss0.056268360465765\n",
      "Epoch:6 Batch:240 Log-loss0.047410447150468826\n",
      "Epoch:6 Batch:320 Log-loss0.055771712213754654\n",
      "Epoch:6 Batch:400 Log-loss0.03949829563498497\n",
      "Epoch:6 Batch:480 Log-loss0.030791571363806725\n",
      "Epoch:6 Batch:560 Log-loss0.03229568898677826\n",
      "Epoch average log loss:0.044981272565200923\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.04155889092983551, best_val_loss:0.04155889092983551, best_auc:0.9859867698818442\n",
      "Epoch:7 Batch:0 Log-loss0.044706087559461594\n",
      "Epoch:7 Batch:80 Log-loss0.04482072964310646\n",
      "Epoch:7 Batch:160 Log-loss0.03476889804005623\n",
      "Epoch:7 Batch:240 Log-loss0.05308767780661583\n",
      "Epoch:7 Batch:320 Log-loss0.04733544960618019\n",
      "Epoch:7 Batch:400 Log-loss0.04073473438620567\n",
      "Epoch:7 Batch:480 Log-loss0.048633888363838196\n",
      "Epoch:7 Batch:560 Log-loss0.044384364038705826\n",
      "Epoch average log loss:0.044326877001939075\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.03993716765669266, best_val_loss:0.03993716765669266, best_auc:0.987167557522111\n",
      "Epoch:8 Batch:0 Log-loss0.048674505203962326\n",
      "Epoch:8 Batch:80 Log-loss0.04808160662651062\n",
      "Epoch:8 Batch:160 Log-loss0.03657810017466545\n",
      "Epoch:8 Batch:240 Log-loss0.05624471604824066\n",
      "Epoch:8 Batch:320 Log-loss0.04456556960940361\n",
      "Epoch:8 Batch:400 Log-loss0.040725138038396835\n",
      "Epoch:8 Batch:480 Log-loss0.03420569375157356\n",
      "Epoch:8 Batch:560 Log-loss0.04455525800585747\n",
      "Epoch average log loss:0.043418767659126646\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.03970318418561908, best_val_loss:0.03970318418561908, best_auc:0.9879785445663226\n",
      "Epoch:9 Batch:0 Log-loss0.04281042888760567\n",
      "Epoch:9 Batch:80 Log-loss0.059381917119026184\n",
      "Epoch:9 Batch:160 Log-loss0.05352580174803734\n",
      "Epoch:9 Batch:240 Log-loss0.03858066722750664\n",
      "Epoch:9 Batch:320 Log-loss0.04412071406841278\n",
      "Epoch:9 Batch:400 Log-loss0.05041211470961571\n",
      "Epoch:9 Batch:480 Log-loss0.04672423377633095\n",
      "Epoch:9 Batch:560 Log-loss0.043157126754522324\n",
      "Epoch average log loss:0.042870320013857315\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.0392150271716064, best_val_loss:0.0392150271716064, best_auc:0.988420055106625\n",
      "Epoch:10 Batch:0 Log-loss0.04093577340245247\n",
      "Epoch:10 Batch:80 Log-loss0.047770533710718155\n",
      "Epoch:10 Batch:160 Log-loss0.04096495360136032\n",
      "Epoch:10 Batch:240 Log-loss0.03698445484042168\n",
      "Epoch:10 Batch:320 Log-loss0.05337987840175629\n",
      "Epoch:10 Batch:400 Log-loss0.05798211321234703\n",
      "Epoch:10 Batch:480 Log-loss0.03534146770834923\n",
      "Epoch:10 Batch:560 Log-loss0.02640831656754017\n",
      "Epoch average log loss:0.042311128774391756\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.03885209105628278, best_val_loss:0.03885209105628278, best_auc:0.9889164090407719\n",
      "Epoch:11 Batch:0 Log-loss0.04893331229686737\n",
      "Epoch:11 Batch:80 Log-loss0.0330306775867939\n",
      "Epoch:11 Batch:160 Log-loss0.037183426320552826\n",
      "Epoch:11 Batch:240 Log-loss0.04365379735827446\n",
      "Epoch:11 Batch:320 Log-loss0.03602832183241844\n",
      "Epoch:11 Batch:400 Log-loss0.02902217209339142\n",
      "Epoch:11 Batch:480 Log-loss0.03815813362598419\n",
      "Epoch:11 Batch:560 Log-loss0.04418642818927765\n",
      "Epoch average log loss:0.041724646730082374\n",
      "In Epoch11, val_loss:0.03889859966947964, best_val_loss:0.03885209105628278, best_auc:0.9889164090407719\n",
      "Epoch:12 Batch:0 Log-loss0.041588883846998215\n",
      "Epoch:12 Batch:80 Log-loss0.04763919487595558\n",
      "Epoch:12 Batch:160 Log-loss0.03318735584616661\n",
      "Epoch:12 Batch:240 Log-loss0.031224163249135017\n",
      "Epoch:12 Batch:320 Log-loss0.042237550020217896\n",
      "Epoch:12 Batch:400 Log-loss0.03446788713335991\n",
      "Epoch:12 Batch:480 Log-loss0.03316266089677811\n",
      "Epoch:12 Batch:560 Log-loss0.05046306923031807\n",
      "Epoch average log loss:0.04140007988483246\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.0387229826131792, best_val_loss:0.0387229826131792, best_auc:0.9891776122791208\n",
      "Epoch:13 Batch:0 Log-loss0.042380522936582565\n",
      "Epoch:13 Batch:80 Log-loss0.03111424297094345\n",
      "Epoch:13 Batch:160 Log-loss0.03920604661107063\n",
      "Epoch:13 Batch:240 Log-loss0.032388947904109955\n",
      "Epoch:13 Batch:320 Log-loss0.03660188242793083\n",
      "Epoch:13 Batch:400 Log-loss0.02800227701663971\n",
      "Epoch:13 Batch:480 Log-loss0.058758292347192764\n",
      "Epoch:13 Batch:560 Log-loss0.048726681619882584\n",
      "Epoch average log loss:0.040936207315618436\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch13, val_loss:0.03854509312646472, best_val_loss:0.03854509312646472, best_auc:0.9892530298576115\n",
      "Epoch:14 Batch:0 Log-loss0.03628009557723999\n",
      "Epoch:14 Batch:80 Log-loss0.05006052926182747\n",
      "Epoch:14 Batch:160 Log-loss0.04204699024558067\n",
      "Epoch:14 Batch:240 Log-loss0.04563164710998535\n",
      "Epoch:14 Batch:320 Log-loss0.033515434712171555\n",
      "Epoch:14 Batch:400 Log-loss0.030722571536898613\n",
      "Epoch:14 Batch:480 Log-loss0.029991619288921356\n",
      "Epoch:14 Batch:560 Log-loss0.04155740886926651\n",
      "Epoch average log loss:0.040514373666207706\n",
      "In Epoch14, val_loss:0.03882964515765184, best_val_loss:0.03854509312646472, best_auc:0.9892530298576115\n",
      "Epoch:15 Batch:0 Log-loss0.05156722292304039\n",
      "Epoch:15 Batch:80 Log-loss0.03978018835186958\n",
      "Epoch:15 Batch:160 Log-loss0.03266591578722\n",
      "Epoch:15 Batch:240 Log-loss0.03702211007475853\n",
      "Epoch:15 Batch:320 Log-loss0.0476844422519207\n",
      "Epoch:15 Batch:400 Log-loss0.04555174335837364\n",
      "Epoch:15 Batch:480 Log-loss0.030027618631720543\n",
      "Epoch:15 Batch:560 Log-loss0.029943540692329407\n",
      "Epoch average log loss:0.04035602648509666\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch15, val_loss:0.03853926619529666, best_val_loss:0.03853926619529666, best_auc:0.989554179060542\n",
      "Epoch:16 Batch:0 Log-loss0.0377388559281826\n",
      "Epoch:16 Batch:80 Log-loss0.04508373513817787\n",
      "Epoch:16 Batch:160 Log-loss0.01979883201420307\n",
      "Epoch:16 Batch:240 Log-loss0.04104058817028999\n",
      "Epoch:16 Batch:320 Log-loss0.03983898460865021\n",
      "Epoch:16 Batch:400 Log-loss0.024622997269034386\n",
      "Epoch:16 Batch:480 Log-loss0.031952206045389175\n",
      "Epoch:16 Batch:560 Log-loss0.02808995544910431\n",
      "Epoch average log loss:0.039918629267984736\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch16, val_loss:0.037926237694796805, best_val_loss:0.037926237694796805, best_auc:0.9897850123345697\n",
      "Epoch:17 Batch:0 Log-loss0.035836536437273026\n",
      "Epoch:17 Batch:80 Log-loss0.03006804548203945\n",
      "Epoch:17 Batch:160 Log-loss0.034197743982076645\n",
      "Epoch:17 Batch:240 Log-loss0.05190204083919525\n",
      "Epoch:17 Batch:320 Log-loss0.04151451960206032\n",
      "Epoch:17 Batch:400 Log-loss0.044216934591531754\n",
      "Epoch:17 Batch:480 Log-loss0.03583996370434761\n",
      "Epoch:17 Batch:560 Log-loss0.029905136674642563\n",
      "Epoch average log loss:0.03962167512758502\n",
      "In Epoch17, val_loss:0.03803516231878809, best_val_loss:0.037926237694796805, best_auc:0.9897850123345697\n",
      "Epoch:18 Batch:0 Log-loss0.0294752586632967\n",
      "Epoch:18 Batch:80 Log-loss0.049178775399923325\n",
      "Epoch:18 Batch:160 Log-loss0.034290142357349396\n",
      "Epoch:18 Batch:240 Log-loss0.045609790831804276\n",
      "Epoch:18 Batch:320 Log-loss0.03634515032172203\n",
      "Epoch:18 Batch:400 Log-loss0.02404644899070263\n",
      "Epoch:18 Batch:480 Log-loss0.030334820970892906\n",
      "Epoch:18 Batch:560 Log-loss0.023693731054663658\n",
      "Epoch average log loss:0.039625546475872395\n",
      "In Epoch18, val_loss:0.03827605597945103, best_val_loss:0.037926237694796805, best_auc:0.9897850123345697\n",
      "Epoch:19 Batch:0 Log-loss0.02865556813776493\n",
      "Epoch:19 Batch:80 Log-loss0.02583337388932705\n",
      "Epoch:19 Batch:160 Log-loss0.039293963462114334\n",
      "Epoch:19 Batch:240 Log-loss0.05812177434563637\n",
      "Epoch:19 Batch:320 Log-loss0.03987984359264374\n",
      "Epoch:19 Batch:400 Log-loss0.04091031476855278\n",
      "Epoch:19 Batch:480 Log-loss0.048562388867139816\n",
      "Epoch:19 Batch:560 Log-loss0.0390067920088768\n",
      "Epoch average log loss:0.039475733863322864\n",
      "In Epoch19, val_loss:0.03797545348664402, best_val_loss:0.037926237694796805, best_auc:0.9897850123345697\n",
      "Epoch:20 Batch:0 Log-loss0.043886974453926086\n",
      "Epoch:20 Batch:80 Log-loss0.034971196204423904\n",
      "Epoch:20 Batch:160 Log-loss0.05002016946673393\n",
      "Epoch:20 Batch:240 Log-loss0.03417382016777992\n",
      "Epoch:20 Batch:320 Log-loss0.03687715157866478\n",
      "Epoch:20 Batch:400 Log-loss0.04106539860367775\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:20 Batch:480 Log-loss0.0327741764485836\n",
      "Epoch:20 Batch:560 Log-loss0.04017656296491623\n",
      "Epoch average log loss:0.0390448432748339\n",
      "In Epoch20, val_loss:0.03795222403640442, best_val_loss:0.037926237694796805, best_auc:0.9897850123345697\n",
      "Epoch:21 Batch:0 Log-loss0.03814211115241051\n",
      "Epoch:21 Batch:80 Log-loss0.035382095724344254\n",
      "Epoch:21 Batch:160 Log-loss0.033323440700769424\n",
      "Epoch:21 Batch:240 Log-loss0.052605804055929184\n",
      "Epoch:21 Batch:320 Log-loss0.03816227987408638\n",
      "Epoch:21 Batch:400 Log-loss0.039860185235738754\n",
      "Epoch:21 Batch:480 Log-loss0.044858891516923904\n",
      "Epoch:21 Batch:560 Log-loss0.042281363159418106\n",
      "Epoch average log loss:0.039070909121073784\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch21, val_loss:0.03791436877659677, best_val_loss:0.03791436877659677, best_auc:0.989891544404899\n",
      "Epoch:22 Batch:0 Log-loss0.029027186334133148\n",
      "Epoch:22 Batch:80 Log-loss0.044422123581171036\n",
      "Epoch:22 Batch:160 Log-loss0.035015057772397995\n",
      "Epoch:22 Batch:240 Log-loss0.029341675341129303\n",
      "Epoch:22 Batch:320 Log-loss0.023687215521931648\n",
      "Epoch:22 Batch:400 Log-loss0.035927191376686096\n",
      "Epoch:22 Batch:480 Log-loss0.031381119042634964\n",
      "Epoch:22 Batch:560 Log-loss0.04212402552366257\n",
      "Epoch average log loss:0.03883014609039362\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch22, val_loss:0.037838971801488326, best_val_loss:0.037838971801488326, best_auc:0.9900087019107738\n",
      "Epoch:23 Batch:0 Log-loss0.03561588376760483\n",
      "Epoch:23 Batch:80 Log-loss0.044989898800849915\n",
      "Epoch:23 Batch:160 Log-loss0.04137049987912178\n",
      "Epoch:23 Batch:240 Log-loss0.04377829656004906\n",
      "Epoch:23 Batch:320 Log-loss0.04857691004872322\n",
      "Epoch:23 Batch:400 Log-loss0.06454996019601822\n",
      "Epoch:23 Batch:480 Log-loss0.03363563492894173\n",
      "Epoch:23 Batch:560 Log-loss0.04339234530925751\n",
      "Epoch average log loss:0.03879192800959572\n",
      "In Epoch23, val_loss:0.03794880332696079, best_val_loss:0.037838971801488326, best_auc:0.9900087019107738\n",
      "Epoch:24 Batch:0 Log-loss0.04390394687652588\n",
      "Epoch:24 Batch:80 Log-loss0.039485130459070206\n",
      "Epoch:24 Batch:160 Log-loss0.03215256333351135\n",
      "Epoch:24 Batch:240 Log-loss0.04508448764681816\n",
      "Epoch:24 Batch:320 Log-loss0.03963356092572212\n",
      "Epoch:24 Batch:400 Log-loss0.03998386487364769\n",
      "Epoch:24 Batch:480 Log-loss0.05736441910266876\n",
      "Epoch:24 Batch:560 Log-loss0.028261898085474968\n",
      "Epoch average log loss:0.0385970508579963\n",
      "In Epoch24, val_loss:0.03833049964653127, best_val_loss:0.037838971801488326, best_auc:0.9900087019107738\n",
      "Epoch:25 Batch:0 Log-loss0.03222271800041199\n",
      "Epoch:25 Batch:80 Log-loss0.029416551813483238\n",
      "Epoch:25 Batch:160 Log-loss0.03772120550274849\n",
      "Epoch:25 Batch:240 Log-loss0.03262540325522423\n",
      "Epoch:25 Batch:320 Log-loss0.03714508190751076\n",
      "Epoch:25 Batch:400 Log-loss0.04438410699367523\n",
      "Epoch:25 Batch:480 Log-loss0.044289156794548035\n",
      "Epoch:25 Batch:560 Log-loss0.03410374000668526\n",
      "Epoch average log loss:0.03848097850235977\n",
      "In Epoch25, val_loss:0.03837571732360634, best_val_loss:0.037838971801488326, best_auc:0.9900087019107738\n",
      "Epoch:26 Batch:0 Log-loss0.028183212503790855\n",
      "Epoch:26 Batch:80 Log-loss0.04694196209311485\n",
      "Epoch:26 Batch:160 Log-loss0.03398086875677109\n",
      "Epoch:26 Batch:240 Log-loss0.033473074436187744\n",
      "Epoch:26 Batch:320 Log-loss0.03726779296994209\n",
      "Epoch:26 Batch:400 Log-loss0.04490518569946289\n",
      "Epoch:26 Batch:480 Log-loss0.029274532571434975\n",
      "Epoch:26 Batch:560 Log-loss0.03414791822433472\n",
      "Epoch average log loss:0.03853222902065941\n",
      "In Epoch26, val_loss:0.03812035662978031, best_val_loss:0.037838971801488326, best_auc:0.9900087019107738\n",
      "Epoch:27 Batch:0 Log-loss0.03618134185671806\n",
      "Epoch:27 Batch:80 Log-loss0.03333355113863945\n",
      "Epoch:27 Batch:160 Log-loss0.045656803995370865\n",
      "Epoch:27 Batch:240 Log-loss0.03271658346056938\n",
      "Epoch:27 Batch:320 Log-loss0.031244583427906036\n",
      "Epoch:27 Batch:400 Log-loss0.030342942103743553\n",
      "Epoch:27 Batch:480 Log-loss0.049202676862478256\n",
      "Epoch:27 Batch:560 Log-loss0.049247100949287415\n",
      "Epoch average log loss:0.03818309363975589\n",
      "In Epoch27, val_loss:0.037911712837134695, best_val_loss:0.037838971801488326, best_auc:0.9900087019107738\n",
      "Epoch:28 Batch:0 Log-loss0.031800154596567154\n",
      "Epoch:28 Batch:80 Log-loss0.049296196550130844\n",
      "Epoch:28 Batch:160 Log-loss0.04914633929729462\n",
      "Epoch:28 Batch:240 Log-loss0.03375256061553955\n",
      "Epoch:28 Batch:320 Log-loss0.03134628012776375\n",
      "Epoch:28 Batch:400 Log-loss0.04514099657535553\n",
      "Epoch:28 Batch:480 Log-loss0.04001479223370552\n",
      "Epoch:28 Batch:560 Log-loss0.024977216497063637\n",
      "Epoch average log loss:0.038336054062736886\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch28, val_loss:0.037684062535458564, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:29 Batch:0 Log-loss0.038278404623270035\n",
      "Epoch:29 Batch:80 Log-loss0.030598973855376244\n",
      "Epoch:29 Batch:160 Log-loss0.027029974386096\n",
      "Epoch:29 Batch:240 Log-loss0.037482112646102905\n",
      "Epoch:29 Batch:320 Log-loss0.03793314844369888\n",
      "Epoch:29 Batch:400 Log-loss0.03001730889081955\n",
      "Epoch:29 Batch:480 Log-loss0.04754193127155304\n",
      "Epoch:29 Batch:560 Log-loss0.02478867955505848\n",
      "Epoch average log loss:0.038380738836713134\n",
      "In Epoch29, val_loss:0.037808704493143996, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:30 Batch:0 Log-loss0.04429149255156517\n",
      "Epoch:30 Batch:80 Log-loss0.035295773297548294\n",
      "Epoch:30 Batch:160 Log-loss0.03731295093894005\n",
      "Epoch:30 Batch:240 Log-loss0.041191548109054565\n",
      "Epoch:30 Batch:320 Log-loss0.034897856414318085\n",
      "Epoch:30 Batch:400 Log-loss0.04089278355240822\n",
      "Epoch:30 Batch:480 Log-loss0.03554414585232735\n",
      "Epoch:30 Batch:560 Log-loss0.03829110786318779\n",
      "Epoch average log loss:0.03793391385115683\n",
      "In Epoch30, val_loss:0.038183572762569416, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:31 Batch:0 Log-loss0.04398608207702637\n",
      "Epoch:31 Batch:80 Log-loss0.03556432947516441\n",
      "Epoch:31 Batch:160 Log-loss0.03856196627020836\n",
      "Epoch:31 Batch:240 Log-loss0.03102649189531803\n",
      "Epoch:31 Batch:320 Log-loss0.04459908977150917\n",
      "Epoch:31 Batch:400 Log-loss0.05128135904669762\n",
      "Epoch:31 Batch:480 Log-loss0.027407215908169746\n",
      "Epoch:31 Batch:560 Log-loss0.05147640034556389\n",
      "Epoch average log loss:0.038046141532582366\n",
      "In Epoch31, val_loss:0.03809960418346748, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:32 Batch:0 Log-loss0.057574402540922165\n",
      "Epoch:32 Batch:80 Log-loss0.03685798868536949\n",
      "Epoch:32 Batch:160 Log-loss0.04228216037154198\n",
      "Epoch:32 Batch:240 Log-loss0.01983744092285633\n",
      "Epoch:32 Batch:320 Log-loss0.028725117444992065\n",
      "Epoch:32 Batch:400 Log-loss0.02298622764647007\n",
      "Epoch:32 Batch:480 Log-loss0.03129763901233673\n",
      "Epoch:32 Batch:560 Log-loss0.042253751307725906\n",
      "Epoch average log loss:0.03782864192035049\n",
      "In Epoch32, val_loss:0.037815799825412166, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:33 Batch:0 Log-loss0.02868231199681759\n",
      "Epoch:33 Batch:80 Log-loss0.03763527795672417\n",
      "Epoch:33 Batch:160 Log-loss0.029718682169914246\n",
      "Epoch:33 Batch:240 Log-loss0.03914104774594307\n",
      "Epoch:33 Batch:320 Log-loss0.032334789633750916\n",
      "Epoch:33 Batch:400 Log-loss0.03796723857522011\n",
      "Epoch:33 Batch:480 Log-loss0.025074152275919914\n",
      "Epoch:33 Batch:560 Log-loss0.04347536340355873\n",
      "Epoch average log loss:0.03787663196812251\n",
      "In Epoch33, val_loss:0.03777111869579219, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:34 Batch:0 Log-loss0.030847301706671715\n",
      "Epoch:34 Batch:80 Log-loss0.028376316651701927\n",
      "Epoch:34 Batch:160 Log-loss0.03417958691716194\n",
      "Epoch:34 Batch:240 Log-loss0.039474423974752426\n",
      "Epoch:34 Batch:320 Log-loss0.03646835684776306\n",
      "Epoch:34 Batch:400 Log-loss0.02595691569149494\n",
      "Epoch:34 Batch:480 Log-loss0.03538360819220543\n",
      "Epoch:34 Batch:560 Log-loss0.02248932048678398\n",
      "Epoch average log loss:0.03780751702641802\n",
      "In Epoch34, val_loss:0.03790298802036056, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:35 Batch:0 Log-loss0.03820085898041725\n",
      "Epoch:35 Batch:80 Log-loss0.03392740339040756\n",
      "Epoch:35 Batch:160 Log-loss0.038310982286930084\n",
      "Epoch:35 Batch:240 Log-loss0.04902249574661255\n",
      "Epoch:35 Batch:320 Log-loss0.03650972619652748\n",
      "Epoch:35 Batch:400 Log-loss0.02929239720106125\n",
      "Epoch:35 Batch:480 Log-loss0.045657236129045486\n",
      "Epoch:35 Batch:560 Log-loss0.029583532363176346\n",
      "Epoch average log loss:0.0377647508168593\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In Epoch35, val_loss:0.037855840185442184, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:36 Batch:0 Log-loss0.05497657135128975\n",
      "Epoch:36 Batch:80 Log-loss0.0437898151576519\n",
      "Epoch:36 Batch:160 Log-loss0.04222365841269493\n",
      "Epoch:36 Batch:240 Log-loss0.03186464682221413\n",
      "Epoch:36 Batch:320 Log-loss0.03034449927508831\n",
      "Epoch:36 Batch:400 Log-loss0.05314840003848076\n",
      "Epoch:36 Batch:480 Log-loss0.04296601936221123\n",
      "Epoch:36 Batch:560 Log-loss0.030323464423418045\n",
      "Epoch average log loss:0.037688776014173135\n",
      "In Epoch36, val_loss:0.03795074429561351, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:37 Batch:0 Log-loss0.039890535175800323\n",
      "Epoch:37 Batch:80 Log-loss0.03803515061736107\n",
      "Epoch:37 Batch:160 Log-loss0.05736413225531578\n",
      "Epoch:37 Batch:240 Log-loss0.03706048056483269\n",
      "Epoch:37 Batch:320 Log-loss0.03357957676053047\n",
      "Epoch:37 Batch:400 Log-loss0.03536348044872284\n",
      "Epoch:37 Batch:480 Log-loss0.028796495869755745\n",
      "Epoch:37 Batch:560 Log-loss0.03146214038133621\n",
      "Epoch average log loss:0.03767980065728937\n",
      "In Epoch37, val_loss:0.03795088591329178, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:38 Batch:0 Log-loss0.050962552428245544\n",
      "Epoch:38 Batch:80 Log-loss0.04892522096633911\n",
      "Epoch:38 Batch:160 Log-loss0.032531868666410446\n",
      "Epoch:38 Batch:240 Log-loss0.045958906412124634\n",
      "Epoch:38 Batch:320 Log-loss0.031176606193184853\n",
      "Epoch:38 Batch:400 Log-loss0.03744059428572655\n",
      "Epoch:38 Batch:480 Log-loss0.03482838347554207\n",
      "Epoch:38 Batch:560 Log-loss0.03208780288696289\n",
      "Epoch average log loss:0.03763357857907457\n",
      "In Epoch38, val_loss:0.03802950298768242, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:39 Batch:0 Log-loss0.034739717841148376\n",
      "Epoch:39 Batch:80 Log-loss0.032992009073495865\n",
      "Epoch:39 Batch:160 Log-loss0.04828329011797905\n",
      "Epoch:39 Batch:240 Log-loss0.034475747495889664\n",
      "Epoch:39 Batch:320 Log-loss0.040365975350141525\n",
      "Epoch:39 Batch:400 Log-loss0.035302210599184036\n",
      "Epoch:39 Batch:480 Log-loss0.016821591183543205\n",
      "Epoch:39 Batch:560 Log-loss0.02879239059984684\n",
      "Epoch average log loss:0.03765831888719861\n",
      "In Epoch39, val_loss:0.03811498398328588, best_val_loss:0.037684062535458564, best_auc:0.9900457880117605\n",
      "Epoch:40 Batch:0 Log-loss0.03673405945301056\n",
      "Epoch:40 Batch:80 Log-loss0.029152443632483482\n",
      "Epoch:40 Batch:160 Log-loss0.06002365052700043\n",
      "Epoch:40 Batch:240 Log-loss0.034372422844171524\n",
      "Epoch:40 Batch:320 Log-loss0.026936739683151245\n",
      "Epoch:40 Batch:400 Log-loss0.028684161603450775\n",
      "Epoch:40 Batch:480 Log-loss0.0333806611597538\n",
      "Epoch:40 Batch:560 Log-loss0.033163800835609436\n",
      "Epoch average log loss:0.03759916147350201\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_648.pt.\n",
      "\n",
      "Choose the torch base model.\n",
      "Training on fold 9\n",
      "Epoch:1 Batch:0 Log-loss0.677842378616333\n",
      "Epoch:1 Batch:80 Log-loss0.10463050752878189\n",
      "Epoch:1 Batch:160 Log-loss0.06665698438882828\n",
      "Epoch:1 Batch:240 Log-loss0.036137718707323074\n",
      "Epoch:1 Batch:320 Log-loss0.050518155097961426\n",
      "Epoch:1 Batch:400 Log-loss0.04709653556346893\n",
      "Epoch:1 Batch:480 Log-loss0.04949408397078514\n",
      "Epoch:1 Batch:560 Log-loss0.050293270498514175\n",
      "Epoch average log loss:0.0846026511929397\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.05050046755988124, best_val_loss:0.05050046755988124, best_auc:0.9704797300944502\n",
      "Epoch:2 Batch:0 Log-loss0.03768691420555115\n",
      "Epoch:2 Batch:80 Log-loss0.04929785057902336\n",
      "Epoch:2 Batch:160 Log-loss0.05624813213944435\n",
      "Epoch:2 Batch:240 Log-loss0.04570619389414787\n",
      "Epoch:2 Batch:320 Log-loss0.05198336020112038\n",
      "Epoch:2 Batch:400 Log-loss0.04710187390446663\n",
      "Epoch:2 Batch:480 Log-loss0.06595747917890549\n",
      "Epoch:2 Batch:560 Log-loss0.040626391768455505\n",
      "Epoch average log loss:0.05043269211892039\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.04733854105353801, best_val_loss:0.04733854105353801, best_auc:0.9751167611082653\n",
      "Epoch:3 Batch:0 Log-loss0.04617832601070404\n",
      "Epoch:3 Batch:80 Log-loss0.04586458578705788\n",
      "Epoch:3 Batch:160 Log-loss0.06089937686920166\n",
      "Epoch:3 Batch:240 Log-loss0.031198419630527496\n",
      "Epoch:3 Batch:320 Log-loss0.030395278707146645\n",
      "Epoch:3 Batch:400 Log-loss0.05435566604137421\n",
      "Epoch:3 Batch:480 Log-loss0.040426257997751236\n",
      "Epoch:3 Batch:560 Log-loss0.04796589910984039\n",
      "Epoch average log loss:0.04787249850840973\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.04574667574448956, best_val_loss:0.04574667574448956, best_auc:0.9794291812575388\n",
      "Epoch:4 Batch:0 Log-loss0.035688627511262894\n",
      "Epoch:4 Batch:80 Log-loss0.05499463155865669\n",
      "Epoch:4 Batch:160 Log-loss0.05692141875624657\n",
      "Epoch:4 Batch:240 Log-loss0.04162402078509331\n",
      "Epoch:4 Batch:320 Log-loss0.03366045281291008\n",
      "Epoch:4 Batch:400 Log-loss0.045593079179525375\n",
      "Epoch:4 Batch:480 Log-loss0.062013354152441025\n",
      "Epoch:4 Batch:560 Log-loss0.034138210117816925\n",
      "Epoch average log loss:0.04598540395631322\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.04470304219172213, best_val_loss:0.04470304219172213, best_auc:0.98199637301135\n",
      "Epoch:5 Batch:0 Log-loss0.0439358614385128\n",
      "Epoch:5 Batch:80 Log-loss0.047482218593358994\n",
      "Epoch:5 Batch:160 Log-loss0.03164978697896004\n",
      "Epoch:5 Batch:240 Log-loss0.037620145827531815\n",
      "Epoch:5 Batch:320 Log-loss0.039946042001247406\n",
      "Epoch:5 Batch:400 Log-loss0.039409834891557693\n",
      "Epoch:5 Batch:480 Log-loss0.04011547192931175\n",
      "Epoch:5 Batch:560 Log-loss0.054191090166568756\n",
      "Epoch average log loss:0.04479344860384507\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.04321622184628834, best_val_loss:0.04321622184628834, best_auc:0.9844011638228521\n",
      "Epoch:6 Batch:0 Log-loss0.03950166329741478\n",
      "Epoch:6 Batch:80 Log-loss0.04399868845939636\n",
      "Epoch:6 Batch:160 Log-loss0.05524519085884094\n",
      "Epoch:6 Batch:240 Log-loss0.04428151622414589\n",
      "Epoch:6 Batch:320 Log-loss0.052233487367630005\n",
      "Epoch:6 Batch:400 Log-loss0.044686947017908096\n",
      "Epoch:6 Batch:480 Log-loss0.055859342217445374\n",
      "Epoch:6 Batch:560 Log-loss0.054939355701208115\n",
      "Epoch average log loss:0.04352427988446184\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch6, val_loss:0.04251425133231091, best_val_loss:0.04251425133231091, best_auc:0.9859662827370692\n",
      "Epoch:7 Batch:0 Log-loss0.0390978567302227\n",
      "Epoch:7 Batch:80 Log-loss0.03588594123721123\n",
      "Epoch:7 Batch:160 Log-loss0.04538007080554962\n",
      "Epoch:7 Batch:240 Log-loss0.04848439618945122\n",
      "Epoch:7 Batch:320 Log-loss0.035526685416698456\n",
      "Epoch:7 Batch:400 Log-loss0.03587348386645317\n",
      "Epoch:7 Batch:480 Log-loss0.04119996353983879\n",
      "Epoch:7 Batch:560 Log-loss0.044025253504514694\n",
      "Epoch average log loss:0.04265120281239173\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch7, val_loss:0.04201561460861228, best_val_loss:0.04201561460861228, best_auc:0.9866843184357164\n",
      "Epoch:8 Batch:0 Log-loss0.034994207322597504\n",
      "Epoch:8 Batch:80 Log-loss0.04288500174880028\n",
      "Epoch:8 Batch:160 Log-loss0.03957794979214668\n",
      "Epoch:8 Batch:240 Log-loss0.05680279806256294\n",
      "Epoch:8 Batch:320 Log-loss0.039460502564907074\n",
      "Epoch:8 Batch:400 Log-loss0.059524547308683395\n",
      "Epoch:8 Batch:480 Log-loss0.04069479927420616\n",
      "Epoch:8 Batch:560 Log-loss0.05137953907251358\n",
      "Epoch average log loss:0.04223212354284312\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.04198266648195665, best_val_loss:0.04198266648195665, best_auc:0.9871624857553303\n",
      "Epoch:9 Batch:0 Log-loss0.03291865065693855\n",
      "Epoch:9 Batch:80 Log-loss0.031417764723300934\n",
      "Epoch:9 Batch:160 Log-loss0.03735470399260521\n",
      "Epoch:9 Batch:240 Log-loss0.028115877881646156\n",
      "Epoch:9 Batch:320 Log-loss0.034559790045022964\n",
      "Epoch:9 Batch:400 Log-loss0.03795201703906059\n",
      "Epoch:9 Batch:480 Log-loss0.045016709715127945\n",
      "Epoch:9 Batch:560 Log-loss0.049198102205991745\n",
      "Epoch average log loss:0.04180619321164808\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.04171546749117249, best_val_loss:0.04171546749117249, best_auc:0.9872679490193325\n",
      "Epoch:10 Batch:0 Log-loss0.04483380541205406\n",
      "Epoch:10 Batch:80 Log-loss0.04528745636343956\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:10 Batch:160 Log-loss0.04282389581203461\n",
      "Epoch:10 Batch:240 Log-loss0.031121863052248955\n",
      "Epoch:10 Batch:320 Log-loss0.05522811785340309\n",
      "Epoch:10 Batch:400 Log-loss0.041295621544122696\n",
      "Epoch:10 Batch:480 Log-loss0.037460751831531525\n",
      "Epoch:10 Batch:560 Log-loss0.04177011176943779\n",
      "Epoch average log loss:0.041290296713954636\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.04146477673379654, best_val_loss:0.04146477673379654, best_auc:0.9879854702331209\n",
      "Epoch:11 Batch:0 Log-loss0.03828712925314903\n",
      "Epoch:11 Batch:80 Log-loss0.029279427602887154\n",
      "Epoch:11 Batch:160 Log-loss0.03931937366724014\n",
      "Epoch:11 Batch:240 Log-loss0.03444394841790199\n",
      "Epoch:11 Batch:320 Log-loss0.051808979362249374\n",
      "Epoch:11 Batch:400 Log-loss0.040487103164196014\n",
      "Epoch:11 Batch:480 Log-loss0.05333732068538666\n",
      "Epoch:11 Batch:560 Log-loss0.038853392004966736\n",
      "Epoch average log loss:0.040830884674297914\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch11, val_loss:0.041113394580980324, best_val_loss:0.041113394580980324, best_auc:0.9882959312316434\n",
      "Epoch:12 Batch:0 Log-loss0.03298095613718033\n",
      "Epoch:12 Batch:80 Log-loss0.04779454693198204\n",
      "Epoch:12 Batch:160 Log-loss0.03428031504154205\n",
      "Epoch:12 Batch:240 Log-loss0.04461465775966644\n",
      "Epoch:12 Batch:320 Log-loss0.045886341482400894\n",
      "Epoch:12 Batch:400 Log-loss0.04814957454800606\n",
      "Epoch:12 Batch:480 Log-loss0.024853602051734924\n",
      "Epoch:12 Batch:560 Log-loss0.037828229367733\n",
      "Epoch average log loss:0.040387955660532625\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.04068602046914995, best_val_loss:0.04068602046914995, best_auc:0.9883029301468982\n",
      "Epoch:13 Batch:0 Log-loss0.04549048840999603\n",
      "Epoch:13 Batch:80 Log-loss0.061070334166288376\n",
      "Epoch:13 Batch:160 Log-loss0.0440683513879776\n",
      "Epoch:13 Batch:240 Log-loss0.04196396470069885\n",
      "Epoch:13 Batch:320 Log-loss0.0368337519466877\n",
      "Epoch:13 Batch:400 Log-loss0.04243134334683418\n",
      "Epoch:13 Batch:480 Log-loss0.04878076910972595\n",
      "Epoch:13 Batch:560 Log-loss0.04423193633556366\n",
      "Epoch average log loss:0.04005203565050449\n",
      "In Epoch13, val_loss:0.04096089973601139, best_val_loss:0.04068602046914995, best_auc:0.9883029301468982\n",
      "Epoch:14 Batch:0 Log-loss0.041499022394418716\n",
      "Epoch:14 Batch:80 Log-loss0.02659977227449417\n",
      "Epoch:14 Batch:160 Log-loss0.04754476621747017\n",
      "Epoch:14 Batch:240 Log-loss0.06462626159191132\n",
      "Epoch:14 Batch:320 Log-loss0.03150920197367668\n",
      "Epoch:14 Batch:400 Log-loss0.03545435145497322\n",
      "Epoch:14 Batch:480 Log-loss0.04616795852780342\n",
      "Epoch:14 Batch:560 Log-loss0.06324633210897446\n",
      "Epoch average log loss:0.03983658616364535\n",
      "In Epoch14, val_loss:0.04203270783001178, best_val_loss:0.04068602046914995, best_auc:0.9883029301468982\n",
      "Epoch:15 Batch:0 Log-loss0.03180462867021561\n",
      "Epoch:15 Batch:80 Log-loss0.031168825924396515\n",
      "Epoch:15 Batch:160 Log-loss0.041617993265390396\n",
      "Epoch:15 Batch:240 Log-loss0.03931914642453194\n",
      "Epoch:15 Batch:320 Log-loss0.0326165072619915\n",
      "Epoch:15 Batch:400 Log-loss0.03732627257704735\n",
      "Epoch:15 Batch:480 Log-loss0.04873822629451752\n",
      "Epoch:15 Batch:560 Log-loss0.03022729605436325\n",
      "Epoch average log loss:0.03964445391403777\n",
      "In Epoch15, val_loss:0.041116889416659794, best_val_loss:0.04068602046914995, best_auc:0.9883029301468982\n",
      "Epoch:16 Batch:0 Log-loss0.037607695907354355\n",
      "Epoch:16 Batch:80 Log-loss0.04284541308879852\n",
      "Epoch:16 Batch:160 Log-loss0.04340958595275879\n",
      "Epoch:16 Batch:240 Log-loss0.03303557634353638\n",
      "Epoch:16 Batch:320 Log-loss0.04930888116359711\n",
      "Epoch:16 Batch:400 Log-loss0.03612102195620537\n",
      "Epoch:16 Batch:480 Log-loss0.04560789093375206\n",
      "Epoch:16 Batch:560 Log-loss0.042768366634845734\n",
      "Epoch average log loss:0.039097649580799046\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch16, val_loss:0.04062620204551958, best_val_loss:0.04062620204551958, best_auc:0.9888732738223941\n",
      "Epoch:17 Batch:0 Log-loss0.048306893557310104\n",
      "Epoch:17 Batch:80 Log-loss0.040027473121881485\n",
      "Epoch:17 Batch:160 Log-loss0.05121392384171486\n",
      "Epoch:17 Batch:240 Log-loss0.04024956375360489\n",
      "Epoch:17 Batch:320 Log-loss0.04062032699584961\n",
      "Epoch:17 Batch:400 Log-loss0.05939185246825218\n",
      "Epoch:17 Batch:480 Log-loss0.05549873039126396\n",
      "Epoch:17 Batch:560 Log-loss0.03156878799200058\n",
      "Epoch average log loss:0.03907184329083455\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch17, val_loss:0.040624047015757316, best_val_loss:0.040624047015757316, best_auc:0.9889294667648053\n",
      "Epoch:18 Batch:0 Log-loss0.03704702854156494\n",
      "Epoch:18 Batch:80 Log-loss0.03917061537504196\n",
      "Epoch:18 Batch:160 Log-loss0.046215638518333435\n",
      "Epoch:18 Batch:240 Log-loss0.04163176193833351\n",
      "Epoch:18 Batch:320 Log-loss0.029996195808053017\n",
      "Epoch:18 Batch:400 Log-loss0.028225988149642944\n",
      "Epoch:18 Batch:480 Log-loss0.03526903688907623\n",
      "Epoch:18 Batch:560 Log-loss0.050019506365060806\n",
      "Epoch average log loss:0.038984606215464214\n",
      "In Epoch18, val_loss:0.04067097444478796, best_val_loss:0.040624047015757316, best_auc:0.9889294667648053\n",
      "Epoch:19 Batch:0 Log-loss0.03436163067817688\n",
      "Epoch:19 Batch:80 Log-loss0.038926202803850174\n",
      "Epoch:19 Batch:160 Log-loss0.04160001501441002\n",
      "Epoch:19 Batch:240 Log-loss0.028287410736083984\n",
      "Epoch:19 Batch:320 Log-loss0.028919709846377373\n",
      "Epoch:19 Batch:400 Log-loss0.036819636821746826\n",
      "Epoch:19 Batch:480 Log-loss0.044794876128435135\n",
      "Epoch:19 Batch:560 Log-loss0.04085922613739967\n",
      "Epoch average log loss:0.038624534230413184\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch19, val_loss:0.04059662514269435, best_val_loss:0.04059662514269435, best_auc:0.9891532284187686\n",
      "Epoch:20 Batch:0 Log-loss0.03120158612728119\n",
      "Epoch:20 Batch:80 Log-loss0.04360171779990196\n",
      "Epoch:20 Batch:160 Log-loss0.06012613698840141\n",
      "Epoch:20 Batch:240 Log-loss0.03150474280118942\n",
      "Epoch:20 Batch:320 Log-loss0.041856423020362854\n",
      "Epoch:20 Batch:400 Log-loss0.049295488744974136\n",
      "Epoch:20 Batch:480 Log-loss0.032550521194934845\n",
      "Epoch:20 Batch:560 Log-loss0.043380554765462875\n",
      "Epoch average log loss:0.03851215581276587\n",
      "In Epoch20, val_loss:0.04064667009147032, best_val_loss:0.04059662514269435, best_auc:0.9891532284187686\n",
      "Epoch:21 Batch:0 Log-loss0.026827014982700348\n",
      "Epoch:21 Batch:80 Log-loss0.030700452625751495\n",
      "Epoch:21 Batch:160 Log-loss0.04818791523575783\n",
      "Epoch:21 Batch:240 Log-loss0.05430862307548523\n",
      "Epoch:21 Batch:320 Log-loss0.0394720584154129\n",
      "Epoch:21 Batch:400 Log-loss0.052398115396499634\n",
      "Epoch:21 Batch:480 Log-loss0.03181442990899086\n",
      "Epoch:21 Batch:560 Log-loss0.019327200949192047\n",
      "Epoch average log loss:0.03838796315747978\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch21, val_loss:0.040314061288075485, best_val_loss:0.040314061288075485, best_auc:0.9892347259714313\n",
      "Epoch:22 Batch:0 Log-loss0.03438674286007881\n",
      "Epoch:22 Batch:80 Log-loss0.041518643498420715\n",
      "Epoch:22 Batch:160 Log-loss0.04716656729578972\n",
      "Epoch:22 Batch:240 Log-loss0.027767354622483253\n",
      "Epoch:22 Batch:320 Log-loss0.03292180970311165\n",
      "Epoch:22 Batch:400 Log-loss0.04605164751410484\n",
      "Epoch:22 Batch:480 Log-loss0.03701641038060188\n",
      "Epoch:22 Batch:560 Log-loss0.04248886555433273\n",
      "Epoch average log loss:0.0382412878463843\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch22, val_loss:0.04020119063752534, best_val_loss:0.04020119063752534, best_auc:0.9893631342856141\n",
      "Epoch:23 Batch:0 Log-loss0.03374942019581795\n",
      "Epoch:23 Batch:80 Log-loss0.02965128980576992\n",
      "Epoch:23 Batch:160 Log-loss0.0343044139444828\n",
      "Epoch:23 Batch:240 Log-loss0.04712975397706032\n",
      "Epoch:23 Batch:320 Log-loss0.03056827373802662\n",
      "Epoch:23 Batch:400 Log-loss0.0331425704061985\n",
      "Epoch:23 Batch:480 Log-loss0.025367185473442078\n",
      "Epoch:23 Batch:560 Log-loss0.0335024893283844\n",
      "Epoch average log loss:0.03806589286375259\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch23, val_loss:0.04019019502415289, best_val_loss:0.04019019502415289, best_auc:0.9893675401559041\n",
      "Epoch:24 Batch:0 Log-loss0.03779659792780876\n",
      "Epoch:24 Batch:80 Log-loss0.03704853728413582\n",
      "Epoch:24 Batch:160 Log-loss0.025128602981567383\n",
      "Epoch:24 Batch:240 Log-loss0.03932122513651848\n",
      "Epoch:24 Batch:320 Log-loss0.03361743316054344\n",
      "Epoch:24 Batch:400 Log-loss0.058460552245378494\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:24 Batch:480 Log-loss0.0466466061770916\n",
      "Epoch:24 Batch:560 Log-loss0.03535909950733185\n",
      "Epoch average log loss:0.038148098016556886\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch24, val_loss:0.04016188253774589, best_val_loss:0.04016188253774589, best_auc:0.9893363387932773\n",
      "Epoch:25 Batch:0 Log-loss0.03850245103240013\n",
      "Epoch:25 Batch:80 Log-loss0.04465041682124138\n",
      "Epoch:25 Batch:160 Log-loss0.040072001516819\n",
      "Epoch:25 Batch:240 Log-loss0.028470242395997047\n",
      "Epoch:25 Batch:320 Log-loss0.04116030037403107\n",
      "Epoch:25 Batch:400 Log-loss0.026606528088450432\n",
      "Epoch:25 Batch:480 Log-loss0.038054484874010086\n",
      "Epoch:25 Batch:560 Log-loss0.0354715920984745\n",
      "Epoch average log loss:0.037899431631168615\n",
      "In Epoch25, val_loss:0.040314034142484294, best_val_loss:0.04016188253774589, best_auc:0.9893363387932773\n",
      "Epoch:26 Batch:0 Log-loss0.032303087413311005\n",
      "Epoch:26 Batch:80 Log-loss0.026805786415934563\n",
      "Epoch:26 Batch:160 Log-loss0.04925544187426567\n",
      "Epoch:26 Batch:240 Log-loss0.04257825389504433\n",
      "Epoch:26 Batch:320 Log-loss0.024929041042923927\n",
      "Epoch:26 Batch:400 Log-loss0.03365882858633995\n",
      "Epoch:26 Batch:480 Log-loss0.03382832556962967\n",
      "Epoch:26 Batch:560 Log-loss0.036977317184209824\n",
      "Epoch average log loss:0.037855572928674516\n",
      "In Epoch26, val_loss:0.040222173655248314, best_val_loss:0.04016188253774589, best_auc:0.9893363387932773\n",
      "Epoch:27 Batch:0 Log-loss0.028725137934088707\n",
      "Epoch:27 Batch:80 Log-loss0.030498052015900612\n",
      "Epoch:27 Batch:160 Log-loss0.032362136989831924\n",
      "Epoch:27 Batch:240 Log-loss0.0421285443007946\n",
      "Epoch:27 Batch:320 Log-loss0.04638294875621796\n",
      "Epoch:27 Batch:400 Log-loss0.032871443778276443\n",
      "Epoch:27 Batch:480 Log-loss0.03495532646775246\n",
      "Epoch:27 Batch:560 Log-loss0.021395433694124222\n",
      "Epoch average log loss:0.03774579140923119\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch27, val_loss:0.040059509779218135, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:28 Batch:0 Log-loss0.03224855288863182\n",
      "Epoch:28 Batch:80 Log-loss0.04748544096946716\n",
      "Epoch:28 Batch:160 Log-loss0.033813830465078354\n",
      "Epoch:28 Batch:240 Log-loss0.02522752247750759\n",
      "Epoch:28 Batch:320 Log-loss0.025120215490460396\n",
      "Epoch:28 Batch:400 Log-loss0.03381581977009773\n",
      "Epoch:28 Batch:480 Log-loss0.04303066432476044\n",
      "Epoch:28 Batch:560 Log-loss0.03830692544579506\n",
      "Epoch average log loss:0.037470285617746416\n",
      "In Epoch28, val_loss:0.040317107672679764, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:29 Batch:0 Log-loss0.04457904398441315\n",
      "Epoch:29 Batch:80 Log-loss0.037761494517326355\n",
      "Epoch:29 Batch:160 Log-loss0.036852601915597916\n",
      "Epoch:29 Batch:240 Log-loss0.030910132452845573\n",
      "Epoch:29 Batch:320 Log-loss0.045016463845968246\n",
      "Epoch:29 Batch:400 Log-loss0.03401254490017891\n",
      "Epoch:29 Batch:480 Log-loss0.03870880603790283\n",
      "Epoch:29 Batch:560 Log-loss0.04042939469218254\n",
      "Epoch average log loss:0.03787387304141053\n",
      "In Epoch29, val_loss:0.04036425632459549, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:30 Batch:0 Log-loss0.04566258192062378\n",
      "Epoch:30 Batch:80 Log-loss0.03213599696755409\n",
      "Epoch:30 Batch:160 Log-loss0.04645654559135437\n",
      "Epoch:30 Batch:240 Log-loss0.040809765458106995\n",
      "Epoch:30 Batch:320 Log-loss0.030501386150717735\n",
      "Epoch:30 Batch:400 Log-loss0.049807921051979065\n",
      "Epoch:30 Batch:480 Log-loss0.023104824125766754\n",
      "Epoch:30 Batch:560 Log-loss0.034759167581796646\n",
      "Epoch average log loss:0.03747424088485007\n",
      "In Epoch30, val_loss:0.04012773208210661, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:31 Batch:0 Log-loss0.06812845915555954\n",
      "Epoch:31 Batch:80 Log-loss0.029928026720881462\n",
      "Epoch:31 Batch:160 Log-loss0.02777133323252201\n",
      "Epoch:31 Batch:240 Log-loss0.04947547987103462\n",
      "Epoch:31 Batch:320 Log-loss0.056027039885520935\n",
      "Epoch:31 Batch:400 Log-loss0.03515635058283806\n",
      "Epoch:31 Batch:480 Log-loss0.04605010151863098\n",
      "Epoch:31 Batch:560 Log-loss0.03620058670639992\n",
      "Epoch average log loss:0.037382282170334034\n",
      "In Epoch31, val_loss:0.04035834252351144, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:32 Batch:0 Log-loss0.04790396988391876\n",
      "Epoch:32 Batch:80 Log-loss0.04459311440587044\n",
      "Epoch:32 Batch:160 Log-loss0.044605184346437454\n",
      "Epoch:32 Batch:240 Log-loss0.04315023496747017\n",
      "Epoch:32 Batch:320 Log-loss0.0494711734354496\n",
      "Epoch:32 Batch:400 Log-loss0.020853325724601746\n",
      "Epoch:32 Batch:480 Log-loss0.028328480198979378\n",
      "Epoch:32 Batch:560 Log-loss0.04013381525874138\n",
      "Epoch average log loss:0.0373666464118287\n",
      "In Epoch32, val_loss:0.040157640690749896, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:33 Batch:0 Log-loss0.024319810792803764\n",
      "Epoch:33 Batch:80 Log-loss0.035227034240961075\n",
      "Epoch:33 Batch:160 Log-loss0.04236409068107605\n",
      "Epoch:33 Batch:240 Log-loss0.043839577585458755\n",
      "Epoch:33 Batch:320 Log-loss0.04304429516196251\n",
      "Epoch:33 Batch:400 Log-loss0.04086269810795784\n",
      "Epoch:33 Batch:480 Log-loss0.04603627324104309\n",
      "Epoch:33 Batch:560 Log-loss0.034914691001176834\n",
      "Epoch average log loss:0.037375700413914664\n",
      "In Epoch33, val_loss:0.04009883668987541, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:34 Batch:0 Log-loss0.029639681801199913\n",
      "Epoch:34 Batch:80 Log-loss0.04470926895737648\n",
      "Epoch:34 Batch:160 Log-loss0.03599005565047264\n",
      "Epoch:34 Batch:240 Log-loss0.03760843351483345\n",
      "Epoch:34 Batch:320 Log-loss0.04404360055923462\n",
      "Epoch:34 Batch:400 Log-loss0.0570288710296154\n",
      "Epoch:34 Batch:480 Log-loss0.030827047303318977\n",
      "Epoch:34 Batch:560 Log-loss0.04792068898677826\n",
      "Epoch average log loss:0.03727063886555178\n",
      "In Epoch34, val_loss:0.04046639528229998, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:35 Batch:0 Log-loss0.030992796644568443\n",
      "Epoch:35 Batch:80 Log-loss0.05202135071158409\n",
      "Epoch:35 Batch:160 Log-loss0.039847858250141144\n",
      "Epoch:35 Batch:240 Log-loss0.037098873406648636\n",
      "Epoch:35 Batch:320 Log-loss0.02730836160480976\n",
      "Epoch:35 Batch:400 Log-loss0.032526250928640366\n",
      "Epoch:35 Batch:480 Log-loss0.033964984118938446\n",
      "Epoch:35 Batch:560 Log-loss0.03669912740588188\n",
      "Epoch average log loss:0.03731412684345352\n",
      "In Epoch35, val_loss:0.04021983574864756, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:36 Batch:0 Log-loss0.03639639541506767\n",
      "Epoch:36 Batch:80 Log-loss0.0372031144797802\n",
      "Epoch:36 Batch:160 Log-loss0.037919532507658005\n",
      "Epoch:36 Batch:240 Log-loss0.025476373732089996\n",
      "Epoch:36 Batch:320 Log-loss0.0378197617828846\n",
      "Epoch:36 Batch:400 Log-loss0.041603781282901764\n",
      "Epoch:36 Batch:480 Log-loss0.04603814706206322\n",
      "Epoch:36 Batch:560 Log-loss0.04141257330775261\n",
      "Epoch average log loss:0.03718172173414912\n",
      "In Epoch36, val_loss:0.04019848111272338, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:37 Batch:0 Log-loss0.03388819471001625\n",
      "Epoch:37 Batch:80 Log-loss0.03116721846163273\n",
      "Epoch:37 Batch:160 Log-loss0.058629363775253296\n",
      "Epoch:37 Batch:240 Log-loss0.03390798345208168\n",
      "Epoch:37 Batch:320 Log-loss0.029790634289383888\n",
      "Epoch:37 Batch:400 Log-loss0.037243179976940155\n",
      "Epoch:37 Batch:480 Log-loss0.043471697717905045\n",
      "Epoch:37 Batch:560 Log-loss0.03585444763302803\n",
      "Epoch average log loss:0.03712967790862812\n",
      "In Epoch37, val_loss:0.04018255876383194, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:38 Batch:0 Log-loss0.047659631818532944\n",
      "Epoch:38 Batch:80 Log-loss0.034056175500154495\n",
      "Epoch:38 Batch:160 Log-loss0.03514738753437996\n",
      "Epoch:38 Batch:240 Log-loss0.03738059848546982\n",
      "Epoch:38 Batch:320 Log-loss0.06724338233470917\n",
      "Epoch:38 Batch:400 Log-loss0.02384502999484539\n",
      "Epoch:38 Batch:480 Log-loss0.04764113202691078\n",
      "Epoch:38 Batch:560 Log-loss0.04870041459798813\n",
      "Epoch average log loss:0.03719705000652799\n",
      "In Epoch38, val_loss:0.04026591315516767, best_val_loss:0.040059509779218135, best_auc:0.9895025935749565\n",
      "Epoch:39 Batch:0 Log-loss0.04129515215754509\n",
      "Epoch:39 Batch:80 Log-loss0.03329072147607803\n",
      "Epoch:39 Batch:160 Log-loss0.0340852253139019\n",
      "Epoch:39 Batch:240 Log-loss0.03768661245703697\n",
      "Epoch:39 Batch:320 Log-loss0.03800671920180321\n",
      "Epoch:39 Batch:400 Log-loss0.043510552495718\n",
      "Epoch:39 Batch:480 Log-loss0.028599560260772705\n",
      "Epoch:39 Batch:560 Log-loss0.032464414834976196\n",
      "Epoch average log loss:0.03699881270793932\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_649.pt.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "models, best_logloss, best_auc, best_val_pred = trainer.train_folds(X=train_sequences, y=training_labels,\n",
    "                    fold_count=10, batch_size=256, get_model_func=get_recurrent_higway_classifier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training on fold 0\n",
      "Epoch:1 Batch:0 Log-loss0.036531466990709305\n",
      "Epoch:1 Batch:80 Log-loss0.03582215681672096\n",
      "Epoch:1 Batch:160 Log-loss0.029018335044384003\n",
      "Epoch:1 Batch:240 Log-loss0.036196328699588776\n",
      "Epoch:1 Batch:320 Log-loss0.03609498217701912\n",
      "Epoch:1 Batch:400 Log-loss0.0301118865609169\n",
      "Epoch:1 Batch:480 Log-loss0.028371067717671394\n",
      "Epoch:1 Batch:560 Log-loss0.02657708339393139\n",
      "Epoch average log loss:0.03519969365359949\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.037408311027521235, best_val_loss:0.037408311027521235, best_auc:0.9905914533781136\n",
      "Epoch:2 Batch:0 Log-loss0.03476415202021599\n",
      "Epoch:2 Batch:80 Log-loss0.0448206402361393\n",
      "Epoch:2 Batch:160 Log-loss0.03184705600142479\n",
      "Epoch:2 Batch:240 Log-loss0.02453266829252243\n",
      "Epoch:2 Batch:320 Log-loss0.031927693635225296\n",
      "Epoch:2 Batch:400 Log-loss0.04199408367276192\n",
      "Epoch:2 Batch:480 Log-loss0.031146720051765442\n",
      "Epoch:2 Batch:560 Log-loss0.038211703300476074\n",
      "Epoch average log loss:0.03862287895088749\n",
      "In Epoch2, val_loss:0.037597328305081654, best_val_loss:0.037408311027521235, best_auc:0.9905914533781136\n",
      "Epoch:3 Batch:0 Log-loss0.03908723220229149\n",
      "Epoch:3 Batch:80 Log-loss0.044257745146751404\n",
      "Epoch:3 Batch:160 Log-loss0.02811143361032009\n",
      "Epoch:3 Batch:240 Log-loss0.03722819313406944\n",
      "Epoch:3 Batch:320 Log-loss0.03752940148115158\n",
      "Epoch:3 Batch:400 Log-loss0.03594081476330757\n",
      "Epoch:3 Batch:480 Log-loss0.037055276334285736\n",
      "Epoch:3 Batch:560 Log-loss0.02806156687438488\n",
      "Epoch average log loss:0.03819808587764523\n",
      "In Epoch3, val_loss:0.037706147913174295, best_val_loss:0.037408311027521235, best_auc:0.9905914533781136\n",
      "Epoch:4 Batch:0 Log-loss0.04407801106572151\n",
      "Epoch:4 Batch:80 Log-loss0.040825869888067245\n",
      "Epoch:4 Batch:160 Log-loss0.04536430537700653\n",
      "Epoch:4 Batch:240 Log-loss0.058482781052589417\n",
      "Epoch:4 Batch:320 Log-loss0.029999515041708946\n",
      "Epoch:4 Batch:400 Log-loss0.058971088379621506\n",
      "Epoch:4 Batch:480 Log-loss0.037144385278224945\n",
      "Epoch:4 Batch:560 Log-loss0.038049206137657166\n",
      "Epoch average log loss:0.03801946338000042\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.03731478001552681, best_val_loss:0.03731478001552681, best_auc:0.9905087098300337\n",
      "Epoch:5 Batch:0 Log-loss0.03366975858807564\n",
      "Epoch:5 Batch:80 Log-loss0.03441440314054489\n",
      "Epoch:5 Batch:160 Log-loss0.02591569535434246\n",
      "Epoch:5 Batch:240 Log-loss0.05303076282143593\n",
      "Epoch:5 Batch:320 Log-loss0.048461124300956726\n",
      "Epoch:5 Batch:400 Log-loss0.03361789882183075\n",
      "Epoch:5 Batch:480 Log-loss0.03201339766383171\n",
      "Epoch:5 Batch:560 Log-loss0.03445018455386162\n",
      "Epoch average log loss:0.03759857211927218\n",
      "In Epoch5, val_loss:0.03739726080901528, best_val_loss:0.03731478001552681, best_auc:0.9905087098300337\n",
      "Epoch:6 Batch:0 Log-loss0.04348793998360634\n",
      "Epoch:6 Batch:80 Log-loss0.04504086449742317\n",
      "Epoch:6 Batch:160 Log-loss0.047153592109680176\n",
      "Epoch:6 Batch:240 Log-loss0.030870608985424042\n",
      "Epoch:6 Batch:320 Log-loss0.02984582632780075\n",
      "Epoch:6 Batch:400 Log-loss0.045341748744249344\n",
      "Epoch:6 Batch:480 Log-loss0.05324370041489601\n",
      "Epoch:6 Batch:560 Log-loss0.041119031608104706\n",
      "Epoch average log loss:0.03722655142524413\n",
      "In Epoch6, val_loss:0.037641762485223405, best_val_loss:0.03731478001552681, best_auc:0.9905087098300337\n",
      "Epoch:7 Batch:0 Log-loss0.05543580278754234\n",
      "Epoch:7 Batch:80 Log-loss0.039940305054187775\n",
      "Epoch:7 Batch:160 Log-loss0.03209333494305611\n",
      "Epoch:7 Batch:240 Log-loss0.032567836344242096\n",
      "Epoch:7 Batch:320 Log-loss0.03408120945096016\n",
      "Epoch:7 Batch:400 Log-loss0.03599647805094719\n",
      "Epoch:7 Batch:480 Log-loss0.04392820969223976\n",
      "Epoch:7 Batch:560 Log-loss0.05547783151268959\n",
      "Epoch average log loss:0.03700107138803495\n",
      "In Epoch7, val_loss:0.03749939073162616, best_val_loss:0.03731478001552681, best_auc:0.9905087098300337\n",
      "Epoch:8 Batch:0 Log-loss0.03637237846851349\n",
      "Epoch:8 Batch:80 Log-loss0.031129265204072\n",
      "Epoch:8 Batch:160 Log-loss0.029288895428180695\n",
      "Epoch:8 Batch:240 Log-loss0.03234580159187317\n",
      "Epoch:8 Batch:320 Log-loss0.035518597811460495\n",
      "Epoch:8 Batch:400 Log-loss0.03533553704619408\n",
      "Epoch:8 Batch:480 Log-loss0.04363507032394409\n",
      "Epoch:8 Batch:560 Log-loss0.04663078859448433\n",
      "Epoch average log loss:0.03674673340283334\n",
      "In Epoch8, val_loss:0.03761652837824383, best_val_loss:0.03731478001552681, best_auc:0.9905087098300337\n",
      "Epoch:9 Batch:0 Log-loss0.02912326343357563\n",
      "Epoch:9 Batch:80 Log-loss0.023095062002539635\n",
      "Epoch:9 Batch:160 Log-loss0.03343208506703377\n",
      "Epoch:9 Batch:240 Log-loss0.03547780588269234\n",
      "Epoch:9 Batch:320 Log-loss0.024698294699192047\n",
      "Epoch:9 Batch:400 Log-loss0.037945035845041275\n",
      "Epoch:9 Batch:480 Log-loss0.04127874597907066\n",
      "Epoch:9 Batch:560 Log-loss0.030459841713309288\n",
      "Epoch average log loss:0.036664095600800854\n",
      "In Epoch9, val_loss:0.03763514299584433, best_val_loss:0.03731478001552681, best_auc:0.9905087098300337\n",
      "Epoch:10 Batch:0 Log-loss0.026739226654171944\n",
      "Epoch:10 Batch:80 Log-loss0.0522577166557312\n",
      "Epoch:10 Batch:160 Log-loss0.038461003452539444\n",
      "Epoch:10 Batch:240 Log-loss0.024192193523049355\n",
      "Epoch:10 Batch:320 Log-loss0.02687169425189495\n",
      "Epoch:10 Batch:400 Log-loss0.03688976168632507\n",
      "Epoch:10 Batch:480 Log-loss0.03064917027950287\n",
      "Epoch:10 Batch:560 Log-loss0.03265950456261635\n",
      "Epoch average log loss:0.03617078578931147\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.03706135061191495, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:11 Batch:0 Log-loss0.038721177726984024\n",
      "Epoch:11 Batch:80 Log-loss0.02795618213713169\n",
      "Epoch:11 Batch:160 Log-loss0.03296096995472908\n",
      "Epoch:11 Batch:240 Log-loss0.03524945303797722\n",
      "Epoch:11 Batch:320 Log-loss0.038118667900562286\n",
      "Epoch:11 Batch:400 Log-loss0.03246568515896797\n",
      "Epoch:11 Batch:480 Log-loss0.039830416440963745\n",
      "Epoch:11 Batch:560 Log-loss0.03015691600739956\n",
      "Epoch average log loss:0.03630637288858582\n",
      "In Epoch11, val_loss:0.03730308960157378, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:12 Batch:0 Log-loss0.03417128697037697\n",
      "Epoch:12 Batch:80 Log-loss0.04016919061541557\n",
      "Epoch:12 Batch:160 Log-loss0.04303596541285515\n",
      "Epoch:12 Batch:240 Log-loss0.03131209686398506\n",
      "Epoch:12 Batch:320 Log-loss0.04099592939019203\n",
      "Epoch:12 Batch:400 Log-loss0.033552877604961395\n",
      "Epoch:12 Batch:480 Log-loss0.026905404403805733\n",
      "Epoch:12 Batch:560 Log-loss0.039183326065540314\n",
      "Epoch average log loss:0.03589283583153571\n",
      "In Epoch12, val_loss:0.037180529783023754, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:13 Batch:0 Log-loss0.03043939732015133\n",
      "Epoch:13 Batch:80 Log-loss0.038936302065849304\n",
      "Epoch:13 Batch:160 Log-loss0.04173377528786659\n",
      "Epoch:13 Batch:240 Log-loss0.048106949776411057\n",
      "Epoch:13 Batch:320 Log-loss0.02806260995566845\n",
      "Epoch:13 Batch:400 Log-loss0.02851550094783306\n",
      "Epoch:13 Batch:480 Log-loss0.024467671290040016\n",
      "Epoch:13 Batch:560 Log-loss0.03155900910496712\n",
      "Epoch average log loss:0.03557944069782804\n",
      "In Epoch13, val_loss:0.03735628068188176, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:14 Batch:0 Log-loss0.04039383307099342\n",
      "Epoch:14 Batch:80 Log-loss0.036995939910411835\n",
      "Epoch:14 Batch:160 Log-loss0.04201497137546539\n",
      "Epoch:14 Batch:240 Log-loss0.02002812549471855\n",
      "Epoch:14 Batch:320 Log-loss0.03103266842663288\n",
      "Epoch:14 Batch:400 Log-loss0.02273581363260746\n",
      "Epoch:14 Batch:480 Log-loss0.026104258373379707\n",
      "Epoch:14 Batch:560 Log-loss0.023508509621024132\n",
      "Epoch average log loss:0.03558794405045254\n",
      "In Epoch14, val_loss:0.03764100956526318, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:15 Batch:0 Log-loss0.04281708598136902\n",
      "Epoch:15 Batch:80 Log-loss0.04314215108752251\n",
      "Epoch:15 Batch:160 Log-loss0.02591629885137081\n",
      "Epoch:15 Batch:240 Log-loss0.03129240497946739\n",
      "Epoch:15 Batch:320 Log-loss0.04743120074272156\n",
      "Epoch:15 Batch:400 Log-loss0.03649500012397766\n",
      "Epoch:15 Batch:480 Log-loss0.03079863451421261\n",
      "Epoch:15 Batch:560 Log-loss0.0458555594086647\n",
      "Epoch average log loss:0.03520938455725887\n",
      "In Epoch15, val_loss:0.037424386282829396, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:16 Batch:0 Log-loss0.04682460054755211\n",
      "Epoch:16 Batch:80 Log-loss0.052255779504776\n",
      "Epoch:16 Batch:160 Log-loss0.04248075559735298\n",
      "Epoch:16 Batch:240 Log-loss0.03116673044860363\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:16 Batch:320 Log-loss0.030546551570296288\n",
      "Epoch:16 Batch:400 Log-loss0.02839331515133381\n",
      "Epoch:16 Batch:480 Log-loss0.034630145877599716\n",
      "Epoch:16 Batch:560 Log-loss0.041434042155742645\n",
      "Epoch average log loss:0.03535553821296032\n",
      "In Epoch16, val_loss:0.03739511261529497, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:17 Batch:0 Log-loss0.030000939965248108\n",
      "Epoch:17 Batch:80 Log-loss0.04034572094678879\n",
      "Epoch:17 Batch:160 Log-loss0.0307935681194067\n",
      "Epoch:17 Batch:240 Log-loss0.029938489198684692\n",
      "Epoch:17 Batch:320 Log-loss0.029702000319957733\n",
      "Epoch:17 Batch:400 Log-loss0.02555868774652481\n",
      "Epoch:17 Batch:480 Log-loss0.028783852234482765\n",
      "Epoch:17 Batch:560 Log-loss0.03363228216767311\n",
      "Epoch average log loss:0.035218066278113316\n",
      "In Epoch17, val_loss:0.03736124952589379, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:18 Batch:0 Log-loss0.028585178777575493\n",
      "Epoch:18 Batch:80 Log-loss0.03286898508667946\n",
      "Epoch:18 Batch:160 Log-loss0.03968498483300209\n",
      "Epoch:18 Batch:240 Log-loss0.03937257081270218\n",
      "Epoch:18 Batch:320 Log-loss0.04628461226820946\n",
      "Epoch:18 Batch:400 Log-loss0.03405385836958885\n",
      "Epoch:18 Batch:480 Log-loss0.0247669517993927\n",
      "Epoch:18 Batch:560 Log-loss0.028492562472820282\n",
      "Epoch average log loss:0.03511891484792743\n",
      "In Epoch18, val_loss:0.037569314059571916, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:19 Batch:0 Log-loss0.03189690783619881\n",
      "Epoch:19 Batch:80 Log-loss0.027319885790348053\n",
      "Epoch:19 Batch:160 Log-loss0.040494780987501144\n",
      "Epoch:19 Batch:240 Log-loss0.03104623593389988\n",
      "Epoch:19 Batch:320 Log-loss0.029986819252371788\n",
      "Epoch:19 Batch:400 Log-loss0.04138898849487305\n",
      "Epoch:19 Batch:480 Log-loss0.03825347125530243\n",
      "Epoch:19 Batch:560 Log-loss0.03550355136394501\n",
      "Epoch average log loss:0.03498670800722071\n",
      "In Epoch19, val_loss:0.03761910316061642, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:20 Batch:0 Log-loss0.0477776937186718\n",
      "Epoch:20 Batch:80 Log-loss0.030350424349308014\n",
      "Epoch:20 Batch:160 Log-loss0.03181663155555725\n",
      "Epoch:20 Batch:240 Log-loss0.037650007754564285\n",
      "Epoch:20 Batch:320 Log-loss0.04098308086395264\n",
      "Epoch:20 Batch:400 Log-loss0.041314009577035904\n",
      "Epoch:20 Batch:480 Log-loss0.040628720074892044\n",
      "Epoch:20 Batch:560 Log-loss0.03505658358335495\n",
      "Epoch average log loss:0.03501635717194793\n",
      "In Epoch20, val_loss:0.03744419366322495, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:21 Batch:0 Log-loss0.046600326895713806\n",
      "Epoch:21 Batch:80 Log-loss0.031370360404253006\n",
      "Epoch:21 Batch:160 Log-loss0.02699771709740162\n",
      "Epoch:21 Batch:240 Log-loss0.04486501216888428\n",
      "Epoch:21 Batch:320 Log-loss0.052255745977163315\n",
      "Epoch:21 Batch:400 Log-loss0.03128497302532196\n",
      "Epoch:21 Batch:480 Log-loss0.04591405391693115\n",
      "Epoch:21 Batch:560 Log-loss0.029119854792952538\n",
      "Epoch average log loss:0.03461381731820958\n",
      "In Epoch21, val_loss:0.03777497061742435, best_val_loss:0.03706135061191495, best_auc:0.9906306823656061\n",
      "Epoch:22 Batch:0 Log-loss0.0344281904399395\n",
      "Epoch:22 Batch:80 Log-loss0.023537347093224525\n",
      "Epoch:22 Batch:160 Log-loss0.03896503895521164\n",
      "Epoch:22 Batch:240 Log-loss0.03315144404768944\n",
      "Epoch:22 Batch:320 Log-loss0.03685428574681282\n",
      "Epoch:22 Batch:400 Log-loss0.03707239404320717\n",
      "Epoch:22 Batch:480 Log-loss0.041828423738479614\n",
      "Epoch:22 Batch:560 Log-loss0.053690921515226364\n",
      "Epoch average log loss:0.034654937544837595\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_640.pt.\n",
      "\n",
      "Training on fold 1\n",
      "Epoch:1 Batch:0 Log-loss0.035626985132694244\n",
      "Epoch:1 Batch:80 Log-loss0.03828069940209389\n",
      "Epoch:1 Batch:160 Log-loss0.026043912395834923\n",
      "Epoch:1 Batch:240 Log-loss0.02567032165825367\n",
      "Epoch:1 Batch:320 Log-loss0.03185640275478363\n",
      "Epoch:1 Batch:400 Log-loss0.036188140511512756\n",
      "Epoch:1 Batch:480 Log-loss0.042369294911623\n",
      "Epoch:1 Batch:560 Log-loss0.032680414617061615\n",
      "Epoch average log loss:0.03557248935934954\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.03924193960656811, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n",
      "Epoch:2 Batch:0 Log-loss0.03344913572072983\n",
      "Epoch:2 Batch:80 Log-loss0.03796151652932167\n",
      "Epoch:2 Batch:160 Log-loss0.048528071492910385\n",
      "Epoch:2 Batch:240 Log-loss0.03850119188427925\n",
      "Epoch:2 Batch:320 Log-loss0.04263437166810036\n",
      "Epoch:2 Batch:400 Log-loss0.04197249189019203\n",
      "Epoch:2 Batch:480 Log-loss0.048088014125823975\n",
      "Epoch:2 Batch:560 Log-loss0.031898465007543564\n",
      "Epoch average log loss:0.03899480524705723\n",
      "In Epoch2, val_loss:0.0400426223368515, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n",
      "Epoch:3 Batch:0 Log-loss0.04600527882575989\n",
      "Epoch:3 Batch:80 Log-loss0.032746851444244385\n",
      "Epoch:3 Batch:160 Log-loss0.041274573653936386\n",
      "Epoch:3 Batch:240 Log-loss0.03917665407061577\n",
      "Epoch:3 Batch:320 Log-loss0.02921479195356369\n",
      "Epoch:3 Batch:400 Log-loss0.023116610944271088\n",
      "Epoch:3 Batch:480 Log-loss0.03989820554852486\n",
      "Epoch:3 Batch:560 Log-loss0.021636169403791428\n",
      "Epoch average log loss:0.03835279327551169\n",
      "In Epoch3, val_loss:0.039356870377671826, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n",
      "Epoch:4 Batch:0 Log-loss0.03682014346122742\n",
      "Epoch:4 Batch:80 Log-loss0.044422924518585205\n",
      "Epoch:4 Batch:160 Log-loss0.04938169941306114\n",
      "Epoch:4 Batch:240 Log-loss0.03645811602473259\n",
      "Epoch:4 Batch:320 Log-loss0.038604412227869034\n",
      "Epoch:4 Batch:400 Log-loss0.03815991058945656\n",
      "Epoch:4 Batch:480 Log-loss0.027318624779582024\n",
      "Epoch:4 Batch:560 Log-loss0.03615381941199303\n",
      "Epoch average log loss:0.03816765992835696\n",
      "In Epoch4, val_loss:0.039247657839455906, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n",
      "Epoch:5 Batch:0 Log-loss0.03621852397918701\n",
      "Epoch:5 Batch:80 Log-loss0.02799328975379467\n",
      "Epoch:5 Batch:160 Log-loss0.03426874056458473\n",
      "Epoch:5 Batch:240 Log-loss0.021088140085339546\n",
      "Epoch:5 Batch:320 Log-loss0.0252585019916296\n",
      "Epoch:5 Batch:400 Log-loss0.03689543530344963\n",
      "Epoch:5 Batch:480 Log-loss0.033834170550107956\n",
      "Epoch:5 Batch:560 Log-loss0.0396488755941391\n",
      "Epoch average log loss:0.037474706998494055\n",
      "In Epoch5, val_loss:0.03929132474004213, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n",
      "Epoch:6 Batch:0 Log-loss0.0379859022796154\n",
      "Epoch:6 Batch:80 Log-loss0.025187795981764793\n",
      "Epoch:6 Batch:160 Log-loss0.03159652277827263\n",
      "Epoch:6 Batch:240 Log-loss0.03173183277249336\n",
      "Epoch:6 Batch:320 Log-loss0.039490919560194016\n",
      "Epoch:6 Batch:400 Log-loss0.04478630796074867\n",
      "Epoch:6 Batch:480 Log-loss0.0442955307662487\n",
      "Epoch:6 Batch:560 Log-loss0.048734329640865326\n",
      "Epoch average log loss:0.037124927208891934\n",
      "In Epoch6, val_loss:0.03987041322403074, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n",
      "Epoch:7 Batch:0 Log-loss0.05495856702327728\n",
      "Epoch:7 Batch:80 Log-loss0.04490601643919945\n",
      "Epoch:7 Batch:160 Log-loss0.026956498622894287\n",
      "Epoch:7 Batch:240 Log-loss0.039721161127090454\n",
      "Epoch:7 Batch:320 Log-loss0.026672683656215668\n",
      "Epoch:7 Batch:400 Log-loss0.038949720561504364\n",
      "Epoch:7 Batch:480 Log-loss0.03404247760772705\n",
      "Epoch:7 Batch:560 Log-loss0.04782827943563461\n",
      "Epoch average log loss:0.037217940044190205\n",
      "In Epoch7, val_loss:0.03958644602371084, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n",
      "Epoch:8 Batch:0 Log-loss0.026013152673840523\n",
      "Epoch:8 Batch:80 Log-loss0.03148961812257767\n",
      "Epoch:8 Batch:160 Log-loss0.03641263023018837\n",
      "Epoch:8 Batch:240 Log-loss0.037662606686353683\n",
      "Epoch:8 Batch:320 Log-loss0.03515670448541641\n",
      "Epoch:8 Batch:400 Log-loss0.025840992107987404\n",
      "Epoch:8 Batch:480 Log-loss0.056693460792303085\n",
      "Epoch:8 Batch:560 Log-loss0.036310482770204544\n",
      "Epoch average log loss:0.03687140898379896\n",
      "In Epoch8, val_loss:0.03948383961725074, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n",
      "Epoch:9 Batch:0 Log-loss0.03031609021127224\n",
      "Epoch:9 Batch:80 Log-loss0.030528319999575615\n",
      "Epoch:9 Batch:160 Log-loss0.02552195079624653\n",
      "Epoch:9 Batch:240 Log-loss0.03366903215646744\n",
      "Epoch:9 Batch:320 Log-loss0.04017817601561546\n",
      "Epoch:9 Batch:400 Log-loss0.035877663642168045\n",
      "Epoch:9 Batch:480 Log-loss0.03376108780503273\n",
      "Epoch:9 Batch:560 Log-loss0.03419195115566254\n",
      "Epoch average log loss:0.03648141515773854\n",
      "In Epoch9, val_loss:0.04012867362537017, best_val_loss:0.03924193960656811, best_auc:0.9904491466550006\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:10 Batch:0 Log-loss0.03167102858424187\n",
      "Epoch:10 Batch:80 Log-loss0.03980262205004692\n",
      "Epoch:10 Batch:160 Log-loss0.03291260078549385\n",
      "Epoch:10 Batch:240 Log-loss0.039584625512361526\n",
      "Epoch:10 Batch:320 Log-loss0.0318765826523304\n",
      "Epoch:10 Batch:400 Log-loss0.034126050770282745\n",
      "Epoch:10 Batch:480 Log-loss0.035141151398420334\n",
      "Epoch:10 Batch:560 Log-loss0.03821359574794769\n",
      "Epoch average log loss:0.036590098466591114\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.03897156411771602, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:11 Batch:0 Log-loss0.03022613376379013\n",
      "Epoch:11 Batch:80 Log-loss0.02622983418405056\n",
      "Epoch:11 Batch:160 Log-loss0.048999711871147156\n",
      "Epoch:11 Batch:240 Log-loss0.026382675394415855\n",
      "Epoch:11 Batch:320 Log-loss0.03980107977986336\n",
      "Epoch:11 Batch:400 Log-loss0.03398565948009491\n",
      "Epoch:11 Batch:480 Log-loss0.03103755973279476\n",
      "Epoch:11 Batch:560 Log-loss0.03763815760612488\n",
      "Epoch average log loss:0.03620597546188427\n",
      "In Epoch11, val_loss:0.040017802888782754, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:12 Batch:0 Log-loss0.020801817998290062\n",
      "Epoch:12 Batch:80 Log-loss0.02715076506137848\n",
      "Epoch:12 Batch:160 Log-loss0.03295287489891052\n",
      "Epoch:12 Batch:240 Log-loss0.02203083597123623\n",
      "Epoch:12 Batch:320 Log-loss0.03413332626223564\n",
      "Epoch:12 Batch:400 Log-loss0.047982051968574524\n",
      "Epoch:12 Batch:480 Log-loss0.02701396308839321\n",
      "Epoch:12 Batch:560 Log-loss0.0617712065577507\n",
      "Epoch average log loss:0.03606528258283755\n",
      "In Epoch12, val_loss:0.03921425720971919, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:13 Batch:0 Log-loss0.032847993075847626\n",
      "Epoch:13 Batch:80 Log-loss0.04962657764554024\n",
      "Epoch:13 Batch:160 Log-loss0.034984465688467026\n",
      "Epoch:13 Batch:240 Log-loss0.03930603340268135\n",
      "Epoch:13 Batch:320 Log-loss0.02433246187865734\n",
      "Epoch:13 Batch:400 Log-loss0.045182570815086365\n",
      "Epoch:13 Batch:480 Log-loss0.029804104939103127\n",
      "Epoch:13 Batch:560 Log-loss0.038661763072013855\n",
      "Epoch average log loss:0.03568862836608397\n",
      "In Epoch13, val_loss:0.03933719963529369, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:14 Batch:0 Log-loss0.038088858127593994\n",
      "Epoch:14 Batch:80 Log-loss0.029780007898807526\n",
      "Epoch:14 Batch:160 Log-loss0.04887649416923523\n",
      "Epoch:14 Batch:240 Log-loss0.04418481886386871\n",
      "Epoch:14 Batch:320 Log-loss0.027222484350204468\n",
      "Epoch:14 Batch:400 Log-loss0.03348756954073906\n",
      "Epoch:14 Batch:480 Log-loss0.04469596967101097\n",
      "Epoch:14 Batch:560 Log-loss0.02792550064623356\n",
      "Epoch average log loss:0.035701028839685026\n",
      "In Epoch14, val_loss:0.039154755967781375, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:15 Batch:0 Log-loss0.04265398904681206\n",
      "Epoch:15 Batch:80 Log-loss0.026009464636445045\n",
      "Epoch:15 Batch:160 Log-loss0.036550890654325485\n",
      "Epoch:15 Batch:240 Log-loss0.040397051721811295\n",
      "Epoch:15 Batch:320 Log-loss0.04681259021162987\n",
      "Epoch:15 Batch:400 Log-loss0.03804856538772583\n",
      "Epoch:15 Batch:480 Log-loss0.024820826947689056\n",
      "Epoch:15 Batch:560 Log-loss0.029955454170703888\n",
      "Epoch average log loss:0.03564962478454358\n",
      "In Epoch15, val_loss:0.03942868945982954, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:16 Batch:0 Log-loss0.038689084351062775\n",
      "Epoch:16 Batch:80 Log-loss0.038117799907922745\n",
      "Epoch:16 Batch:160 Log-loss0.027007104828953743\n",
      "Epoch:16 Batch:240 Log-loss0.02191345952451229\n",
      "Epoch:16 Batch:320 Log-loss0.020532606169581413\n",
      "Epoch:16 Batch:400 Log-loss0.035342223942279816\n",
      "Epoch:16 Batch:480 Log-loss0.04195210337638855\n",
      "Epoch:16 Batch:560 Log-loss0.02789352647960186\n",
      "Epoch average log loss:0.035542287424738916\n",
      "In Epoch16, val_loss:0.039004498494983296, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:17 Batch:0 Log-loss0.023003561422228813\n",
      "Epoch:17 Batch:80 Log-loss0.036239199340343475\n",
      "Epoch:17 Batch:160 Log-loss0.03979151323437691\n",
      "Epoch:17 Batch:240 Log-loss0.03633042052388191\n",
      "Epoch:17 Batch:320 Log-loss0.02467917650938034\n",
      "Epoch:17 Batch:400 Log-loss0.03483619913458824\n",
      "Epoch:17 Batch:480 Log-loss0.028398437425494194\n",
      "Epoch:17 Batch:560 Log-loss0.047985292971134186\n",
      "Epoch average log loss:0.03540255965531937\n",
      "In Epoch17, val_loss:0.03916384435676679, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:18 Batch:0 Log-loss0.02251979149878025\n",
      "Epoch:18 Batch:80 Log-loss0.0414484441280365\n",
      "Epoch:18 Batch:160 Log-loss0.0318797267973423\n",
      "Epoch:18 Batch:240 Log-loss0.034422118216753006\n",
      "Epoch:18 Batch:320 Log-loss0.04312273859977722\n",
      "Epoch:18 Batch:400 Log-loss0.04238312318921089\n",
      "Epoch:18 Batch:480 Log-loss0.03267336264252663\n",
      "Epoch:18 Batch:560 Log-loss0.04228413850069046\n",
      "Epoch average log loss:0.03527888083896999\n",
      "In Epoch18, val_loss:0.03947754794329841, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:19 Batch:0 Log-loss0.03244315832853317\n",
      "Epoch:19 Batch:80 Log-loss0.027452828362584114\n",
      "Epoch:19 Batch:160 Log-loss0.029111815616488457\n",
      "Epoch:19 Batch:240 Log-loss0.03816917911171913\n",
      "Epoch:19 Batch:320 Log-loss0.026890305802226067\n",
      "Epoch:19 Batch:400 Log-loss0.042081624269485474\n",
      "Epoch:19 Batch:480 Log-loss0.039040759205818176\n",
      "Epoch:19 Batch:560 Log-loss0.041687335819005966\n",
      "Epoch average log loss:0.035086622869130224\n",
      "In Epoch19, val_loss:0.04053286617929735, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:20 Batch:0 Log-loss0.04864754155278206\n",
      "Epoch:20 Batch:80 Log-loss0.03738313540816307\n",
      "Epoch:20 Batch:160 Log-loss0.04530135914683342\n",
      "Epoch:20 Batch:240 Log-loss0.0334196500480175\n",
      "Epoch:20 Batch:320 Log-loss0.032913584262132645\n",
      "Epoch:20 Batch:400 Log-loss0.03131778538227081\n",
      "Epoch:20 Batch:480 Log-loss0.042635928839445114\n",
      "Epoch:20 Batch:560 Log-loss0.028362367302179337\n",
      "Epoch average log loss:0.0350690876965278\n",
      "In Epoch20, val_loss:0.0398252559022352, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:21 Batch:0 Log-loss0.028499409556388855\n",
      "Epoch:21 Batch:80 Log-loss0.028583331033587456\n",
      "Epoch:21 Batch:160 Log-loss0.044424671679735184\n",
      "Epoch:21 Batch:240 Log-loss0.026046274229884148\n",
      "Epoch:21 Batch:320 Log-loss0.04195655882358551\n",
      "Epoch:21 Batch:400 Log-loss0.03146921098232269\n",
      "Epoch:21 Batch:480 Log-loss0.04468996822834015\n",
      "Epoch:21 Batch:560 Log-loss0.04306117445230484\n",
      "Epoch average log loss:0.03499318260998864\n",
      "In Epoch21, val_loss:0.03958865853476445, best_val_loss:0.03897156411771602, best_auc:0.9904688482991184\n",
      "Epoch:22 Batch:0 Log-loss0.033647019416093826\n",
      "Epoch:22 Batch:80 Log-loss0.03697453811764717\n",
      "Epoch:22 Batch:160 Log-loss0.03794126957654953\n",
      "Epoch:22 Batch:240 Log-loss0.02991245687007904\n",
      "Epoch:22 Batch:320 Log-loss0.03373514488339424\n",
      "Epoch:22 Batch:400 Log-loss0.04243149235844612\n",
      "Epoch:22 Batch:480 Log-loss0.033607367426157\n",
      "Epoch:22 Batch:560 Log-loss0.022420736029744148\n",
      "Epoch average log loss:0.034895641250269754\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_641.pt.\n",
      "\n",
      "Training on fold 2\n",
      "Epoch:1 Batch:0 Log-loss0.028883716091513634\n",
      "Epoch:1 Batch:80 Log-loss0.02492702752351761\n",
      "Epoch:1 Batch:160 Log-loss0.02424725703895092\n",
      "Epoch:1 Batch:240 Log-loss0.028597524389624596\n",
      "Epoch:1 Batch:320 Log-loss0.03494271636009216\n",
      "Epoch:1 Batch:400 Log-loss0.0445033460855484\n",
      "Epoch:1 Batch:480 Log-loss0.05515245720744133\n",
      "Epoch:1 Batch:560 Log-loss0.04049067199230194\n",
      "Epoch average log loss:0.03752359235659242\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.038671918756833484, best_val_loss:0.038671918756833484, best_auc:0.9906350623176355\n",
      "Epoch:2 Batch:0 Log-loss0.025249535217881203\n",
      "Epoch:2 Batch:80 Log-loss0.03349922224879265\n",
      "Epoch:2 Batch:160 Log-loss0.029316090047359467\n",
      "Epoch:2 Batch:240 Log-loss0.048878852277994156\n",
      "Epoch:2 Batch:320 Log-loss0.045896317809820175\n",
      "Epoch:2 Batch:400 Log-loss0.039427950978279114\n",
      "Epoch:2 Batch:480 Log-loss0.03011968731880188\n",
      "Epoch:2 Batch:560 Log-loss0.03775358945131302\n",
      "Epoch average log loss:0.04048467791705791\n",
      "In Epoch2, val_loss:0.03868840871804299, best_val_loss:0.038671918756833484, best_auc:0.9906350623176355\n",
      "Epoch:3 Batch:0 Log-loss0.04278213903307915\n",
      "Epoch:3 Batch:80 Log-loss0.03662635013461113\n",
      "Epoch:3 Batch:160 Log-loss0.03270016983151436\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:3 Batch:240 Log-loss0.032005272805690765\n",
      "Epoch:3 Batch:320 Log-loss0.0390213280916214\n",
      "Epoch:3 Batch:400 Log-loss0.03865356370806694\n",
      "Epoch:3 Batch:480 Log-loss0.0460469089448452\n",
      "Epoch:3 Batch:560 Log-loss0.02496490813791752\n",
      "Epoch average log loss:0.039944548555649814\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.03834335084060546, best_val_loss:0.03834335084060546, best_auc:0.990730663749567\n",
      "Epoch:4 Batch:0 Log-loss0.03521014377474785\n",
      "Epoch:4 Batch:80 Log-loss0.048332978039979935\n",
      "Epoch:4 Batch:160 Log-loss0.051618512719869614\n",
      "Epoch:4 Batch:240 Log-loss0.033066276460886\n",
      "Epoch:4 Batch:320 Log-loss0.04460427537560463\n",
      "Epoch:4 Batch:400 Log-loss0.03888536989688873\n",
      "Epoch:4 Batch:480 Log-loss0.032224517315626144\n",
      "Epoch:4 Batch:560 Log-loss0.0522904247045517\n",
      "Epoch average log loss:0.039509261949985684\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.0379504426149232, best_val_loss:0.0379504426149232, best_auc:0.9909377505154269\n",
      "Epoch:5 Batch:0 Log-loss0.027869736775755882\n",
      "Epoch:5 Batch:80 Log-loss0.06167254224419594\n",
      "Epoch:5 Batch:160 Log-loss0.05525290593504906\n",
      "Epoch:5 Batch:240 Log-loss0.03310811147093773\n",
      "Epoch:5 Batch:320 Log-loss0.03005312941968441\n",
      "Epoch:5 Batch:400 Log-loss0.044846415519714355\n",
      "Epoch:5 Batch:480 Log-loss0.05000854656100273\n",
      "Epoch:5 Batch:560 Log-loss0.03467236086726189\n",
      "Epoch average log loss:0.03900071149026709\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.03775904332067562, best_val_loss:0.03775904332067562, best_auc:0.9911597108621577\n",
      "Epoch:6 Batch:0 Log-loss0.0410468652844429\n",
      "Epoch:6 Batch:80 Log-loss0.03976452723145485\n",
      "Epoch:6 Batch:160 Log-loss0.024678679183125496\n",
      "Epoch:6 Batch:240 Log-loss0.030070632696151733\n",
      "Epoch:6 Batch:320 Log-loss0.048615243285894394\n",
      "Epoch:6 Batch:400 Log-loss0.04817819595336914\n",
      "Epoch:6 Batch:480 Log-loss0.036037225276231766\n",
      "Epoch:6 Batch:560 Log-loss0.03687088564038277\n",
      "Epoch average log loss:0.03903456479643604\n",
      "In Epoch6, val_loss:0.038373425187467815, best_val_loss:0.03775904332067562, best_auc:0.9911597108621577\n",
      "Epoch:7 Batch:0 Log-loss0.0473572313785553\n",
      "Epoch:7 Batch:80 Log-loss0.03697242587804794\n",
      "Epoch:7 Batch:160 Log-loss0.051110997796058655\n",
      "Epoch:7 Batch:240 Log-loss0.049939077347517014\n",
      "Epoch:7 Batch:320 Log-loss0.045605387538671494\n",
      "Epoch:7 Batch:400 Log-loss0.030873892828822136\n",
      "Epoch:7 Batch:480 Log-loss0.04640620946884155\n",
      "Epoch:7 Batch:560 Log-loss0.0436076782643795\n",
      "Epoch average log loss:0.03854173519648611\n",
      "In Epoch7, val_loss:0.037985447244244144, best_val_loss:0.03775904332067562, best_auc:0.9911597108621577\n",
      "Epoch:8 Batch:0 Log-loss0.04870182275772095\n",
      "Epoch:8 Batch:80 Log-loss0.03812522068619728\n",
      "Epoch:8 Batch:160 Log-loss0.04538673534989357\n",
      "Epoch:8 Batch:240 Log-loss0.03200311213731766\n",
      "Epoch:8 Batch:320 Log-loss0.030348027125000954\n",
      "Epoch:8 Batch:400 Log-loss0.04436039552092552\n",
      "Epoch:8 Batch:480 Log-loss0.036177244037389755\n",
      "Epoch:8 Batch:560 Log-loss0.05166351795196533\n",
      "Epoch average log loss:0.0382403743652893\n",
      "In Epoch8, val_loss:0.03777604955032421, best_val_loss:0.03775904332067562, best_auc:0.9911597108621577\n",
      "Epoch:9 Batch:0 Log-loss0.046179693192243576\n",
      "Epoch:9 Batch:80 Log-loss0.0391143336892128\n",
      "Epoch:9 Batch:160 Log-loss0.035816553980112076\n",
      "Epoch:9 Batch:240 Log-loss0.0481049083173275\n",
      "Epoch:9 Batch:320 Log-loss0.0499703474342823\n",
      "Epoch:9 Batch:400 Log-loss0.0434359572827816\n",
      "Epoch:9 Batch:480 Log-loss0.0397024005651474\n",
      "Epoch:9 Batch:560 Log-loss0.03494461625814438\n",
      "Epoch average log loss:0.037881507358646815\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.03767924390509154, best_val_loss:0.03767924390509154, best_auc:0.9911778479580584\n",
      "Epoch:10 Batch:0 Log-loss0.02553788386285305\n",
      "Epoch:10 Batch:80 Log-loss0.03114454448223114\n",
      "Epoch:10 Batch:160 Log-loss0.043627288192510605\n",
      "Epoch:10 Batch:240 Log-loss0.043965090066194534\n",
      "Epoch:10 Batch:320 Log-loss0.04329986870288849\n",
      "Epoch:10 Batch:400 Log-loss0.02651750110089779\n",
      "Epoch:10 Batch:480 Log-loss0.03528742119669914\n",
      "Epoch:10 Batch:560 Log-loss0.04217866063117981\n",
      "Epoch average log loss:0.03756901265255042\n",
      "In Epoch10, val_loss:0.037691845679837736, best_val_loss:0.03767924390509154, best_auc:0.9911778479580584\n",
      "Epoch:11 Batch:0 Log-loss0.036956917494535446\n",
      "Epoch:11 Batch:80 Log-loss0.045160114765167236\n",
      "Epoch:11 Batch:160 Log-loss0.061157554388046265\n",
      "Epoch:11 Batch:240 Log-loss0.03738059476017952\n",
      "Epoch:11 Batch:320 Log-loss0.02797752618789673\n",
      "Epoch:11 Batch:400 Log-loss0.024483663961291313\n",
      "Epoch:11 Batch:480 Log-loss0.051957353949546814\n",
      "Epoch:11 Batch:560 Log-loss0.045084018260240555\n",
      "Epoch average log loss:0.03766180091936674\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch11, val_loss:0.037602736538535894, best_val_loss:0.037602736538535894, best_auc:0.9913340981119011\n",
      "Epoch:12 Batch:0 Log-loss0.0326203778386116\n",
      "Epoch:12 Batch:80 Log-loss0.03277045115828514\n",
      "Epoch:12 Batch:160 Log-loss0.030371317639946938\n",
      "Epoch:12 Batch:240 Log-loss0.03162987902760506\n",
      "Epoch:12 Batch:320 Log-loss0.031969375908374786\n",
      "Epoch:12 Batch:400 Log-loss0.03578484430909157\n",
      "Epoch:12 Batch:480 Log-loss0.03777768835425377\n",
      "Epoch:12 Batch:560 Log-loss0.04903336614370346\n",
      "Epoch average log loss:0.03718017227615097\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch12, val_loss:0.03755628232418578, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:13 Batch:0 Log-loss0.022390000522136688\n",
      "Epoch:13 Batch:80 Log-loss0.0325922854244709\n",
      "Epoch:13 Batch:160 Log-loss0.03216461464762688\n",
      "Epoch:13 Batch:240 Log-loss0.025836460292339325\n",
      "Epoch:13 Batch:320 Log-loss0.028630098327994347\n",
      "Epoch:13 Batch:400 Log-loss0.038700710982084274\n",
      "Epoch:13 Batch:480 Log-loss0.03056333027780056\n",
      "Epoch:13 Batch:560 Log-loss0.03593162074685097\n",
      "Epoch average log loss:0.03699992422979059\n",
      "In Epoch13, val_loss:0.03856321555874187, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:14 Batch:0 Log-loss0.04128136858344078\n",
      "Epoch:14 Batch:80 Log-loss0.03811439499258995\n",
      "Epoch:14 Batch:160 Log-loss0.06143251433968544\n",
      "Epoch:14 Batch:240 Log-loss0.032752566039562225\n",
      "Epoch:14 Batch:320 Log-loss0.039617348462343216\n",
      "Epoch:14 Batch:400 Log-loss0.0569654144346714\n",
      "Epoch:14 Batch:480 Log-loss0.02731451392173767\n",
      "Epoch:14 Batch:560 Log-loss0.030886273831129074\n",
      "Epoch average log loss:0.03697886757186747\n",
      "In Epoch14, val_loss:0.03808642077527443, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:15 Batch:0 Log-loss0.04936458170413971\n",
      "Epoch:15 Batch:80 Log-loss0.028094975277781487\n",
      "Epoch:15 Batch:160 Log-loss0.04064654931426048\n",
      "Epoch:15 Batch:240 Log-loss0.03979684039950371\n",
      "Epoch:15 Batch:320 Log-loss0.03623379394412041\n",
      "Epoch:15 Batch:400 Log-loss0.035010237246751785\n",
      "Epoch:15 Batch:480 Log-loss0.027150174602866173\n",
      "Epoch:15 Batch:560 Log-loss0.037347208708524704\n",
      "Epoch average log loss:0.03664585511266653\n",
      "In Epoch15, val_loss:0.0379653618549717, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:16 Batch:0 Log-loss0.03327383100986481\n",
      "Epoch:16 Batch:80 Log-loss0.042190831154584885\n",
      "Epoch:16 Batch:160 Log-loss0.03212277591228485\n",
      "Epoch:16 Batch:240 Log-loss0.030463343486189842\n",
      "Epoch:16 Batch:320 Log-loss0.04194827005267143\n",
      "Epoch:16 Batch:400 Log-loss0.04251076281070709\n",
      "Epoch:16 Batch:480 Log-loss0.0440693162381649\n",
      "Epoch:16 Batch:560 Log-loss0.02401268295943737\n",
      "Epoch average log loss:0.03675237641603287\n",
      "In Epoch16, val_loss:0.03786213239604825, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:17 Batch:0 Log-loss0.03156306967139244\n",
      "Epoch:17 Batch:80 Log-loss0.026638248935341835\n",
      "Epoch:17 Batch:160 Log-loss0.030400261282920837\n",
      "Epoch:17 Batch:240 Log-loss0.018669752404093742\n",
      "Epoch:17 Batch:320 Log-loss0.035271868109703064\n",
      "Epoch:17 Batch:400 Log-loss0.042646657675504684\n",
      "Epoch:17 Batch:480 Log-loss0.031096892431378365\n",
      "Epoch:17 Batch:560 Log-loss0.044704120606184006\n",
      "Epoch average log loss:0.03635879551792251\n",
      "In Epoch17, val_loss:0.037606122577827465, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:18 Batch:0 Log-loss0.03631376847624779\n",
      "Epoch:18 Batch:80 Log-loss0.02810768596827984\n",
      "Epoch:18 Batch:160 Log-loss0.026982231065630913\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:18 Batch:240 Log-loss0.030680930241942406\n",
      "Epoch:18 Batch:320 Log-loss0.042495403438806534\n",
      "Epoch:18 Batch:400 Log-loss0.04091886058449745\n",
      "Epoch:18 Batch:480 Log-loss0.03740249201655388\n",
      "Epoch:18 Batch:560 Log-loss0.049633562564849854\n",
      "Epoch average log loss:0.036289786494203974\n",
      "In Epoch18, val_loss:0.03763787818940871, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:19 Batch:0 Log-loss0.04028857871890068\n",
      "Epoch:19 Batch:80 Log-loss0.029782412573695183\n",
      "Epoch:19 Batch:160 Log-loss0.0276960376650095\n",
      "Epoch:19 Batch:240 Log-loss0.0395294725894928\n",
      "Epoch:19 Batch:320 Log-loss0.062283992767333984\n",
      "Epoch:19 Batch:400 Log-loss0.04187283292412758\n",
      "Epoch:19 Batch:480 Log-loss0.03349567577242851\n",
      "Epoch:19 Batch:560 Log-loss0.05303627625107765\n",
      "Epoch average log loss:0.036181867871034355\n",
      "In Epoch19, val_loss:0.037795202638426784, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:20 Batch:0 Log-loss0.04225537180900574\n",
      "Epoch:20 Batch:80 Log-loss0.034826185554265976\n",
      "Epoch:20 Batch:160 Log-loss0.025621598586440086\n",
      "Epoch:20 Batch:240 Log-loss0.036497991532087326\n",
      "Epoch:20 Batch:320 Log-loss0.03583785146474838\n",
      "Epoch:20 Batch:400 Log-loss0.03781317174434662\n",
      "Epoch:20 Batch:480 Log-loss0.0419403612613678\n",
      "Epoch:20 Batch:560 Log-loss0.03908979892730713\n",
      "Epoch average log loss:0.03618457189295441\n",
      "In Epoch20, val_loss:0.037784034394750955, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:21 Batch:0 Log-loss0.03078138642013073\n",
      "Epoch:21 Batch:80 Log-loss0.023425934836268425\n",
      "Epoch:21 Batch:160 Log-loss0.03742293640971184\n",
      "Epoch:21 Batch:240 Log-loss0.03650260344147682\n",
      "Epoch:21 Batch:320 Log-loss0.045512016862630844\n",
      "Epoch:21 Batch:400 Log-loss0.033851977437734604\n",
      "Epoch:21 Batch:480 Log-loss0.04269875958561897\n",
      "Epoch:21 Batch:560 Log-loss0.03569115325808525\n",
      "Epoch average log loss:0.035916680240604494\n",
      "In Epoch21, val_loss:0.03774134185664921, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:22 Batch:0 Log-loss0.031071163713932037\n",
      "Epoch:22 Batch:80 Log-loss0.025549909099936485\n",
      "Epoch:22 Batch:160 Log-loss0.03324827924370766\n",
      "Epoch:22 Batch:240 Log-loss0.026035455986857414\n",
      "Epoch:22 Batch:320 Log-loss0.04135914891958237\n",
      "Epoch:22 Batch:400 Log-loss0.05890090763568878\n",
      "Epoch:22 Batch:480 Log-loss0.03149508684873581\n",
      "Epoch:22 Batch:560 Log-loss0.036718226969242096\n",
      "Epoch average log loss:0.035915038994114315\n",
      "In Epoch22, val_loss:0.03782105015215478, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:23 Batch:0 Log-loss0.04458479583263397\n",
      "Epoch:23 Batch:80 Log-loss0.025982273742556572\n",
      "Epoch:23 Batch:160 Log-loss0.03978949412703514\n",
      "Epoch:23 Batch:240 Log-loss0.035578202456235886\n",
      "Epoch:23 Batch:320 Log-loss0.0332987904548645\n",
      "Epoch:23 Batch:400 Log-loss0.039372701197862625\n",
      "Epoch:23 Batch:480 Log-loss0.023922888562083244\n",
      "Epoch:23 Batch:560 Log-loss0.034787457436323166\n",
      "Epoch average log loss:0.035732152768676834\n",
      "In Epoch23, val_loss:0.037663987525373, best_val_loss:0.03755628232418578, best_auc:0.991321868146562\n",
      "Epoch:24 Batch:0 Log-loss0.05314834043383598\n",
      "Epoch:24 Batch:80 Log-loss0.03450440242886543\n",
      "Epoch:24 Batch:160 Log-loss0.027647359296679497\n",
      "Epoch:24 Batch:240 Log-loss0.03439122810959816\n",
      "Epoch:24 Batch:320 Log-loss0.037502188235521317\n",
      "Epoch:24 Batch:400 Log-loss0.02555975876748562\n",
      "Epoch:24 Batch:480 Log-loss0.020609397441148758\n",
      "Epoch:24 Batch:560 Log-loss0.03160480782389641\n",
      "Epoch average log loss:0.03581970087418865\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_642.pt.\n",
      "\n",
      "Training on fold 3\n",
      "Epoch:1 Batch:0 Log-loss0.03947494551539421\n",
      "Epoch:1 Batch:80 Log-loss0.054464858025312424\n",
      "Epoch:1 Batch:160 Log-loss0.016123661771416664\n",
      "Epoch:1 Batch:240 Log-loss0.027640221640467644\n",
      "Epoch:1 Batch:320 Log-loss0.032790180295705795\n",
      "Epoch:1 Batch:400 Log-loss0.02542540617287159\n",
      "Epoch:1 Batch:480 Log-loss0.02836250513792038\n",
      "Epoch:1 Batch:560 Log-loss0.039309415966272354\n",
      "Epoch average log loss:0.03503922972761627\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.040579157746533145, best_val_loss:0.040579157746533145, best_auc:0.9883954367989746\n",
      "Epoch:2 Batch:0 Log-loss0.03558769449591637\n",
      "Epoch:2 Batch:80 Log-loss0.028157318010926247\n",
      "Epoch:2 Batch:160 Log-loss0.0334915891289711\n",
      "Epoch:2 Batch:240 Log-loss0.03248453140258789\n",
      "Epoch:2 Batch:320 Log-loss0.041080787777900696\n",
      "Epoch:2 Batch:400 Log-loss0.03724556788802147\n",
      "Epoch:2 Batch:480 Log-loss0.029260367155075073\n",
      "Epoch:2 Batch:560 Log-loss0.0588865801692009\n",
      "Epoch average log loss:0.03863335102796554\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.04052772755224245, best_val_loss:0.04052772755224245, best_auc:0.9881734416752536\n",
      "Epoch:3 Batch:0 Log-loss0.04627963528037071\n",
      "Epoch:3 Batch:80 Log-loss0.024163536727428436\n",
      "Epoch:3 Batch:160 Log-loss0.02892494760453701\n",
      "Epoch:3 Batch:240 Log-loss0.03735874965786934\n",
      "Epoch:3 Batch:320 Log-loss0.026981329545378685\n",
      "Epoch:3 Batch:400 Log-loss0.04657632112503052\n",
      "Epoch:3 Batch:480 Log-loss0.043922606855630875\n",
      "Epoch:3 Batch:560 Log-loss0.027830153703689575\n",
      "Epoch average log loss:0.03776203237607011\n",
      "In Epoch3, val_loss:0.0411094014356578, best_val_loss:0.04052772755224245, best_auc:0.9881734416752536\n",
      "Epoch:4 Batch:0 Log-loss0.045577019453048706\n",
      "Epoch:4 Batch:80 Log-loss0.05467700585722923\n",
      "Epoch:4 Batch:160 Log-loss0.034986916929483414\n",
      "Epoch:4 Batch:240 Log-loss0.04928092658519745\n",
      "Epoch:4 Batch:320 Log-loss0.041316237300634384\n",
      "Epoch:4 Batch:400 Log-loss0.03954320400953293\n",
      "Epoch:4 Batch:480 Log-loss0.046788305044174194\n",
      "Epoch:4 Batch:560 Log-loss0.03894753381609917\n",
      "Epoch average log loss:0.037847701390273866\n",
      "In Epoch4, val_loss:0.040670798145402275, best_val_loss:0.04052772755224245, best_auc:0.9881734416752536\n",
      "Epoch:5 Batch:0 Log-loss0.04686901345849037\n",
      "Epoch:5 Batch:80 Log-loss0.050912510603666306\n",
      "Epoch:5 Batch:160 Log-loss0.04547111317515373\n",
      "Epoch:5 Batch:240 Log-loss0.04103565961122513\n",
      "Epoch:5 Batch:320 Log-loss0.03860987350344658\n",
      "Epoch:5 Batch:400 Log-loss0.0370270200073719\n",
      "Epoch:5 Batch:480 Log-loss0.036716360598802567\n",
      "Epoch:5 Batch:560 Log-loss0.03795710206031799\n",
      "Epoch average log loss:0.03737234348830368\n",
      "In Epoch5, val_loss:0.04094601404670873, best_val_loss:0.04052772755224245, best_auc:0.9881734416752536\n",
      "Epoch:6 Batch:0 Log-loss0.029073871672153473\n",
      "Epoch:6 Batch:80 Log-loss0.03865678235888481\n",
      "Epoch:6 Batch:160 Log-loss0.024249644950032234\n",
      "Epoch:6 Batch:240 Log-loss0.042570922523736954\n",
      "Epoch:6 Batch:320 Log-loss0.037265099585056305\n",
      "Epoch:6 Batch:400 Log-loss0.029195301234722137\n",
      "Epoch:6 Batch:480 Log-loss0.04425312206149101\n",
      "Epoch:6 Batch:560 Log-loss0.039404626935720444\n",
      "Epoch average log loss:0.0371445541702477\n",
      "In Epoch6, val_loss:0.040613575589987724, best_val_loss:0.04052772755224245, best_auc:0.9881734416752536\n",
      "Epoch:7 Batch:0 Log-loss0.02766125462949276\n",
      "Epoch:7 Batch:80 Log-loss0.04502272978425026\n",
      "Epoch:7 Batch:160 Log-loss0.04414593055844307\n",
      "Epoch:7 Batch:240 Log-loss0.03642166778445244\n",
      "Epoch:7 Batch:320 Log-loss0.03057139366865158\n",
      "Epoch:7 Batch:400 Log-loss0.04231312870979309\n",
      "Epoch:7 Batch:480 Log-loss0.03693712130188942\n",
      "Epoch:7 Batch:560 Log-loss0.03465138375759125\n",
      "Epoch average log loss:0.03686030036104577\n",
      "In Epoch7, val_loss:0.040844753987343045, best_val_loss:0.04052772755224245, best_auc:0.9881734416752536\n",
      "Epoch:8 Batch:0 Log-loss0.053547028452157974\n",
      "Epoch:8 Batch:80 Log-loss0.034797679632902145\n",
      "Epoch:8 Batch:160 Log-loss0.028844794258475304\n",
      "Epoch:8 Batch:240 Log-loss0.0386822335422039\n",
      "Epoch:8 Batch:320 Log-loss0.04586678370833397\n",
      "Epoch:8 Batch:400 Log-loss0.04835519194602966\n",
      "Epoch:8 Batch:480 Log-loss0.03222687169909477\n",
      "Epoch:8 Batch:560 Log-loss0.036040227860212326\n",
      "Epoch average log loss:0.0366849406705504\n",
      "In Epoch8, val_loss:0.04095926473292447, best_val_loss:0.04052772755224245, best_auc:0.9881734416752536\n",
      "Epoch:9 Batch:0 Log-loss0.03819284215569496\n",
      "Epoch:9 Batch:80 Log-loss0.031074441969394684\n",
      "Epoch:9 Batch:160 Log-loss0.04730287566781044\n",
      "Epoch:9 Batch:240 Log-loss0.042917776852846146\n",
      "Epoch:9 Batch:320 Log-loss0.03131013736128807\n",
      "Epoch:9 Batch:400 Log-loss0.03531276062130928\n",
      "Epoch:9 Batch:480 Log-loss0.04479474201798439\n",
      "Epoch:9 Batch:560 Log-loss0.042685385793447495\n",
      "Epoch average log loss:0.03635832427867821\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.04031868534488856, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:10 Batch:0 Log-loss0.037676919251680374\n",
      "Epoch:10 Batch:80 Log-loss0.026269955560564995\n",
      "Epoch:10 Batch:160 Log-loss0.032758038491010666\n",
      "Epoch:10 Batch:240 Log-loss0.03625533729791641\n",
      "Epoch:10 Batch:320 Log-loss0.03885022923350334\n",
      "Epoch:10 Batch:400 Log-loss0.03763432055711746\n",
      "Epoch:10 Batch:480 Log-loss0.03844169154763222\n",
      "Epoch:10 Batch:560 Log-loss0.029993753880262375\n",
      "Epoch average log loss:0.03610843778920493\n",
      "In Epoch10, val_loss:0.0408912177192061, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:11 Batch:0 Log-loss0.04773542657494545\n",
      "Epoch:11 Batch:80 Log-loss0.03310179337859154\n",
      "Epoch:11 Batch:160 Log-loss0.040052250027656555\n",
      "Epoch:11 Batch:240 Log-loss0.030351713299751282\n",
      "Epoch:11 Batch:320 Log-loss0.056189220398664474\n",
      "Epoch:11 Batch:400 Log-loss0.024402394890785217\n",
      "Epoch:11 Batch:480 Log-loss0.030284272506833076\n",
      "Epoch:11 Batch:560 Log-loss0.04159332439303398\n",
      "Epoch average log loss:0.03584429250165288\n",
      "In Epoch11, val_loss:0.04075033815547518, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:12 Batch:0 Log-loss0.024157406762242317\n",
      "Epoch:12 Batch:80 Log-loss0.038902461528778076\n",
      "Epoch:12 Batch:160 Log-loss0.04434642568230629\n",
      "Epoch:12 Batch:240 Log-loss0.026910722255706787\n",
      "Epoch:12 Batch:320 Log-loss0.029507294297218323\n",
      "Epoch:12 Batch:400 Log-loss0.03301509469747543\n",
      "Epoch:12 Batch:480 Log-loss0.02061251737177372\n",
      "Epoch:12 Batch:560 Log-loss0.03997401148080826\n",
      "Epoch average log loss:0.035850260396754104\n",
      "In Epoch12, val_loss:0.040624241049023194, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:13 Batch:0 Log-loss0.03293347358703613\n",
      "Epoch:13 Batch:80 Log-loss0.02257128618657589\n",
      "Epoch:13 Batch:160 Log-loss0.047872111201286316\n",
      "Epoch:13 Batch:240 Log-loss0.03812186419963837\n",
      "Epoch:13 Batch:320 Log-loss0.039202842861413956\n",
      "Epoch:13 Batch:400 Log-loss0.03707672655582428\n",
      "Epoch:13 Batch:480 Log-loss0.051144346594810486\n",
      "Epoch:13 Batch:560 Log-loss0.03367004916071892\n",
      "Epoch average log loss:0.035551946029798796\n",
      "In Epoch13, val_loss:0.04076856469901393, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:14 Batch:0 Log-loss0.04692264273762703\n",
      "Epoch:14 Batch:80 Log-loss0.04235001280903816\n",
      "Epoch:14 Batch:160 Log-loss0.028478438034653664\n",
      "Epoch:14 Batch:240 Log-loss0.04450022056698799\n",
      "Epoch:14 Batch:320 Log-loss0.05222268030047417\n",
      "Epoch:14 Batch:400 Log-loss0.03670946881175041\n",
      "Epoch:14 Batch:480 Log-loss0.04040183871984482\n",
      "Epoch:14 Batch:560 Log-loss0.027951881289482117\n",
      "Epoch average log loss:0.035588968255823215\n",
      "In Epoch14, val_loss:0.04130926669281977, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:15 Batch:0 Log-loss0.034103330224752426\n",
      "Epoch:15 Batch:80 Log-loss0.0292892437428236\n",
      "Epoch:15 Batch:160 Log-loss0.040980856865644455\n",
      "Epoch:15 Batch:240 Log-loss0.03315562754869461\n",
      "Epoch:15 Batch:320 Log-loss0.032203227281570435\n",
      "Epoch:15 Batch:400 Log-loss0.0329601913690567\n",
      "Epoch:15 Batch:480 Log-loss0.035249534994363785\n",
      "Epoch:15 Batch:560 Log-loss0.02000151202082634\n",
      "Epoch average log loss:0.03532198511197099\n",
      "In Epoch15, val_loss:0.04099235790921459, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:16 Batch:0 Log-loss0.04624561592936516\n",
      "Epoch:16 Batch:80 Log-loss0.035420145839452744\n",
      "Epoch:16 Batch:160 Log-loss0.026347694918513298\n",
      "Epoch:16 Batch:240 Log-loss0.03443874418735504\n",
      "Epoch:16 Batch:320 Log-loss0.034959450364112854\n",
      "Epoch:16 Batch:400 Log-loss0.04302293062210083\n",
      "Epoch:16 Batch:480 Log-loss0.02552635967731476\n",
      "Epoch:16 Batch:560 Log-loss0.028992274776101112\n",
      "Epoch average log loss:0.03522947883299951\n",
      "In Epoch16, val_loss:0.041265106035263295, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:17 Batch:0 Log-loss0.03226509690284729\n",
      "Epoch:17 Batch:80 Log-loss0.03175849840044975\n",
      "Epoch:17 Batch:160 Log-loss0.03677361458539963\n",
      "Epoch:17 Batch:240 Log-loss0.04616744443774223\n",
      "Epoch:17 Batch:320 Log-loss0.038561828434467316\n",
      "Epoch:17 Batch:400 Log-loss0.035472121089696884\n",
      "Epoch:17 Batch:480 Log-loss0.030606890097260475\n",
      "Epoch:17 Batch:560 Log-loss0.029783494770526886\n",
      "Epoch average log loss:0.03510041280948956\n",
      "In Epoch17, val_loss:0.0419737181221796, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:18 Batch:0 Log-loss0.03914853185415268\n",
      "Epoch:18 Batch:80 Log-loss0.02550322748720646\n",
      "Epoch:18 Batch:160 Log-loss0.036711014807224274\n",
      "Epoch:18 Batch:240 Log-loss0.03745883330702782\n",
      "Epoch:18 Batch:320 Log-loss0.04246428981423378\n",
      "Epoch:18 Batch:400 Log-loss0.04331068694591522\n",
      "Epoch:18 Batch:480 Log-loss0.02563236467540264\n",
      "Epoch:18 Batch:560 Log-loss0.041588194668293\n",
      "Epoch average log loss:0.03499074494466185\n",
      "In Epoch18, val_loss:0.040979396215336576, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:19 Batch:0 Log-loss0.031008228659629822\n",
      "Epoch:19 Batch:80 Log-loss0.03863659128546715\n",
      "Epoch:19 Batch:160 Log-loss0.05324491858482361\n",
      "Epoch:19 Batch:240 Log-loss0.03560011461377144\n",
      "Epoch:19 Batch:320 Log-loss0.04290671646595001\n",
      "Epoch:19 Batch:400 Log-loss0.04173646494746208\n",
      "Epoch:19 Batch:480 Log-loss0.037986163049936295\n",
      "Epoch:19 Batch:560 Log-loss0.035214148461818695\n",
      "Epoch average log loss:0.034774121991358695\n",
      "In Epoch19, val_loss:0.04157966824072044, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:20 Batch:0 Log-loss0.039690542966127396\n",
      "Epoch:20 Batch:80 Log-loss0.03799114748835564\n",
      "Epoch:20 Batch:160 Log-loss0.03615431860089302\n",
      "Epoch:20 Batch:240 Log-loss0.01851480081677437\n",
      "Epoch:20 Batch:320 Log-loss0.024119028821587563\n",
      "Epoch:20 Batch:400 Log-loss0.03743697330355644\n",
      "Epoch:20 Batch:480 Log-loss0.03818633407354355\n",
      "Epoch:20 Batch:560 Log-loss0.040624454617500305\n",
      "Epoch average log loss:0.03470889662691791\n",
      "In Epoch20, val_loss:0.04102177725869749, best_val_loss:0.04031868534488856, best_auc:0.988507101980355\n",
      "Epoch:21 Batch:0 Log-loss0.0431065559387207\n",
      "Epoch:21 Batch:80 Log-loss0.0477595292031765\n",
      "Epoch:21 Batch:160 Log-loss0.02689531445503235\n",
      "Epoch:21 Batch:240 Log-loss0.040841858834028244\n",
      "Epoch:21 Batch:320 Log-loss0.039446718990802765\n",
      "Epoch:21 Batch:400 Log-loss0.02305573970079422\n",
      "Epoch:21 Batch:480 Log-loss0.030508780851960182\n",
      "Epoch:21 Batch:560 Log-loss0.04222488030791283\n",
      "Epoch average log loss:0.034839341985726996\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_643.pt.\n",
      "\n",
      "Training on fold 4\n",
      "Epoch:1 Batch:0 Log-loss0.03514398634433746\n",
      "Epoch:1 Batch:80 Log-loss0.027597812935709953\n",
      "Epoch:1 Batch:160 Log-loss0.04532581567764282\n",
      "Epoch:1 Batch:240 Log-loss0.04779863357543945\n",
      "Epoch:1 Batch:320 Log-loss0.045417558401823044\n",
      "Epoch:1 Batch:400 Log-loss0.029956365004181862\n",
      "Epoch:1 Batch:480 Log-loss0.030506765469908714\n",
      "Epoch:1 Batch:560 Log-loss0.025315508246421814\n",
      "Epoch average log loss:0.03516570514932807\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.037972753504137126, best_val_loss:0.037972753504137126, best_auc:0.9901462779772943\n",
      "Epoch:2 Batch:0 Log-loss0.03448239341378212\n",
      "Epoch:2 Batch:80 Log-loss0.03891701623797417\n",
      "Epoch:2 Batch:160 Log-loss0.041827473789453506\n",
      "Epoch:2 Batch:240 Log-loss0.0502757728099823\n",
      "Epoch:2 Batch:320 Log-loss0.05907377600669861\n",
      "Epoch:2 Batch:400 Log-loss0.03738711401820183\n",
      "Epoch:2 Batch:480 Log-loss0.04936307668685913\n",
      "Epoch:2 Batch:560 Log-loss0.041400372982025146\n",
      "Epoch average log loss:0.03859397578053177\n",
      "In Epoch2, val_loss:0.03819122114955265, best_val_loss:0.037972753504137126, best_auc:0.9901462779772943\n",
      "Epoch:3 Batch:0 Log-loss0.022537657991051674\n",
      "Epoch:3 Batch:80 Log-loss0.03961651772260666\n",
      "Epoch:3 Batch:160 Log-loss0.026181139051914215\n",
      "Epoch:3 Batch:240 Log-loss0.021308481693267822\n",
      "Epoch:3 Batch:320 Log-loss0.03389531373977661\n",
      "Epoch:3 Batch:400 Log-loss0.02969621866941452\n",
      "Epoch:3 Batch:480 Log-loss0.03820526599884033\n",
      "Epoch:3 Batch:560 Log-loss0.02901245653629303\n",
      "Epoch average log loss:0.03799981511464076\n",
      "In Epoch3, val_loss:0.039190467598933604, best_val_loss:0.037972753504137126, best_auc:0.9901462779772943\n",
      "Epoch:4 Batch:0 Log-loss0.03765934705734253\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:4 Batch:80 Log-loss0.037233926355838776\n",
      "Epoch:4 Batch:160 Log-loss0.044129639863967896\n",
      "Epoch:4 Batch:240 Log-loss0.05626616254448891\n",
      "Epoch:4 Batch:320 Log-loss0.03646199777722359\n",
      "Epoch:4 Batch:400 Log-loss0.0284369308501482\n",
      "Epoch:4 Batch:480 Log-loss0.041317906230688095\n",
      "Epoch:4 Batch:560 Log-loss0.032546672970056534\n",
      "Epoch average log loss:0.03771523407527379\n",
      "In Epoch4, val_loss:0.03833784164551411, best_val_loss:0.037972753504137126, best_auc:0.9901462779772943\n",
      "Epoch:5 Batch:0 Log-loss0.03403029963374138\n",
      "Epoch:5 Batch:80 Log-loss0.03243027999997139\n",
      "Epoch:5 Batch:160 Log-loss0.035253364592790604\n",
      "Epoch:5 Batch:240 Log-loss0.036598023027181625\n",
      "Epoch:5 Batch:320 Log-loss0.04244200512766838\n",
      "Epoch:5 Batch:400 Log-loss0.04261108115315437\n",
      "Epoch:5 Batch:480 Log-loss0.032846126705408096\n",
      "Epoch:5 Batch:560 Log-loss0.04084589332342148\n",
      "Epoch average log loss:0.03758782513073779\n",
      "In Epoch5, val_loss:0.03798649841787571, best_val_loss:0.037972753504137126, best_auc:0.9901462779772943\n",
      "Epoch:6 Batch:0 Log-loss0.03948407247662544\n",
      "Epoch:6 Batch:80 Log-loss0.02469666115939617\n",
      "Epoch:6 Batch:160 Log-loss0.035204220563173294\n",
      "Epoch:6 Batch:240 Log-loss0.02522438019514084\n",
      "Epoch:6 Batch:320 Log-loss0.05126126483082771\n",
      "Epoch:6 Batch:400 Log-loss0.046738628298044205\n",
      "Epoch:6 Batch:480 Log-loss0.04359649121761322\n",
      "Epoch:6 Batch:560 Log-loss0.02621430903673172\n",
      "Epoch average log loss:0.037204606202431025\n",
      "In Epoch6, val_loss:0.03817640659860094, best_val_loss:0.037972753504137126, best_auc:0.9901462779772943\n",
      "Epoch:7 Batch:0 Log-loss0.028605608269572258\n",
      "Epoch:7 Batch:80 Log-loss0.03567931056022644\n",
      "Epoch:7 Batch:160 Log-loss0.05020475387573242\n",
      "Epoch:7 Batch:240 Log-loss0.036981042474508286\n",
      "Epoch:7 Batch:320 Log-loss0.037290796637535095\n",
      "Epoch:7 Batch:400 Log-loss0.03789384290575981\n",
      "Epoch:7 Batch:480 Log-loss0.035851094871759415\n",
      "Epoch:7 Batch:560 Log-loss0.053937651216983795\n",
      "Epoch average log loss:0.036751866869495387\n",
      "In Epoch7, val_loss:0.03805878490955985, best_val_loss:0.037972753504137126, best_auc:0.9901462779772943\n",
      "Epoch:8 Batch:0 Log-loss0.0654972717165947\n",
      "Epoch:8 Batch:80 Log-loss0.04306827113032341\n",
      "Epoch:8 Batch:160 Log-loss0.043974053114652634\n",
      "Epoch:8 Batch:240 Log-loss0.030701858922839165\n",
      "Epoch:8 Batch:320 Log-loss0.04446418583393097\n",
      "Epoch:8 Batch:400 Log-loss0.045221418142318726\n",
      "Epoch:8 Batch:480 Log-loss0.023224854841828346\n",
      "Epoch:8 Batch:560 Log-loss0.033205147832632065\n",
      "Epoch average log loss:0.03688325347112758\n",
      "In Epoch8, val_loss:0.038429815454681535, best_val_loss:0.037972753504137126, best_auc:0.9901462779772943\n",
      "Epoch:9 Batch:0 Log-loss0.03131287172436714\n",
      "Epoch:9 Batch:80 Log-loss0.0465337373316288\n",
      "Epoch:9 Batch:160 Log-loss0.037149686366319656\n",
      "Epoch:9 Batch:240 Log-loss0.047536805272102356\n",
      "Epoch:9 Batch:320 Log-loss0.039461951702833176\n",
      "Epoch:9 Batch:400 Log-loss0.030075885355472565\n",
      "Epoch:9 Batch:480 Log-loss0.03754733130335808\n",
      "Epoch:9 Batch:560 Log-loss0.05021059885621071\n",
      "Epoch average log loss:0.03662668495172901\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.03788784150779539, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:10 Batch:0 Log-loss0.039094045758247375\n",
      "Epoch:10 Batch:80 Log-loss0.02924962341785431\n",
      "Epoch:10 Batch:160 Log-loss0.041780684143304825\n",
      "Epoch:10 Batch:240 Log-loss0.038855455815792084\n",
      "Epoch:10 Batch:320 Log-loss0.050876617431640625\n",
      "Epoch:10 Batch:400 Log-loss0.035579875111579895\n",
      "Epoch:10 Batch:480 Log-loss0.030676916241645813\n",
      "Epoch:10 Batch:560 Log-loss0.03768166899681091\n",
      "Epoch average log loss:0.03618388861317986\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/zake7749/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1694: RuntimeWarning: divide by zero encountered in log\n",
      "  loss = -(transformed_labels * np.log(y_pred)).sum(axis=1)\n",
      "/home/zake7749/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1694: RuntimeWarning: invalid value encountered in multiply\n",
      "  loss = -(transformed_labels * np.log(y_pred)).sum(axis=1)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In Epoch10, val_loss:nan, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:11 Batch:0 Log-loss0.04602649435400963\n",
      "Epoch:11 Batch:80 Log-loss0.03017587773501873\n",
      "Epoch:11 Batch:160 Log-loss0.03831406682729721\n",
      "Epoch:11 Batch:240 Log-loss0.04492254555225372\n",
      "Epoch:11 Batch:320 Log-loss0.0332631878554821\n",
      "Epoch:11 Batch:400 Log-loss0.03734120354056358\n",
      "Epoch:11 Batch:480 Log-loss0.034766972064971924\n",
      "Epoch:11 Batch:560 Log-loss0.04941846802830696\n",
      "Epoch average log loss:0.036114837320721044\n",
      "In Epoch11, val_loss:nan, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:12 Batch:0 Log-loss0.044260889291763306\n",
      "Epoch:12 Batch:80 Log-loss0.04089733585715294\n",
      "Epoch:12 Batch:160 Log-loss0.04165925458073616\n",
      "Epoch:12 Batch:240 Log-loss0.043058544397354126\n",
      "Epoch:12 Batch:320 Log-loss0.042819809168577194\n",
      "Epoch:12 Batch:400 Log-loss0.04282712936401367\n",
      "Epoch:12 Batch:480 Log-loss0.03803418204188347\n",
      "Epoch:12 Batch:560 Log-loss0.03983001410961151\n",
      "Epoch average log loss:0.03601960337421458\n",
      "In Epoch12, val_loss:0.0383642895676956, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:13 Batch:0 Log-loss0.0323689766228199\n",
      "Epoch:13 Batch:80 Log-loss0.027679460123181343\n",
      "Epoch:13 Batch:160 Log-loss0.03615042567253113\n",
      "Epoch:13 Batch:240 Log-loss0.03660395368933678\n",
      "Epoch:13 Batch:320 Log-loss0.03255987539887428\n",
      "Epoch:13 Batch:400 Log-loss0.02792801521718502\n",
      "Epoch:13 Batch:480 Log-loss0.035374801605939865\n",
      "Epoch:13 Batch:560 Log-loss0.04404008761048317\n",
      "Epoch average log loss:0.03596294669800305\n",
      "In Epoch13, val_loss:nan, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:14 Batch:0 Log-loss0.031621355563402176\n",
      "Epoch:14 Batch:80 Log-loss0.04101204872131348\n",
      "Epoch:14 Batch:160 Log-loss0.03720135986804962\n",
      "Epoch:14 Batch:240 Log-loss0.02798532508313656\n",
      "Epoch:14 Batch:320 Log-loss0.028507068753242493\n",
      "Epoch:14 Batch:400 Log-loss0.027643578127026558\n",
      "Epoch:14 Batch:480 Log-loss0.02605041116476059\n",
      "Epoch:14 Batch:560 Log-loss0.026066619902849197\n",
      "Epoch average log loss:0.03573906817473471\n",
      "In Epoch14, val_loss:nan, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:15 Batch:0 Log-loss0.037041667848825455\n",
      "Epoch:15 Batch:80 Log-loss0.03498713672161102\n",
      "Epoch:15 Batch:160 Log-loss0.03816750645637512\n",
      "Epoch:15 Batch:240 Log-loss0.0332440547645092\n",
      "Epoch:15 Batch:320 Log-loss0.039789799600839615\n",
      "Epoch:15 Batch:400 Log-loss0.030349433422088623\n",
      "Epoch:15 Batch:480 Log-loss0.05339256301522255\n",
      "Epoch:15 Batch:560 Log-loss0.03787291795015335\n",
      "Epoch average log loss:0.03564823657008154\n",
      "In Epoch15, val_loss:nan, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:16 Batch:0 Log-loss0.04183949530124664\n",
      "Epoch:16 Batch:80 Log-loss0.03851412236690521\n",
      "Epoch:16 Batch:160 Log-loss0.023943467065691948\n",
      "Epoch:16 Batch:240 Log-loss0.0380551777780056\n",
      "Epoch:16 Batch:320 Log-loss0.03346620127558708\n",
      "Epoch:16 Batch:400 Log-loss0.02952052839100361\n",
      "Epoch:16 Batch:480 Log-loss0.03975849971175194\n",
      "Epoch:16 Batch:560 Log-loss0.030453354120254517\n",
      "Epoch average log loss:0.035400569212755986\n",
      "In Epoch16, val_loss:nan, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:17 Batch:0 Log-loss0.03406311944127083\n",
      "Epoch:17 Batch:80 Log-loss0.032394666224718094\n",
      "Epoch:17 Batch:160 Log-loss0.0392896942794323\n",
      "Epoch:17 Batch:240 Log-loss0.03713761642575264\n",
      "Epoch:17 Batch:320 Log-loss0.04439886286854744\n",
      "Epoch:17 Batch:400 Log-loss0.032256875187158585\n",
      "Epoch:17 Batch:480 Log-loss0.02470305562019348\n",
      "Epoch:17 Batch:560 Log-loss0.027178576216101646\n",
      "Epoch average log loss:0.03525096291476595\n",
      "In Epoch17, val_loss:nan, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:18 Batch:0 Log-loss0.03606104850769043\n",
      "Epoch:18 Batch:80 Log-loss0.03492378816008568\n",
      "Epoch:18 Batch:160 Log-loss0.04463754966855049\n",
      "Epoch:18 Batch:240 Log-loss0.03170472010970116\n",
      "Epoch:18 Batch:320 Log-loss0.028683258220553398\n",
      "Epoch:18 Batch:400 Log-loss0.029256513342261314\n",
      "Epoch:18 Batch:480 Log-loss0.027254223823547363\n",
      "Epoch:18 Batch:560 Log-loss0.03038468398153782\n",
      "Epoch average log loss:0.03506272526657475\n",
      "In Epoch18, val_loss:nan, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:19 Batch:0 Log-loss0.05828293785452843\n",
      "Epoch:19 Batch:80 Log-loss0.04288588464260101\n",
      "Epoch:19 Batch:160 Log-loss0.028858615085482597\n",
      "Epoch:19 Batch:240 Log-loss0.03452349081635475\n",
      "Epoch:19 Batch:320 Log-loss0.03347210958600044\n",
      "Epoch:19 Batch:400 Log-loss0.034320030361413956\n",
      "Epoch:19 Batch:480 Log-loss0.0326056107878685\n",
      "Epoch:19 Batch:560 Log-loss0.02573421038687229\n",
      "Epoch average log loss:0.035161723625580114\n",
      "In Epoch19, val_loss:0.03897861511933428, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:20 Batch:0 Log-loss0.032761458307504654\n",
      "Epoch:20 Batch:80 Log-loss0.01735646091401577\n",
      "Epoch:20 Batch:160 Log-loss0.03317594900727272\n",
      "Epoch:20 Batch:240 Log-loss0.02604081481695175\n",
      "Epoch:20 Batch:320 Log-loss0.03362560644745827\n",
      "Epoch:20 Batch:400 Log-loss0.0348038449883461\n",
      "Epoch:20 Batch:480 Log-loss0.03118257410824299\n",
      "Epoch:20 Batch:560 Log-loss0.036912526935338974\n",
      "Epoch average log loss:0.035149704931037765\n",
      "In Epoch20, val_loss:0.03854015422286456, best_val_loss:0.03788784150779539, best_auc:0.990084701347369\n",
      "Epoch:21 Batch:0 Log-loss0.027844125404953957\n",
      "Epoch:21 Batch:80 Log-loss0.03306746110320091\n",
      "Epoch:21 Batch:160 Log-loss0.04081331938505173\n",
      "Epoch:21 Batch:240 Log-loss0.03292566165328026\n",
      "Epoch:21 Batch:320 Log-loss0.05485758185386658\n",
      "Epoch:21 Batch:400 Log-loss0.032975081354379654\n",
      "Epoch:21 Batch:480 Log-loss0.04571808874607086\n",
      "Epoch:21 Batch:560 Log-loss0.042023856192827225\n",
      "Epoch average log loss:0.03472102863181915\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_644.pt.\n",
      "\n",
      "Training on fold 5\n",
      "Epoch:1 Batch:0 Log-loss0.027776164934039116\n",
      "Epoch:1 Batch:80 Log-loss0.025543412193655968\n",
      "Epoch:1 Batch:160 Log-loss0.030007148161530495\n",
      "Epoch:1 Batch:240 Log-loss0.020555386319756508\n",
      "Epoch:1 Batch:320 Log-loss0.055155832320451736\n",
      "Epoch:1 Batch:400 Log-loss0.030342763289809227\n",
      "Epoch:1 Batch:480 Log-loss0.04210066422820091\n",
      "Epoch:1 Batch:560 Log-loss0.02244664542376995\n",
      "Epoch average log loss:0.03545875013639618\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.038514724888425465, best_val_loss:0.038514724888425465, best_auc:0.9902369841943298\n",
      "Epoch:2 Batch:0 Log-loss0.049923282116651535\n",
      "Epoch:2 Batch:80 Log-loss0.04532671347260475\n",
      "Epoch:2 Batch:160 Log-loss0.028566768392920494\n",
      "Epoch:2 Batch:240 Log-loss0.04306675121188164\n",
      "Epoch:2 Batch:320 Log-loss0.043991949409246445\n",
      "Epoch:2 Batch:400 Log-loss0.04308345913887024\n",
      "Epoch:2 Batch:480 Log-loss0.04414394497871399\n",
      "Epoch:2 Batch:560 Log-loss0.03686550259590149\n",
      "Epoch average log loss:0.03921486120172111\n",
      "In Epoch2, val_loss:0.03889626043923513, best_val_loss:0.038514724888425465, best_auc:0.9902369841943298\n",
      "Epoch:3 Batch:0 Log-loss0.04336472228169441\n",
      "Epoch:3 Batch:80 Log-loss0.037020716816186905\n",
      "Epoch:3 Batch:160 Log-loss0.05121530964970589\n",
      "Epoch:3 Batch:240 Log-loss0.044884685426950455\n",
      "Epoch:3 Batch:320 Log-loss0.04022479057312012\n",
      "Epoch:3 Batch:400 Log-loss0.043699946254491806\n",
      "Epoch:3 Batch:480 Log-loss0.038734760135412216\n",
      "Epoch:3 Batch:560 Log-loss0.051900558173656464\n",
      "Epoch average log loss:0.03853066601524396\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.03775466213959252, best_val_loss:0.03775466213959252, best_auc:0.99036589680514\n",
      "Epoch:4 Batch:0 Log-loss0.02524293214082718\n",
      "Epoch:4 Batch:80 Log-loss0.04575403407216072\n",
      "Epoch:4 Batch:160 Log-loss0.06310447305440903\n",
      "Epoch:4 Batch:240 Log-loss0.03445322439074516\n",
      "Epoch:4 Batch:320 Log-loss0.05049453303217888\n",
      "Epoch:4 Batch:400 Log-loss0.03919060155749321\n",
      "Epoch:4 Batch:480 Log-loss0.039878960698843\n",
      "Epoch:4 Batch:560 Log-loss0.03181666508316994\n",
      "Epoch average log loss:0.03818224970184799\n",
      "In Epoch4, val_loss:0.037775225046277595, best_val_loss:0.03775466213959252, best_auc:0.99036589680514\n",
      "Epoch:5 Batch:0 Log-loss0.03245433419942856\n",
      "Epoch:5 Batch:80 Log-loss0.024397941306233406\n",
      "Epoch:5 Batch:160 Log-loss0.05207344889640808\n",
      "Epoch:5 Batch:240 Log-loss0.036910515278577805\n",
      "Epoch:5 Batch:320 Log-loss0.04811998829245567\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:5 Batch:400 Log-loss0.030093243345618248\n",
      "Epoch:5 Batch:480 Log-loss0.018039489164948463\n",
      "Epoch:5 Batch:560 Log-loss0.04151514172554016\n",
      "Epoch average log loss:0.03797521377314946\n",
      "In Epoch5, val_loss:0.03819615840981692, best_val_loss:0.03775466213959252, best_auc:0.99036589680514\n",
      "Epoch:6 Batch:0 Log-loss0.02336401306092739\n",
      "Epoch:6 Batch:80 Log-loss0.04573683440685272\n",
      "Epoch:6 Batch:160 Log-loss0.04076691344380379\n",
      "Epoch:6 Batch:240 Log-loss0.04391122981905937\n",
      "Epoch:6 Batch:320 Log-loss0.044730644673109055\n",
      "Epoch:6 Batch:400 Log-loss0.0320717953145504\n",
      "Epoch:6 Batch:480 Log-loss0.037110552191734314\n",
      "Epoch:6 Batch:560 Log-loss0.028581203892827034\n",
      "Epoch average log loss:0.03761226306004184\n",
      "In Epoch6, val_loss:0.038166955782558146, best_val_loss:0.03775466213959252, best_auc:0.99036589680514\n",
      "Epoch:7 Batch:0 Log-loss0.03393092006444931\n",
      "Epoch:7 Batch:80 Log-loss0.033407505601644516\n",
      "Epoch:7 Batch:160 Log-loss0.049434076994657516\n",
      "Epoch:7 Batch:240 Log-loss0.02716623991727829\n",
      "Epoch:7 Batch:320 Log-loss0.03352135792374611\n",
      "Epoch:7 Batch:400 Log-loss0.03979310765862465\n",
      "Epoch:7 Batch:480 Log-loss0.03929677605628967\n",
      "Epoch:7 Batch:560 Log-loss0.028248393908143044\n",
      "Epoch average log loss:0.03725933112270598\n",
      "In Epoch7, val_loss:0.0380024717031771, best_val_loss:0.03775466213959252, best_auc:0.99036589680514\n",
      "Epoch:8 Batch:0 Log-loss0.03642166405916214\n",
      "Epoch:8 Batch:80 Log-loss0.03606632351875305\n",
      "Epoch:8 Batch:160 Log-loss0.0368802510201931\n",
      "Epoch:8 Batch:240 Log-loss0.029714256525039673\n",
      "Epoch:8 Batch:320 Log-loss0.03137939050793648\n",
      "Epoch:8 Batch:400 Log-loss0.040452662855386734\n",
      "Epoch:8 Batch:480 Log-loss0.03336815908551216\n",
      "Epoch:8 Batch:560 Log-loss0.034810569137334824\n",
      "Epoch average log loss:0.03729425824380347\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch8, val_loss:0.03735007805821661, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:9 Batch:0 Log-loss0.034935902804136276\n",
      "Epoch:9 Batch:80 Log-loss0.04697777330875397\n",
      "Epoch:9 Batch:160 Log-loss0.03275592625141144\n",
      "Epoch:9 Batch:240 Log-loss0.044527318328619\n",
      "Epoch:9 Batch:320 Log-loss0.035998500883579254\n",
      "Epoch:9 Batch:400 Log-loss0.04124021902680397\n",
      "Epoch:9 Batch:480 Log-loss0.03286362439393997\n",
      "Epoch:9 Batch:560 Log-loss0.04471774771809578\n",
      "Epoch average log loss:0.03684330206929839\n",
      "In Epoch9, val_loss:0.03788186136487015, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:10 Batch:0 Log-loss0.03048037551343441\n",
      "Epoch:10 Batch:80 Log-loss0.04640297591686249\n",
      "Epoch:10 Batch:160 Log-loss0.03633590415120125\n",
      "Epoch:10 Batch:240 Log-loss0.024393053725361824\n",
      "Epoch:10 Batch:320 Log-loss0.030322464182972908\n",
      "Epoch:10 Batch:400 Log-loss0.054054271429777145\n",
      "Epoch:10 Batch:480 Log-loss0.038462862372398376\n",
      "Epoch:10 Batch:560 Log-loss0.049771472811698914\n",
      "Epoch average log loss:0.03660152692879949\n",
      "In Epoch10, val_loss:0.037682144896100556, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:11 Batch:0 Log-loss0.030376365408301353\n",
      "Epoch:11 Batch:80 Log-loss0.04455322399735451\n",
      "Epoch:11 Batch:160 Log-loss0.042758259922266006\n",
      "Epoch:11 Batch:240 Log-loss0.03523153439164162\n",
      "Epoch:11 Batch:320 Log-loss0.03667489066720009\n",
      "Epoch:11 Batch:400 Log-loss0.02760455757379532\n",
      "Epoch:11 Batch:480 Log-loss0.03204086795449257\n",
      "Epoch:11 Batch:560 Log-loss0.02860497497022152\n",
      "Epoch average log loss:0.03645132256206125\n",
      "In Epoch11, val_loss:0.037740545861838136, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:12 Batch:0 Log-loss0.03636792674660683\n",
      "Epoch:12 Batch:80 Log-loss0.040545541793107986\n",
      "Epoch:12 Batch:160 Log-loss0.02541634626686573\n",
      "Epoch:12 Batch:240 Log-loss0.040028538554906845\n",
      "Epoch:12 Batch:320 Log-loss0.05054611340165138\n",
      "Epoch:12 Batch:400 Log-loss0.051725346595048904\n",
      "Epoch:12 Batch:480 Log-loss0.037352342158555984\n",
      "Epoch:12 Batch:560 Log-loss0.041176438331604004\n",
      "Epoch average log loss:0.0363370576158299\n",
      "In Epoch12, val_loss:0.037830726390851895, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:13 Batch:0 Log-loss0.04364998638629913\n",
      "Epoch:13 Batch:80 Log-loss0.03279808536171913\n",
      "Epoch:13 Batch:160 Log-loss0.03362492844462395\n",
      "Epoch:13 Batch:240 Log-loss0.05291716754436493\n",
      "Epoch:13 Batch:320 Log-loss0.05451872572302818\n",
      "Epoch:13 Batch:400 Log-loss0.03907907381653786\n",
      "Epoch:13 Batch:480 Log-loss0.04480894282460213\n",
      "Epoch:13 Batch:560 Log-loss0.033925995230674744\n",
      "Epoch average log loss:0.03604786200781486\n",
      "In Epoch13, val_loss:0.03801768189347907, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:14 Batch:0 Log-loss0.045754533261060715\n",
      "Epoch:14 Batch:80 Log-loss0.035136137157678604\n",
      "Epoch:14 Batch:160 Log-loss0.03336283192038536\n",
      "Epoch:14 Batch:240 Log-loss0.03295069932937622\n",
      "Epoch:14 Batch:320 Log-loss0.028052357956767082\n",
      "Epoch:14 Batch:400 Log-loss0.04356221482157707\n",
      "Epoch:14 Batch:480 Log-loss0.045577604323625565\n",
      "Epoch:14 Batch:560 Log-loss0.02760123461484909\n",
      "Epoch average log loss:0.03601144682948611\n",
      "In Epoch14, val_loss:0.037628714522748276, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:15 Batch:0 Log-loss0.03755142539739609\n",
      "Epoch:15 Batch:80 Log-loss0.037362564355134964\n",
      "Epoch:15 Batch:160 Log-loss0.031456101685762405\n",
      "Epoch:15 Batch:240 Log-loss0.044027794152498245\n",
      "Epoch:15 Batch:320 Log-loss0.039101749658584595\n",
      "Epoch:15 Batch:400 Log-loss0.04053801670670509\n",
      "Epoch:15 Batch:480 Log-loss0.03443256765604019\n",
      "Epoch:15 Batch:560 Log-loss0.022086599841713905\n",
      "Epoch average log loss:0.03591556968832654\n",
      "In Epoch15, val_loss:0.03757274472773122, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:16 Batch:0 Log-loss0.027440277859568596\n",
      "Epoch:16 Batch:80 Log-loss0.0401613675057888\n",
      "Epoch:16 Batch:160 Log-loss0.04179481044411659\n",
      "Epoch:16 Batch:240 Log-loss0.03931271657347679\n",
      "Epoch:16 Batch:320 Log-loss0.0276130810379982\n",
      "Epoch:16 Batch:400 Log-loss0.024150624871253967\n",
      "Epoch:16 Batch:480 Log-loss0.04114317521452904\n",
      "Epoch:16 Batch:560 Log-loss0.058628153055906296\n",
      "Epoch average log loss:0.035750166875576334\n",
      "In Epoch16, val_loss:0.037902737552503944, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:17 Batch:0 Log-loss0.036102697253227234\n",
      "Epoch:17 Batch:80 Log-loss0.026854239404201508\n",
      "Epoch:17 Batch:160 Log-loss0.04641719162464142\n",
      "Epoch:17 Batch:240 Log-loss0.040792156010866165\n",
      "Epoch:17 Batch:320 Log-loss0.03619476035237312\n",
      "Epoch:17 Batch:400 Log-loss0.02099546045064926\n",
      "Epoch:17 Batch:480 Log-loss0.030593520030379295\n",
      "Epoch:17 Batch:560 Log-loss0.03725666180253029\n",
      "Epoch average log loss:0.03560289092960634\n",
      "In Epoch17, val_loss:0.03786403294320669, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:18 Batch:0 Log-loss0.02904563955962658\n",
      "Epoch:18 Batch:80 Log-loss0.024980908259749413\n",
      "Epoch:18 Batch:160 Log-loss0.031612787395715714\n",
      "Epoch:18 Batch:240 Log-loss0.03756066784262657\n",
      "Epoch:18 Batch:320 Log-loss0.02659495361149311\n",
      "Epoch:18 Batch:400 Log-loss0.04102637618780136\n",
      "Epoch:18 Batch:480 Log-loss0.03894906863570213\n",
      "Epoch:18 Batch:560 Log-loss0.03825173154473305\n",
      "Epoch average log loss:0.035350331368057854\n",
      "In Epoch18, val_loss:0.03764366119357328, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:19 Batch:0 Log-loss0.0464901477098465\n",
      "Epoch:19 Batch:80 Log-loss0.03677889704704285\n",
      "Epoch:19 Batch:160 Log-loss0.02987760305404663\n",
      "Epoch:19 Batch:240 Log-loss0.04379424825310707\n",
      "Epoch:19 Batch:320 Log-loss0.04994272068142891\n",
      "Epoch:19 Batch:400 Log-loss0.0330217108130455\n",
      "Epoch:19 Batch:480 Log-loss0.036431875079870224\n",
      "Epoch:19 Batch:560 Log-loss0.038968540728092194\n",
      "Epoch average log loss:0.035369868781084994\n",
      "In Epoch19, val_loss:0.038155554854553776, best_val_loss:0.03735007805821661, best_auc:0.9905330197351695\n",
      "Epoch:20 Batch:0 Log-loss0.039566922932863235\n",
      "Epoch:20 Batch:80 Log-loss0.03678261861205101\n",
      "Epoch:20 Batch:160 Log-loss0.03218292072415352\n",
      "Epoch:20 Batch:240 Log-loss0.03364696726202965\n",
      "Epoch:20 Batch:320 Log-loss0.03934815153479576\n",
      "Epoch:20 Batch:400 Log-loss0.027195652946829796\n",
      "Epoch:20 Batch:480 Log-loss0.03600214794278145\n",
      "Epoch:20 Batch:560 Log-loss0.029351796954870224\n",
      "Epoch average log loss:0.03533217134952013\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_645.pt.\n",
      "\n",
      "Training on fold 6\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:1 Batch:0 Log-loss0.05207217112183571\n",
      "Epoch:1 Batch:80 Log-loss0.03845914080739021\n",
      "Epoch:1 Batch:160 Log-loss0.035605866461992264\n",
      "Epoch:1 Batch:240 Log-loss0.04167464002966881\n",
      "Epoch:1 Batch:320 Log-loss0.052951741963624954\n",
      "Epoch:1 Batch:400 Log-loss0.03308295086026192\n",
      "Epoch:1 Batch:480 Log-loss0.02545510232448578\n",
      "Epoch:1 Batch:560 Log-loss0.026346463710069656\n",
      "Epoch average log loss:0.03555475313748632\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.03954641780618133, best_val_loss:0.03954641780618133, best_auc:0.9896234366979048\n",
      "Epoch:2 Batch:0 Log-loss0.04751436412334442\n",
      "Epoch:2 Batch:80 Log-loss0.030747773125767708\n",
      "Epoch:2 Batch:160 Log-loss0.027692407369613647\n",
      "Epoch:2 Batch:240 Log-loss0.03106490522623062\n",
      "Epoch:2 Batch:320 Log-loss0.03448696807026863\n",
      "Epoch:2 Batch:400 Log-loss0.03421015664935112\n",
      "Epoch:2 Batch:480 Log-loss0.033812977373600006\n",
      "Epoch:2 Batch:560 Log-loss0.04125376418232918\n",
      "Epoch average log loss:0.039165138829100346\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.038015952095756624, best_val_loss:0.038015952095756624, best_auc:0.9898251781333262\n",
      "Epoch:3 Batch:0 Log-loss0.03238809481263161\n",
      "Epoch:3 Batch:80 Log-loss0.03574803099036217\n",
      "Epoch:3 Batch:160 Log-loss0.06373708695173264\n",
      "Epoch:3 Batch:240 Log-loss0.029484866186976433\n",
      "Epoch:3 Batch:320 Log-loss0.030245589092373848\n",
      "Epoch:3 Batch:400 Log-loss0.048609185963869095\n",
      "Epoch:3 Batch:480 Log-loss0.028385542333126068\n",
      "Epoch:3 Batch:560 Log-loss0.035502925515174866\n",
      "Epoch average log loss:0.038305223201002395\n",
      "In Epoch3, val_loss:0.0393088158907652, best_val_loss:0.038015952095756624, best_auc:0.9898251781333262\n",
      "Epoch:4 Batch:0 Log-loss0.03685513883829117\n",
      "Epoch:4 Batch:80 Log-loss0.051338206976652145\n",
      "Epoch:4 Batch:160 Log-loss0.036551810801029205\n",
      "Epoch:4 Batch:240 Log-loss0.017386773601174355\n",
      "Epoch:4 Batch:320 Log-loss0.03170201927423477\n",
      "Epoch:4 Batch:400 Log-loss0.04366162046790123\n",
      "Epoch:4 Batch:480 Log-loss0.03755616396665573\n",
      "Epoch:4 Batch:560 Log-loss0.03187774866819382\n",
      "Epoch average log loss:0.03785542780533433\n",
      "In Epoch4, val_loss:0.0385055659090058, best_val_loss:0.038015952095756624, best_auc:0.9898251781333262\n",
      "Epoch:5 Batch:0 Log-loss0.03119986318051815\n",
      "Epoch:5 Batch:80 Log-loss0.03540634736418724\n",
      "Epoch:5 Batch:160 Log-loss0.03142485022544861\n",
      "Epoch:5 Batch:240 Log-loss0.04134673625230789\n",
      "Epoch:5 Batch:320 Log-loss0.043356191366910934\n",
      "Epoch:5 Batch:400 Log-loss0.03153924271464348\n",
      "Epoch:5 Batch:480 Log-loss0.04513835534453392\n",
      "Epoch:5 Batch:560 Log-loss0.04288198798894882\n",
      "Epoch average log loss:0.03772204195847735\n",
      "In Epoch5, val_loss:0.03805723452541319, best_val_loss:0.038015952095756624, best_auc:0.9898251781333262\n",
      "Epoch:6 Batch:0 Log-loss0.042926888912916183\n",
      "Epoch:6 Batch:80 Log-loss0.026024440303444862\n",
      "Epoch:6 Batch:160 Log-loss0.04812707006931305\n",
      "Epoch:6 Batch:240 Log-loss0.05892042815685272\n",
      "Epoch:6 Batch:320 Log-loss0.04093048721551895\n",
      "Epoch:6 Batch:400 Log-loss0.03673272952437401\n",
      "Epoch:6 Batch:480 Log-loss0.041222214698791504\n",
      "Epoch:6 Batch:560 Log-loss0.038335639983415604\n",
      "Epoch average log loss:0.037375043498884354\n",
      "In Epoch6, val_loss:0.03833932243053037, best_val_loss:0.038015952095756624, best_auc:0.9898251781333262\n",
      "Epoch:7 Batch:0 Log-loss0.03944830223917961\n",
      "Epoch:7 Batch:80 Log-loss0.024116309359669685\n",
      "Epoch:7 Batch:160 Log-loss0.04668441042304039\n",
      "Epoch:7 Batch:240 Log-loss0.04153451696038246\n",
      "Epoch:7 Batch:320 Log-loss0.03495209291577339\n",
      "Epoch:7 Batch:400 Log-loss0.03979625925421715\n",
      "Epoch:7 Batch:480 Log-loss0.039176929742097855\n",
      "Epoch:7 Batch:560 Log-loss0.045309148728847504\n",
      "Epoch average log loss:0.03723689351629998\n",
      "In Epoch7, val_loss:0.03873484674753414, best_val_loss:0.038015952095756624, best_auc:0.9898251781333262\n",
      "Epoch:8 Batch:0 Log-loss0.033207181841135025\n",
      "Epoch:8 Batch:80 Log-loss0.023422231897711754\n",
      "Epoch:8 Batch:160 Log-loss0.03605421259999275\n",
      "Epoch:8 Batch:240 Log-loss0.03728464990854263\n",
      "Epoch:8 Batch:320 Log-loss0.030615441501140594\n",
      "Epoch:8 Batch:400 Log-loss0.0418345183134079\n",
      "Epoch:8 Batch:480 Log-loss0.03596286103129387\n",
      "Epoch:8 Batch:560 Log-loss0.03916940838098526\n",
      "Epoch average log loss:0.037032156667139916\n",
      "In Epoch8, val_loss:0.03845395137607019, best_val_loss:0.038015952095756624, best_auc:0.9898251781333262\n",
      "Epoch:9 Batch:0 Log-loss0.02992979623377323\n",
      "Epoch:9 Batch:80 Log-loss0.04375095292925835\n",
      "Epoch:9 Batch:160 Log-loss0.02899547852575779\n",
      "Epoch:9 Batch:240 Log-loss0.028223685920238495\n",
      "Epoch:9 Batch:320 Log-loss0.036418329924345016\n",
      "Epoch:9 Batch:400 Log-loss0.029361678287386894\n",
      "Epoch:9 Batch:480 Log-loss0.01926790550351143\n",
      "Epoch:9 Batch:560 Log-loss0.03003719076514244\n",
      "Epoch average log loss:0.03675589538179338\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch9, val_loss:0.0379442347536952, best_val_loss:0.0379442347536952, best_auc:0.990182612600166\n",
      "Epoch:10 Batch:0 Log-loss0.032210931181907654\n",
      "Epoch:10 Batch:80 Log-loss0.03916360065340996\n",
      "Epoch:10 Batch:160 Log-loss0.02302675135433674\n",
      "Epoch:10 Batch:240 Log-loss0.03169663995504379\n",
      "Epoch:10 Batch:320 Log-loss0.04150436818599701\n",
      "Epoch:10 Batch:400 Log-loss0.038665130734443665\n",
      "Epoch:10 Batch:480 Log-loss0.027296824380755424\n",
      "Epoch:10 Batch:560 Log-loss0.03982679173350334\n",
      "Epoch average log loss:0.036425489829187946\n",
      "In Epoch10, val_loss:0.03830530895671046, best_val_loss:0.0379442347536952, best_auc:0.990182612600166\n",
      "Epoch:11 Batch:0 Log-loss0.021877245977520943\n",
      "Epoch:11 Batch:80 Log-loss0.04228295758366585\n",
      "Epoch:11 Batch:160 Log-loss0.04653821513056755\n",
      "Epoch:11 Batch:240 Log-loss0.03611621633172035\n",
      "Epoch:11 Batch:320 Log-loss0.02401808649301529\n",
      "Epoch:11 Batch:400 Log-loss0.02652093581855297\n",
      "Epoch:11 Batch:480 Log-loss0.032027099281549454\n",
      "Epoch:11 Batch:560 Log-loss0.028302019461989403\n",
      "Epoch average log loss:0.036356503371748006\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch11, val_loss:0.03782607360579734, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:12 Batch:0 Log-loss0.042187467217445374\n",
      "Epoch:12 Batch:80 Log-loss0.026153378188610077\n",
      "Epoch:12 Batch:160 Log-loss0.03633743152022362\n",
      "Epoch:12 Batch:240 Log-loss0.03163761645555496\n",
      "Epoch:12 Batch:320 Log-loss0.030791983008384705\n",
      "Epoch:12 Batch:400 Log-loss0.034969959408044815\n",
      "Epoch:12 Batch:480 Log-loss0.03115115314722061\n",
      "Epoch:12 Batch:560 Log-loss0.024327117949724197\n",
      "Epoch average log loss:0.03597216296974304\n",
      "In Epoch12, val_loss:0.03798521262805602, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:13 Batch:0 Log-loss0.04406728222966194\n",
      "Epoch:13 Batch:80 Log-loss0.025689100846648216\n",
      "Epoch:13 Batch:160 Log-loss0.026115072891116142\n",
      "Epoch:13 Batch:240 Log-loss0.03589160367846489\n",
      "Epoch:13 Batch:320 Log-loss0.045891184359788895\n",
      "Epoch:13 Batch:400 Log-loss0.032528527081012726\n",
      "Epoch:13 Batch:480 Log-loss0.028596311807632446\n",
      "Epoch:13 Batch:560 Log-loss0.045493535697460175\n",
      "Epoch average log loss:0.03587581589818001\n",
      "In Epoch13, val_loss:0.03854812985070886, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:14 Batch:0 Log-loss0.04365922138094902\n",
      "Epoch:14 Batch:80 Log-loss0.04031050577759743\n",
      "Epoch:14 Batch:160 Log-loss0.02744915336370468\n",
      "Epoch:14 Batch:240 Log-loss0.04794668033719063\n",
      "Epoch:14 Batch:320 Log-loss0.034204062074422836\n",
      "Epoch:14 Batch:400 Log-loss0.026057668030261993\n",
      "Epoch:14 Batch:480 Log-loss0.03551222011446953\n",
      "Epoch:14 Batch:560 Log-loss0.04752814397215843\n",
      "Epoch average log loss:0.03593318335312818\n",
      "In Epoch14, val_loss:0.03885138705767344, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:15 Batch:0 Log-loss0.028089866042137146\n",
      "Epoch:15 Batch:80 Log-loss0.0411536768078804\n",
      "Epoch:15 Batch:160 Log-loss0.031411558389663696\n",
      "Epoch:15 Batch:240 Log-loss0.035175666213035583\n",
      "Epoch:15 Batch:320 Log-loss0.02826409973204136\n",
      "Epoch:15 Batch:400 Log-loss0.04220425710082054\n",
      "Epoch:15 Batch:480 Log-loss0.05470307171344757\n",
      "Epoch:15 Batch:560 Log-loss0.028357598930597305\n",
      "Epoch average log loss:0.03565845471208117\n",
      "In Epoch15, val_loss:0.03835160391040795, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:16 Batch:0 Log-loss0.03457709401845932\n",
      "Epoch:16 Batch:80 Log-loss0.03833502531051636\n",
      "Epoch:16 Batch:160 Log-loss0.05324616655707359\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:16 Batch:240 Log-loss0.03371323272585869\n",
      "Epoch:16 Batch:320 Log-loss0.04443761706352234\n",
      "Epoch:16 Batch:400 Log-loss0.02794976532459259\n",
      "Epoch:16 Batch:480 Log-loss0.028991661965847015\n",
      "Epoch:16 Batch:560 Log-loss0.022236615419387817\n",
      "Epoch average log loss:0.035471910278179815\n",
      "In Epoch16, val_loss:0.038446196350147344, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:17 Batch:0 Log-loss0.03249521926045418\n",
      "Epoch:17 Batch:80 Log-loss0.03803713619709015\n",
      "Epoch:17 Batch:160 Log-loss0.03942335024476051\n",
      "Epoch:17 Batch:240 Log-loss0.04239571467041969\n",
      "Epoch:17 Batch:320 Log-loss0.04388946667313576\n",
      "Epoch:17 Batch:400 Log-loss0.028058985248208046\n",
      "Epoch:17 Batch:480 Log-loss0.041725043207407\n",
      "Epoch:17 Batch:560 Log-loss0.04402457922697067\n",
      "Epoch average log loss:0.03555354495931949\n",
      "In Epoch17, val_loss:0.038251327148832816, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:18 Batch:0 Log-loss0.0230376198887825\n",
      "Epoch:18 Batch:80 Log-loss0.029103966429829597\n",
      "Epoch:18 Batch:160 Log-loss0.042058128863573074\n",
      "Epoch:18 Batch:240 Log-loss0.04272277280688286\n",
      "Epoch:18 Batch:320 Log-loss0.02812904864549637\n",
      "Epoch:18 Batch:400 Log-loss0.031271517276763916\n",
      "Epoch:18 Batch:480 Log-loss0.03569360077381134\n",
      "Epoch:18 Batch:560 Log-loss0.017574109137058258\n",
      "Epoch average log loss:0.03518398333939591\n",
      "In Epoch18, val_loss:0.039017177810013015, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:19 Batch:0 Log-loss0.03911616653203964\n",
      "Epoch:19 Batch:80 Log-loss0.033615678548812866\n",
      "Epoch:19 Batch:160 Log-loss0.03729037940502167\n",
      "Epoch:19 Batch:240 Log-loss0.0446944497525692\n",
      "Epoch:19 Batch:320 Log-loss0.02370043657720089\n",
      "Epoch:19 Batch:400 Log-loss0.04073096066713333\n",
      "Epoch:19 Batch:480 Log-loss0.027325330302119255\n",
      "Epoch:19 Batch:560 Log-loss0.022741353139281273\n",
      "Epoch average log loss:0.03509540160718773\n",
      "In Epoch19, val_loss:0.03849106694561039, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:20 Batch:0 Log-loss0.0309449415653944\n",
      "Epoch:20 Batch:80 Log-loss0.033044133335351944\n",
      "Epoch:20 Batch:160 Log-loss0.035708580166101456\n",
      "Epoch:20 Batch:240 Log-loss0.040661025792360306\n",
      "Epoch:20 Batch:320 Log-loss0.037188123911619186\n",
      "Epoch:20 Batch:400 Log-loss0.031092235818505287\n",
      "Epoch:20 Batch:480 Log-loss0.03431392088532448\n",
      "Epoch:20 Batch:560 Log-loss0.029813459143042564\n",
      "Epoch average log loss:0.03529593396399702\n",
      "In Epoch20, val_loss:0.03837749942339095, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:21 Batch:0 Log-loss0.022513652220368385\n",
      "Epoch:21 Batch:80 Log-loss0.019735265523195267\n",
      "Epoch:21 Batch:160 Log-loss0.03296039253473282\n",
      "Epoch:21 Batch:240 Log-loss0.029489533975720406\n",
      "Epoch:21 Batch:320 Log-loss0.04437538981437683\n",
      "Epoch:21 Batch:400 Log-loss0.02907736785709858\n",
      "Epoch:21 Batch:480 Log-loss0.046617720276117325\n",
      "Epoch:21 Batch:560 Log-loss0.041716139763593674\n",
      "Epoch average log loss:0.03487958811955261\n",
      "In Epoch21, val_loss:0.03805522894690755, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:22 Batch:0 Log-loss0.02865346521139145\n",
      "Epoch:22 Batch:80 Log-loss0.019195234403014183\n",
      "Epoch:22 Batch:160 Log-loss0.036288123577833176\n",
      "Epoch:22 Batch:240 Log-loss0.028558187186717987\n",
      "Epoch:22 Batch:320 Log-loss0.03736360743641853\n",
      "Epoch:22 Batch:400 Log-loss0.044908832758665085\n",
      "Epoch:22 Batch:480 Log-loss0.029483407735824585\n",
      "Epoch:22 Batch:560 Log-loss0.04548387601971626\n",
      "Epoch average log loss:0.035019901155361105\n",
      "In Epoch22, val_loss:0.038323679180662734, best_val_loss:0.03782607360579734, best_auc:0.9902244769918855\n",
      "Epoch:23 Batch:0 Log-loss0.030534783378243446\n",
      "Epoch:23 Batch:80 Log-loss0.02358369342982769\n",
      "Epoch:23 Batch:160 Log-loss0.03061174601316452\n",
      "Epoch:23 Batch:240 Log-loss0.042998190969228745\n",
      "Epoch:23 Batch:320 Log-loss0.03400757536292076\n",
      "Epoch:23 Batch:400 Log-loss0.03612194210290909\n",
      "Epoch:23 Batch:480 Log-loss0.026529597118496895\n",
      "Epoch:23 Batch:560 Log-loss0.033183760941028595\n",
      "Epoch average log loss:0.03487670691351273\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_646.pt.\n",
      "\n",
      "Training on fold 7\n",
      "Epoch:1 Batch:0 Log-loss0.03505679592490196\n",
      "Epoch:1 Batch:80 Log-loss0.032049763947725296\n",
      "Epoch:1 Batch:160 Log-loss0.03396584093570709\n",
      "Epoch:1 Batch:240 Log-loss0.03287816420197487\n",
      "Epoch:1 Batch:320 Log-loss0.035539306700229645\n",
      "Epoch:1 Batch:400 Log-loss0.039999861270189285\n",
      "Epoch:1 Batch:480 Log-loss0.03834345564246178\n",
      "Epoch:1 Batch:560 Log-loss0.04080106317996979\n",
      "Epoch average log loss:0.035133160090273514\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.040117405356634876, best_val_loss:0.040117405356634876, best_auc:0.9883087396519182\n",
      "Epoch:2 Batch:0 Log-loss0.033712077885866165\n",
      "Epoch:2 Batch:80 Log-loss0.03542940318584442\n",
      "Epoch:2 Batch:160 Log-loss0.0317981131374836\n",
      "Epoch:2 Batch:240 Log-loss0.05097554624080658\n",
      "Epoch:2 Batch:320 Log-loss0.053007762879133224\n",
      "Epoch:2 Batch:400 Log-loss0.027371356263756752\n",
      "Epoch:2 Batch:480 Log-loss0.03520480543375015\n",
      "Epoch:2 Batch:560 Log-loss0.033259060233831406\n",
      "Epoch average log loss:0.03846364584086197\n",
      "In Epoch2, val_loss:0.04029530066437918, best_val_loss:0.040117405356634876, best_auc:0.9883087396519182\n",
      "Epoch:3 Batch:0 Log-loss0.035082001239061356\n",
      "Epoch:3 Batch:80 Log-loss0.04362929239869118\n",
      "Epoch:3 Batch:160 Log-loss0.04022618755698204\n",
      "Epoch:3 Batch:240 Log-loss0.022534407675266266\n",
      "Epoch:3 Batch:320 Log-loss0.042121876031160355\n",
      "Epoch:3 Batch:400 Log-loss0.028911100700497627\n",
      "Epoch:3 Batch:480 Log-loss0.03164483234286308\n",
      "Epoch:3 Batch:560 Log-loss0.03660240024328232\n",
      "Epoch average log loss:0.03827732162816184\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.03970959235260172, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:4 Batch:0 Log-loss0.04295705631375313\n",
      "Epoch:4 Batch:80 Log-loss0.04227222129702568\n",
      "Epoch:4 Batch:160 Log-loss0.03569643199443817\n",
      "Epoch:4 Batch:240 Log-loss0.03799532726407051\n",
      "Epoch:4 Batch:320 Log-loss0.020701520144939423\n",
      "Epoch:4 Batch:400 Log-loss0.04030301421880722\n",
      "Epoch:4 Batch:480 Log-loss0.053648125380277634\n",
      "Epoch:4 Batch:560 Log-loss0.02877168543636799\n",
      "Epoch average log loss:0.037695136812648605\n",
      "In Epoch4, val_loss:0.04030890054851455, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:5 Batch:0 Log-loss0.02914491854608059\n",
      "Epoch:5 Batch:80 Log-loss0.03554144129157066\n",
      "Epoch:5 Batch:160 Log-loss0.021988870576024055\n",
      "Epoch:5 Batch:240 Log-loss0.033090535551309586\n",
      "Epoch:5 Batch:320 Log-loss0.04816390573978424\n",
      "Epoch:5 Batch:400 Log-loss0.03607340157032013\n",
      "Epoch:5 Batch:480 Log-loss0.045407045632600784\n",
      "Epoch:5 Batch:560 Log-loss0.026932809501886368\n",
      "Epoch average log loss:0.03746332267099725\n",
      "In Epoch5, val_loss:0.0397253932386339, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:6 Batch:0 Log-loss0.022680163383483887\n",
      "Epoch:6 Batch:80 Log-loss0.03773641213774681\n",
      "Epoch:6 Batch:160 Log-loss0.028106510639190674\n",
      "Epoch:6 Batch:240 Log-loss0.024599740281701088\n",
      "Epoch:6 Batch:320 Log-loss0.03572878614068031\n",
      "Epoch:6 Batch:400 Log-loss0.028722137212753296\n",
      "Epoch:6 Batch:480 Log-loss0.0334358848631382\n",
      "Epoch:6 Batch:560 Log-loss0.057513799518346786\n",
      "Epoch average log loss:0.03705681894041066\n",
      "In Epoch6, val_loss:0.039922579646930544, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:7 Batch:0 Log-loss0.028861144557595253\n",
      "Epoch:7 Batch:80 Log-loss0.03598487749695778\n",
      "Epoch:7 Batch:160 Log-loss0.03288282826542854\n",
      "Epoch:7 Batch:240 Log-loss0.04933273792266846\n",
      "Epoch:7 Batch:320 Log-loss0.04666830226778984\n",
      "Epoch:7 Batch:400 Log-loss0.03538510203361511\n",
      "Epoch:7 Batch:480 Log-loss0.04580047354102135\n",
      "Epoch:7 Batch:560 Log-loss0.03381863608956337\n",
      "Epoch average log loss:0.036849505504194115\n",
      "In Epoch7, val_loss:0.04033882165208041, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:8 Batch:0 Log-loss0.03916117921471596\n",
      "Epoch:8 Batch:80 Log-loss0.024451667442917824\n",
      "Epoch:8 Batch:160 Log-loss0.03710034117102623\n",
      "Epoch:8 Batch:240 Log-loss0.031936369836330414\n",
      "Epoch:8 Batch:320 Log-loss0.032975226640701294\n",
      "Epoch:8 Batch:400 Log-loss0.04301680624485016\n",
      "Epoch:8 Batch:480 Log-loss0.03337729349732399\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:8 Batch:560 Log-loss0.018029995262622833\n",
      "Epoch average log loss:0.03682059693216745\n",
      "In Epoch8, val_loss:0.040302184031095596, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:9 Batch:0 Log-loss0.033474210649728775\n",
      "Epoch:9 Batch:80 Log-loss0.02909381128847599\n",
      "Epoch:9 Batch:160 Log-loss0.02780190296471119\n",
      "Epoch:9 Batch:240 Log-loss0.027900410816073418\n",
      "Epoch:9 Batch:320 Log-loss0.035261120647192\n",
      "Epoch:9 Batch:400 Log-loss0.034019362181425095\n",
      "Epoch:9 Batch:480 Log-loss0.029447287321090698\n",
      "Epoch:9 Batch:560 Log-loss0.04000287875533104\n",
      "Epoch average log loss:0.03633360559553174\n",
      "In Epoch9, val_loss:0.0400507137951366, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:10 Batch:0 Log-loss0.031076958402991295\n",
      "Epoch:10 Batch:80 Log-loss0.0312681682407856\n",
      "Epoch:10 Batch:160 Log-loss0.04380783438682556\n",
      "Epoch:10 Batch:240 Log-loss0.032192766666412354\n",
      "Epoch:10 Batch:320 Log-loss0.04843224585056305\n",
      "Epoch:10 Batch:400 Log-loss0.04303550347685814\n",
      "Epoch:10 Batch:480 Log-loss0.05341682210564613\n",
      "Epoch:10 Batch:560 Log-loss0.03907017782330513\n",
      "Epoch average log loss:0.036145607398689855\n",
      "In Epoch10, val_loss:0.04020719891788784, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:11 Batch:0 Log-loss0.028322838246822357\n",
      "Epoch:11 Batch:80 Log-loss0.04716317355632782\n",
      "Epoch:11 Batch:160 Log-loss0.02744961529970169\n",
      "Epoch:11 Batch:240 Log-loss0.034271664917469025\n",
      "Epoch:11 Batch:320 Log-loss0.045559097081422806\n",
      "Epoch:11 Batch:400 Log-loss0.030173050239682198\n",
      "Epoch:11 Batch:480 Log-loss0.03815468028187752\n",
      "Epoch:11 Batch:560 Log-loss0.03545717895030975\n",
      "Epoch average log loss:0.03614146191560264\n",
      "In Epoch11, val_loss:0.040333990598658144, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:12 Batch:0 Log-loss0.027230001986026764\n",
      "Epoch:12 Batch:80 Log-loss0.04127097502350807\n",
      "Epoch:12 Batch:160 Log-loss0.03358836844563484\n",
      "Epoch:12 Batch:240 Log-loss0.030668092891573906\n",
      "Epoch:12 Batch:320 Log-loss0.02736787311732769\n",
      "Epoch:12 Batch:400 Log-loss0.0388588048517704\n",
      "Epoch:12 Batch:480 Log-loss0.03292927145957947\n",
      "Epoch:12 Batch:560 Log-loss0.031610287725925446\n",
      "Epoch average log loss:0.03563913737224149\n",
      "In Epoch12, val_loss:0.040804884312536344, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:13 Batch:0 Log-loss0.045787811279296875\n",
      "Epoch:13 Batch:80 Log-loss0.03896408528089523\n",
      "Epoch:13 Batch:160 Log-loss0.03921462222933769\n",
      "Epoch:13 Batch:240 Log-loss0.03644378110766411\n",
      "Epoch:13 Batch:320 Log-loss0.03385987505316734\n",
      "Epoch:13 Batch:400 Log-loss0.028341328725218773\n",
      "Epoch:13 Batch:480 Log-loss0.04336225986480713\n",
      "Epoch:13 Batch:560 Log-loss0.04299730062484741\n",
      "Epoch average log loss:0.03562383906849261\n",
      "In Epoch13, val_loss:0.04041303061639174, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:14 Batch:0 Log-loss0.0336168073117733\n",
      "Epoch:14 Batch:80 Log-loss0.051586467772722244\n",
      "Epoch:14 Batch:160 Log-loss0.041314397007226944\n",
      "Epoch:14 Batch:240 Log-loss0.03056691586971283\n",
      "Epoch:14 Batch:320 Log-loss0.03099648468196392\n",
      "Epoch:14 Batch:400 Log-loss0.039340727031230927\n",
      "Epoch:14 Batch:480 Log-loss0.029131175950169563\n",
      "Epoch:14 Batch:560 Log-loss0.03420166298747063\n",
      "Epoch average log loss:0.03543788889156921\n",
      "In Epoch14, val_loss:0.04043223776363044, best_val_loss:0.03970959235260172, best_auc:0.9884084612497316\n",
      "Epoch:15 Batch:0 Log-loss0.036242399364709854\n",
      "Epoch:15 Batch:80 Log-loss0.03521217033267021\n",
      "Epoch:15 Batch:160 Log-loss0.017370210960507393\n",
      "Epoch:15 Batch:240 Log-loss0.03660312667489052\n",
      "Epoch:15 Batch:320 Log-loss0.03416673466563225\n",
      "Epoch:15 Batch:400 Log-loss0.049826402217149734\n",
      "Epoch:15 Batch:480 Log-loss0.0398978628218174\n",
      "Epoch:15 Batch:560 Log-loss0.03285229951143265\n",
      "Epoch average log loss:0.03538340799776571\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_647.pt.\n",
      "\n",
      "Training on fold 8\n",
      "Epoch:1 Batch:0 Log-loss0.06488711386919022\n",
      "Epoch:1 Batch:80 Log-loss0.041825708001852036\n",
      "Epoch:1 Batch:160 Log-loss0.024930529296398163\n",
      "Epoch:1 Batch:240 Log-loss0.03125006705522537\n",
      "Epoch:1 Batch:320 Log-loss0.04528192803263664\n",
      "Epoch:1 Batch:400 Log-loss0.03434397652745247\n",
      "Epoch:1 Batch:480 Log-loss0.03789108246564865\n",
      "Epoch:1 Batch:560 Log-loss0.038841284811496735\n",
      "Epoch average log loss:0.03554575523261779\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.03863905014400414, best_val_loss:0.03863905014400414, best_auc:0.9891163275779903\n",
      "Epoch:2 Batch:0 Log-loss0.03358664736151695\n",
      "Epoch:2 Batch:80 Log-loss0.03720766678452492\n",
      "Epoch:2 Batch:160 Log-loss0.030355028808116913\n",
      "Epoch:2 Batch:240 Log-loss0.04001541808247566\n",
      "Epoch:2 Batch:320 Log-loss0.032924260944128036\n",
      "Epoch:2 Batch:400 Log-loss0.03965051844716072\n",
      "Epoch:2 Batch:480 Log-loss0.039620015770196915\n",
      "Epoch:2 Batch:560 Log-loss0.03428269550204277\n",
      "Epoch average log loss:0.03923427036830357\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch2, val_loss:0.03780414246263122, best_val_loss:0.03780414246263122, best_auc:0.9897340979674488\n",
      "Epoch:3 Batch:0 Log-loss0.022928060963749886\n",
      "Epoch:3 Batch:80 Log-loss0.04198277369141579\n",
      "Epoch:3 Batch:160 Log-loss0.030837751924991608\n",
      "Epoch:3 Batch:240 Log-loss0.04787914454936981\n",
      "Epoch:3 Batch:320 Log-loss0.02057330310344696\n",
      "Epoch:3 Batch:400 Log-loss0.05216123163700104\n",
      "Epoch:3 Batch:480 Log-loss0.045488204807043076\n",
      "Epoch:3 Batch:560 Log-loss0.05448384955525398\n",
      "Epoch average log loss:0.0383804704594825\n",
      "In Epoch3, val_loss:0.03859603550675358, best_val_loss:0.03780414246263122, best_auc:0.9897340979674488\n",
      "Epoch:4 Batch:0 Log-loss0.02800523303449154\n",
      "Epoch:4 Batch:80 Log-loss0.046156615018844604\n",
      "Epoch:4 Batch:160 Log-loss0.039434630423784256\n",
      "Epoch:4 Batch:240 Log-loss0.03486433997750282\n",
      "Epoch:4 Batch:320 Log-loss0.03185887262225151\n",
      "Epoch:4 Batch:400 Log-loss0.047223836183547974\n",
      "Epoch:4 Batch:480 Log-loss0.03272591903805733\n",
      "Epoch:4 Batch:560 Log-loss0.034766532480716705\n",
      "Epoch average log loss:0.03829191752177264\n",
      "In Epoch4, val_loss:0.0381889239321642, best_val_loss:0.03780414246263122, best_auc:0.9897340979674488\n",
      "Epoch:5 Batch:0 Log-loss0.029251093044877052\n",
      "Epoch:5 Batch:80 Log-loss0.04397038742899895\n",
      "Epoch:5 Batch:160 Log-loss0.04270613193511963\n",
      "Epoch:5 Batch:240 Log-loss0.036230940371751785\n",
      "Epoch:5 Batch:320 Log-loss0.03794470429420471\n",
      "Epoch:5 Batch:400 Log-loss0.050119202584028244\n",
      "Epoch:5 Batch:480 Log-loss0.028288140892982483\n",
      "Epoch:5 Batch:560 Log-loss0.041788410395383835\n",
      "Epoch average log loss:0.03790822842690562\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.03751126514220742, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:6 Batch:0 Log-loss0.04070843383669853\n",
      "Epoch:6 Batch:80 Log-loss0.042429644614458084\n",
      "Epoch:6 Batch:160 Log-loss0.035093359649181366\n",
      "Epoch:6 Batch:240 Log-loss0.03931872919201851\n",
      "Epoch:6 Batch:320 Log-loss0.03440815582871437\n",
      "Epoch:6 Batch:400 Log-loss0.0328565277159214\n",
      "Epoch:6 Batch:480 Log-loss0.0383247509598732\n",
      "Epoch:6 Batch:560 Log-loss0.044313374906778336\n",
      "Epoch average log loss:0.03740105995841857\n",
      "In Epoch6, val_loss:0.038955003026234496, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:7 Batch:0 Log-loss0.04786694422364235\n",
      "Epoch:7 Batch:80 Log-loss0.047138918191194534\n",
      "Epoch:7 Batch:160 Log-loss0.026884274557232857\n",
      "Epoch:7 Batch:240 Log-loss0.03304827958345413\n",
      "Epoch:7 Batch:320 Log-loss0.036494992673397064\n",
      "Epoch:7 Batch:400 Log-loss0.033519160002470016\n",
      "Epoch:7 Batch:480 Log-loss0.036750294268131256\n",
      "Epoch:7 Batch:560 Log-loss0.02913857437670231\n",
      "Epoch average log loss:0.03726488488526749\n",
      "In Epoch7, val_loss:0.038303222868629416, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:8 Batch:0 Log-loss0.035365741699934006\n",
      "Epoch:8 Batch:80 Log-loss0.0263389740139246\n",
      "Epoch:8 Batch:160 Log-loss0.028445446863770485\n",
      "Epoch:8 Batch:240 Log-loss0.03212195262312889\n",
      "Epoch:8 Batch:320 Log-loss0.026277877390384674\n",
      "Epoch:8 Batch:400 Log-loss0.042592961341142654\n",
      "Epoch:8 Batch:480 Log-loss0.0335368849337101\n",
      "Epoch:8 Batch:560 Log-loss0.029809366911649704\n",
      "Epoch average log loss:0.03729307265020907\n",
      "In Epoch8, val_loss:0.03849804327433406, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:9 Batch:0 Log-loss0.035227056592702866\n",
      "Epoch:9 Batch:80 Log-loss0.03321220353245735\n",
      "Epoch:9 Batch:160 Log-loss0.040791433304548264\n",
      "Epoch:9 Batch:240 Log-loss0.03755727782845497\n",
      "Epoch:9 Batch:320 Log-loss0.02223108522593975\n",
      "Epoch:9 Batch:400 Log-loss0.036377791315317154\n",
      "Epoch:9 Batch:480 Log-loss0.035230815410614014\n",
      "Epoch:9 Batch:560 Log-loss0.040438611060380936\n",
      "Epoch average log loss:0.036815993665784065\n",
      "In Epoch9, val_loss:0.03840831145742161, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:10 Batch:0 Log-loss0.03399014472961426\n",
      "Epoch:10 Batch:80 Log-loss0.018776964396238327\n",
      "Epoch:10 Batch:160 Log-loss0.05743438005447388\n",
      "Epoch:10 Batch:240 Log-loss0.03840179368853569\n",
      "Epoch:10 Batch:320 Log-loss0.03482433781027794\n",
      "Epoch:10 Batch:400 Log-loss0.03970171883702278\n",
      "Epoch:10 Batch:480 Log-loss0.029150905087590218\n",
      "Epoch:10 Batch:560 Log-loss0.03424927219748497\n",
      "Epoch average log loss:0.03674386511224189\n",
      "In Epoch10, val_loss:0.03830415780585586, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:11 Batch:0 Log-loss0.04021839797496796\n",
      "Epoch:11 Batch:80 Log-loss0.034669626504182816\n",
      "Epoch:11 Batch:160 Log-loss0.03301805630326271\n",
      "Epoch:11 Batch:240 Log-loss0.04166630282998085\n",
      "Epoch:11 Batch:320 Log-loss0.04627664387226105\n",
      "Epoch:11 Batch:400 Log-loss0.02671860158443451\n",
      "Epoch:11 Batch:480 Log-loss0.0311717689037323\n",
      "Epoch:11 Batch:560 Log-loss0.026508860290050507\n",
      "Epoch average log loss:0.03656070454578315\n",
      "In Epoch11, val_loss:0.03923555921626203, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:12 Batch:0 Log-loss0.035708267241716385\n",
      "Epoch:12 Batch:80 Log-loss0.04936208203434944\n",
      "Epoch:12 Batch:160 Log-loss0.02622394450008869\n",
      "Epoch:12 Batch:240 Log-loss0.05499271675944328\n",
      "Epoch:12 Batch:320 Log-loss0.03864910081028938\n",
      "Epoch:12 Batch:400 Log-loss0.02892233245074749\n",
      "Epoch:12 Batch:480 Log-loss0.03700194135308266\n",
      "Epoch:12 Batch:560 Log-loss0.03338764235377312\n",
      "Epoch average log loss:0.03621539663124297\n",
      "In Epoch12, val_loss:0.03791519431891528, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:13 Batch:0 Log-loss0.033848050981760025\n",
      "Epoch:13 Batch:80 Log-loss0.039963673800230026\n",
      "Epoch:13 Batch:160 Log-loss0.037564754486083984\n",
      "Epoch:13 Batch:240 Log-loss0.0224506426602602\n",
      "Epoch:13 Batch:320 Log-loss0.022251911461353302\n",
      "Epoch:13 Batch:400 Log-loss0.02708568423986435\n",
      "Epoch:13 Batch:480 Log-loss0.025850778445601463\n",
      "Epoch:13 Batch:560 Log-loss0.03182292729616165\n",
      "Epoch average log loss:0.03599110495831285\n",
      "In Epoch13, val_loss:0.03824969790521773, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:14 Batch:0 Log-loss0.026165010407567024\n",
      "Epoch:14 Batch:80 Log-loss0.030472712591290474\n",
      "Epoch:14 Batch:160 Log-loss0.0379832424223423\n",
      "Epoch:14 Batch:240 Log-loss0.039011090993881226\n",
      "Epoch:14 Batch:320 Log-loss0.03749664127826691\n",
      "Epoch:14 Batch:400 Log-loss0.030053937807679176\n",
      "Epoch:14 Batch:480 Log-loss0.03954797238111496\n",
      "Epoch:14 Batch:560 Log-loss0.02921169251203537\n",
      "Epoch average log loss:0.036019061213093145\n",
      "In Epoch14, val_loss:0.03822479229993869, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:15 Batch:0 Log-loss0.027345361188054085\n",
      "Epoch:15 Batch:80 Log-loss0.04423002526164055\n",
      "Epoch:15 Batch:160 Log-loss0.037210144102573395\n",
      "Epoch:15 Batch:240 Log-loss0.045759186148643494\n",
      "Epoch:15 Batch:320 Log-loss0.04527231678366661\n",
      "Epoch:15 Batch:400 Log-loss0.04682986065745354\n",
      "Epoch:15 Batch:480 Log-loss0.02865074761211872\n",
      "Epoch:15 Batch:560 Log-loss0.03321831300854683\n",
      "Epoch average log loss:0.03590958241838962\n",
      "In Epoch15, val_loss:0.037773803816732836, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:16 Batch:0 Log-loss0.03415023535490036\n",
      "Epoch:16 Batch:80 Log-loss0.030621351674199104\n",
      "Epoch:16 Batch:160 Log-loss0.03591173514723778\n",
      "Epoch:16 Batch:240 Log-loss0.027647338807582855\n",
      "Epoch:16 Batch:320 Log-loss0.03823221102356911\n",
      "Epoch:16 Batch:400 Log-loss0.034494828432798386\n",
      "Epoch:16 Batch:480 Log-loss0.04428953304886818\n",
      "Epoch:16 Batch:560 Log-loss0.04707137122750282\n",
      "Epoch average log loss:0.035595939098857346\n",
      "In Epoch16, val_loss:0.03818538842638301, best_val_loss:0.03751126514220742, best_auc:0.9898700414891296\n",
      "Epoch:17 Batch:0 Log-loss0.03613070026040077\n",
      "Epoch:17 Batch:80 Log-loss0.030107533559203148\n",
      "Epoch:17 Batch:160 Log-loss0.02647324651479721\n",
      "Epoch:17 Batch:240 Log-loss0.03630480542778969\n",
      "Epoch:17 Batch:320 Log-loss0.02425616979598999\n",
      "Epoch:17 Batch:400 Log-loss0.03592250123620033\n",
      "Epoch:17 Batch:480 Log-loss0.030591636896133423\n",
      "Epoch:17 Batch:560 Log-loss0.03849773854017258\n",
      "Epoch average log loss:0.03533907373369272\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_648.pt.\n",
      "\n",
      "Training on fold 9\n",
      "Epoch:1 Batch:0 Log-loss0.0498681478202343\n",
      "Epoch:1 Batch:80 Log-loss0.03308827802538872\n",
      "Epoch:1 Batch:160 Log-loss0.04498946666717529\n",
      "Epoch:1 Batch:240 Log-loss0.03502911701798439\n",
      "Epoch:1 Batch:320 Log-loss0.03598938509821892\n",
      "Epoch:1 Batch:400 Log-loss0.035275865346193314\n",
      "Epoch:1 Batch:480 Log-loss0.045959603041410446\n",
      "Epoch:1 Batch:560 Log-loss0.03923507407307625\n",
      "Epoch average log loss:0.03516888461807477\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch1, val_loss:0.04112522275402075, best_val_loss:0.04112522275402075, best_auc:0.9886620695208367\n",
      "Epoch:2 Batch:0 Log-loss0.055412691086530685\n",
      "Epoch:2 Batch:80 Log-loss0.04502759501338005\n",
      "Epoch:2 Batch:160 Log-loss0.043473947793245316\n",
      "Epoch:2 Batch:240 Log-loss0.044117674231529236\n",
      "Epoch:2 Batch:320 Log-loss0.027925027534365654\n",
      "Epoch:2 Batch:400 Log-loss0.037262994796037674\n",
      "Epoch:2 Batch:480 Log-loss0.03713422641158104\n",
      "Epoch:2 Batch:560 Log-loss0.028129911050200462\n",
      "Epoch average log loss:0.03853693995292166\n",
      "In Epoch2, val_loss:0.04169428789552823, best_val_loss:0.04112522275402075, best_auc:0.9886620695208367\n",
      "Epoch:3 Batch:0 Log-loss0.03416014462709427\n",
      "Epoch:3 Batch:80 Log-loss0.04351849481463432\n",
      "Epoch:3 Batch:160 Log-loss0.028908558189868927\n",
      "Epoch:3 Batch:240 Log-loss0.03574265539646149\n",
      "Epoch:3 Batch:320 Log-loss0.03435227647423744\n",
      "Epoch:3 Batch:400 Log-loss0.021951481699943542\n",
      "Epoch:3 Batch:480 Log-loss0.03125807270407677\n",
      "Epoch:3 Batch:560 Log-loss0.04276639595627785\n",
      "Epoch average log loss:0.037921784989469284\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch3, val_loss:0.04054298888504141, best_val_loss:0.04054298888504141, best_auc:0.9893460978724491\n",
      "Epoch:4 Batch:0 Log-loss0.04632754623889923\n",
      "Epoch:4 Batch:80 Log-loss0.035842251032590866\n",
      "Epoch:4 Batch:160 Log-loss0.030086537823081017\n",
      "Epoch:4 Batch:240 Log-loss0.041170842945575714\n",
      "Epoch:4 Batch:320 Log-loss0.03835776075720787\n",
      "Epoch:4 Batch:400 Log-loss0.04130389913916588\n",
      "Epoch:4 Batch:480 Log-loss0.03736275061964989\n",
      "Epoch:4 Batch:560 Log-loss0.04384532570838928\n",
      "Epoch average log loss:0.03774998034683189\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch4, val_loss:0.04053019376400733, best_val_loss:0.04053019376400733, best_auc:0.9893801854247984\n",
      "Epoch:5 Batch:0 Log-loss0.042301058769226074\n",
      "Epoch:5 Batch:80 Log-loss0.03606249764561653\n",
      "Epoch:5 Batch:160 Log-loss0.04028811678290367\n",
      "Epoch:5 Batch:240 Log-loss0.03507004305720329\n",
      "Epoch:5 Batch:320 Log-loss0.03987054526805878\n",
      "Epoch:5 Batch:400 Log-loss0.038884907960891724\n",
      "Epoch:5 Batch:480 Log-loss0.032979317009449005\n",
      "Epoch:5 Batch:560 Log-loss0.04727953299880028\n",
      "Epoch average log loss:0.03752260026355673\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch5, val_loss:0.04034946693479865, best_val_loss:0.04034946693479865, best_auc:0.9893968306623105\n",
      "Epoch:6 Batch:0 Log-loss0.041499555110931396\n",
      "Epoch:6 Batch:80 Log-loss0.04162311181426048\n",
      "Epoch:6 Batch:160 Log-loss0.03715226426720619\n",
      "Epoch:6 Batch:240 Log-loss0.03356056287884712\n",
      "Epoch:6 Batch:320 Log-loss0.02643711306154728\n",
      "Epoch:6 Batch:400 Log-loss0.03710552677512169\n",
      "Epoch:6 Batch:480 Log-loss0.04843147099018097\n",
      "Epoch:6 Batch:560 Log-loss0.04796629399061203\n",
      "Epoch average log loss:0.03705962027722438\n",
      "In Epoch6, val_loss:0.04043969472336135, best_val_loss:0.04034946693479865, best_auc:0.9893968306623105\n",
      "Epoch:7 Batch:0 Log-loss0.030485590919852257\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:7 Batch:80 Log-loss0.04403877258300781\n",
      "Epoch:7 Batch:160 Log-loss0.04668818414211273\n",
      "Epoch:7 Batch:240 Log-loss0.0374080166220665\n",
      "Epoch:7 Batch:320 Log-loss0.02633543312549591\n",
      "Epoch:7 Batch:400 Log-loss0.04821203649044037\n",
      "Epoch:7 Batch:480 Log-loss0.037637703120708466\n",
      "Epoch:7 Batch:560 Log-loss0.042374834418296814\n",
      "Epoch average log loss:0.036936644611081905\n",
      "In Epoch7, val_loss:0.04104022416694721, best_val_loss:0.04034946693479865, best_auc:0.9893968306623105\n",
      "Epoch:8 Batch:0 Log-loss0.032038066536188126\n",
      "Epoch:8 Batch:80 Log-loss0.03901800885796547\n",
      "Epoch:8 Batch:160 Log-loss0.04085184633731842\n",
      "Epoch:8 Batch:240 Log-loss0.041280340403318405\n",
      "Epoch:8 Batch:320 Log-loss0.03356344252824783\n",
      "Epoch:8 Batch:400 Log-loss0.02586238831281662\n",
      "Epoch:8 Batch:480 Log-loss0.02725793607532978\n",
      "Epoch:8 Batch:560 Log-loss0.03114374913275242\n",
      "Epoch average log loss:0.03686091272559549\n",
      "In Epoch8, val_loss:0.04097913077627538, best_val_loss:0.04034946693479865, best_auc:0.9893968306623105\n",
      "Epoch:9 Batch:0 Log-loss0.0356074720621109\n",
      "Epoch:9 Batch:80 Log-loss0.04709295928478241\n",
      "Epoch:9 Batch:160 Log-loss0.04092113673686981\n",
      "Epoch:9 Batch:240 Log-loss0.024883324280381203\n",
      "Epoch:9 Batch:320 Log-loss0.04105282202363014\n",
      "Epoch:9 Batch:400 Log-loss0.03952944278717041\n",
      "Epoch:9 Batch:480 Log-loss0.042015139013528824\n",
      "Epoch:9 Batch:560 Log-loss0.027793534100055695\n",
      "Epoch average log loss:0.0364009072072804\n",
      "In Epoch9, val_loss:0.0406956180794022, best_val_loss:0.04034946693479865, best_auc:0.9893968306623105\n",
      "Epoch:10 Batch:0 Log-loss0.02426602877676487\n",
      "Epoch:10 Batch:80 Log-loss0.0351468101143837\n",
      "Epoch:10 Batch:160 Log-loss0.04002348706126213\n",
      "Epoch:10 Batch:240 Log-loss0.03186838701367378\n",
      "Epoch:10 Batch:320 Log-loss0.0501554012298584\n",
      "Epoch:10 Batch:400 Log-loss0.04872320592403412\n",
      "Epoch:10 Batch:480 Log-loss0.03067481704056263\n",
      "Epoch:10 Batch:560 Log-loss0.02019612118601799\n",
      "Epoch average log loss:0.036392422509379685\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch10, val_loss:0.04033728788455621, best_val_loss:0.04033728788455621, best_auc:0.9895293591042278\n",
      "Epoch:11 Batch:0 Log-loss0.021429402753710747\n",
      "Epoch:11 Batch:80 Log-loss0.033192072063684464\n",
      "Epoch:11 Batch:160 Log-loss0.035927046090364456\n",
      "Epoch:11 Batch:240 Log-loss0.0225560013204813\n",
      "Epoch:11 Batch:320 Log-loss0.04427952691912651\n",
      "Epoch:11 Batch:400 Log-loss0.046299517154693604\n",
      "Epoch:11 Batch:480 Log-loss0.047213900834321976\n",
      "Epoch:11 Batch:560 Log-loss0.04039004072546959\n",
      "Epoch average log loss:0.036116489019644045\n",
      "Model has been saved as temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt.\n",
      "\n",
      "In Epoch11, val_loss:0.04012715772403915, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:12 Batch:0 Log-loss0.031491346657276154\n",
      "Epoch:12 Batch:80 Log-loss0.04375268518924713\n",
      "Epoch:12 Batch:160 Log-loss0.03546306863427162\n",
      "Epoch:12 Batch:240 Log-loss0.048387590795755386\n",
      "Epoch:12 Batch:320 Log-loss0.0356375016272068\n",
      "Epoch:12 Batch:400 Log-loss0.027038171887397766\n",
      "Epoch:12 Batch:480 Log-loss0.038987722247838974\n",
      "Epoch:12 Batch:560 Log-loss0.028453856706619263\n",
      "Epoch average log loss:0.03575929475482553\n",
      "In Epoch12, val_loss:0.04118821830684052, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:13 Batch:0 Log-loss0.026244478300213814\n",
      "Epoch:13 Batch:80 Log-loss0.024160834029316902\n",
      "Epoch:13 Batch:160 Log-loss0.03616837039589882\n",
      "Epoch:13 Batch:240 Log-loss0.030019676312804222\n",
      "Epoch:13 Batch:320 Log-loss0.027345867827534676\n",
      "Epoch:13 Batch:400 Log-loss0.029812829568982124\n",
      "Epoch:13 Batch:480 Log-loss0.03490768000483513\n",
      "Epoch:13 Batch:560 Log-loss0.0512387789785862\n",
      "Epoch average log loss:0.03580731160473079\n",
      "In Epoch13, val_loss:0.04180910489480507, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:14 Batch:0 Log-loss0.035297732800245285\n",
      "Epoch:14 Batch:80 Log-loss0.026988238096237183\n",
      "Epoch:14 Batch:160 Log-loss0.041562967002391815\n",
      "Epoch:14 Batch:240 Log-loss0.041281700134277344\n",
      "Epoch:14 Batch:320 Log-loss0.023599820211529732\n",
      "Epoch:14 Batch:400 Log-loss0.028523651883006096\n",
      "Epoch:14 Batch:480 Log-loss0.03639598563313484\n",
      "Epoch:14 Batch:560 Log-loss0.05700261518359184\n",
      "Epoch average log loss:0.03553349836064237\n",
      "In Epoch14, val_loss:0.040862459746689075, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:15 Batch:0 Log-loss0.02863347716629505\n",
      "Epoch:15 Batch:80 Log-loss0.03756405785679817\n",
      "Epoch:15 Batch:160 Log-loss0.03989553079009056\n",
      "Epoch:15 Batch:240 Log-loss0.033194370567798615\n",
      "Epoch:15 Batch:320 Log-loss0.03420557081699371\n",
      "Epoch:15 Batch:400 Log-loss0.04482691362500191\n",
      "Epoch:15 Batch:480 Log-loss0.038796279579401016\n",
      "Epoch:15 Batch:560 Log-loss0.03889124467968941\n",
      "Epoch average log loss:0.035518081931929506\n",
      "In Epoch15, val_loss:0.04089407149887974, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:16 Batch:0 Log-loss0.03840313479304314\n",
      "Epoch:16 Batch:80 Log-loss0.03236844018101692\n",
      "Epoch:16 Batch:160 Log-loss0.030568810179829597\n",
      "Epoch:16 Batch:240 Log-loss0.020228056237101555\n",
      "Epoch:16 Batch:320 Log-loss0.03871630132198334\n",
      "Epoch:16 Batch:400 Log-loss0.03491203486919403\n",
      "Epoch:16 Batch:480 Log-loss0.030595839023590088\n",
      "Epoch:16 Batch:560 Log-loss0.0326327309012413\n",
      "Epoch average log loss:0.03541430014385177\n",
      "In Epoch16, val_loss:0.04074978107791911, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:17 Batch:0 Log-loss0.034712761640548706\n",
      "Epoch:17 Batch:80 Log-loss0.04266089200973511\n",
      "Epoch:17 Batch:160 Log-loss0.03756854683160782\n",
      "Epoch:17 Batch:240 Log-loss0.03435380384325981\n",
      "Epoch:17 Batch:320 Log-loss0.046918243169784546\n",
      "Epoch:17 Batch:400 Log-loss0.046880125999450684\n",
      "Epoch:17 Batch:480 Log-loss0.034529197961091995\n",
      "Epoch:17 Batch:560 Log-loss0.0444762296974659\n",
      "Epoch average log loss:0.03526105677176799\n",
      "In Epoch17, val_loss:0.04071574010457185, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:18 Batch:0 Log-loss0.033005908131599426\n",
      "Epoch:18 Batch:80 Log-loss0.030914761126041412\n",
      "Epoch:18 Batch:160 Log-loss0.04029466211795807\n",
      "Epoch:18 Batch:240 Log-loss0.02288946323096752\n",
      "Epoch:18 Batch:320 Log-loss0.030843263491988182\n",
      "Epoch:18 Batch:400 Log-loss0.03217266872525215\n",
      "Epoch:18 Batch:480 Log-loss0.030630184337496758\n",
      "Epoch:18 Batch:560 Log-loss0.028905194252729416\n",
      "Epoch average log loss:0.035200106172955464\n",
      "In Epoch18, val_loss:0.04116508601970839, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:19 Batch:0 Log-loss0.027829894796013832\n",
      "Epoch:19 Batch:80 Log-loss0.03331774100661278\n",
      "Epoch:19 Batch:160 Log-loss0.04688282310962677\n",
      "Epoch:19 Batch:240 Log-loss0.05290345475077629\n",
      "Epoch:19 Batch:320 Log-loss0.0420704185962677\n",
      "Epoch:19 Batch:400 Log-loss0.026546111330389977\n",
      "Epoch:19 Batch:480 Log-loss0.031230399385094643\n",
      "Epoch:19 Batch:560 Log-loss0.03431027755141258\n",
      "Epoch average log loss:0.0349937688193417\n",
      "In Epoch19, val_loss:0.040716170532796615, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:20 Batch:0 Log-loss0.043845269829034805\n",
      "Epoch:20 Batch:80 Log-loss0.03360271081328392\n",
      "Epoch:20 Batch:160 Log-loss0.040238697081804276\n",
      "Epoch:20 Batch:240 Log-loss0.04366825520992279\n",
      "Epoch:20 Batch:320 Log-loss0.03357905149459839\n",
      "Epoch:20 Batch:400 Log-loss0.04217592254281044\n",
      "Epoch:20 Batch:480 Log-loss0.032154422253370285\n",
      "Epoch:20 Batch:560 Log-loss0.03507544845342636\n",
      "Epoch average log loss:0.0349110573679874\n",
      "In Epoch20, val_loss:0.041048599920047765, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:21 Batch:0 Log-loss0.04515450820326805\n",
      "Epoch:21 Batch:80 Log-loss0.04174225404858589\n",
      "Epoch:21 Batch:160 Log-loss0.045114461332559586\n",
      "Epoch:21 Batch:240 Log-loss0.03345750272274017\n",
      "Epoch:21 Batch:320 Log-loss0.02868153154850006\n",
      "Epoch:21 Batch:400 Log-loss0.03914862126111984\n",
      "Epoch:21 Batch:480 Log-loss0.039659030735492706\n",
      "Epoch:21 Batch:560 Log-loss0.03572624921798706\n",
      "Epoch average log loss:0.034845265040972406\n",
      "In Epoch21, val_loss:0.04143439941640703, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:22 Batch:0 Log-loss0.031078891828656197\n",
      "Epoch:22 Batch:80 Log-loss0.025072479620575905\n",
      "Epoch:22 Batch:160 Log-loss0.03648410364985466\n",
      "Epoch:22 Batch:240 Log-loss0.02326861023902893\n",
      "Epoch:22 Batch:320 Log-loss0.04325728490948677\n",
      "Epoch:22 Batch:400 Log-loss0.03137240186333656\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch:22 Batch:480 Log-loss0.028182735666632652\n",
      "Epoch:22 Batch:560 Log-loss0.0524577759206295\n",
      "Epoch average log loss:0.03471323919116653\n",
      "In Epoch22, val_loss:0.04071607518488194, best_val_loss:0.04012715772403915, best_auc:0.9895871182630483\n",
      "Epoch:23 Batch:0 Log-loss0.039027612656354904\n",
      "Epoch:23 Batch:80 Log-loss0.02714204601943493\n",
      "Epoch:23 Batch:160 Log-loss0.04840656742453575\n",
      "Epoch:23 Batch:240 Log-loss0.027546949684619904\n",
      "Epoch:23 Batch:320 Log-loss0.025439567863941193\n",
      "Epoch:23 Batch:400 Log-loss0.024166835471987724\n",
      "Epoch:23 Batch:480 Log-loss0.044597651809453964\n",
      "Epoch:23 Batch:560 Log-loss0.028611093759536743\n",
      "Epoch average log loss:0.03460068312540118\n",
      "A pre-trained model at temporary_checkpoints/FASTTEXT_RHN_64_64-TEMP.pt has been loaded.\n",
      "\n",
      "Model has been saved as checkpoints/FASTTEXT_RHN_64_649.pt.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# keep training\n",
    "\n",
    "new_models, best_logloss, best_auc, best_val_pred = trainer.keep_train_folds(X=train_sequences, y=training_labels,\n",
    "                    fold_count=10, batch_size=256, old_models=models)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.038431989078936295"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_logloss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make predections"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training AUC 0.9898540098272272\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "train_fold_preditcions = np.concatenate(best_val_pred, axis=0)\n",
    "training_auc = roc_auc_score(training_labels, train_fold_preditcions)\n",
    "print(\"Training AUC\", training_auc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "path = 'Dataset/'\n",
    "TRAIN_DATA_FILE=path + 'train.csv'\n",
    "TEST_DATA_FILE=path + 'test.csv'\n",
    "test_df = pd.read_csv(TEST_DATA_FILE)\n",
    "train_df = pd.read_csv(TRAIN_DATA_FILE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicting testing results...\n"
     ]
    },
    {
     "ename": "NameError",
     "evalue": "name 'val_loss' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-53-605350b6ca3a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     20\u001b[0m \u001b[0mtest_predicts\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"id\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtest_ids\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m \u001b[0mtest_predicts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtest_predicts\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"id\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0mCLASSES\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 22\u001b[0;31m \u001b[0msubmit_path\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msubmit_path_prefix\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m\"-L{:4f}-A{:4f}.csv\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mval_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtotal_auc\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     23\u001b[0m \u001b[0mtest_predicts\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msubmit_path\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'val_loss' is not defined"
     ]
    }
   ],
   "source": [
    "#test_data = test_df\n",
    "CLASSES = [\"toxic\", \"severe_toxic\", \"obscene\", \"threat\", \"insult\", \"identity_hate\"]\n",
    "submit_path_prefix = \"results/rhn/Fasttext-tunedRHN-\" + str(MAX_SEQUENCE_LENGTH) \n",
    "\n",
    "print(\"Predicting testing results...\")\n",
    "test_predicts_list = []\n",
    "for fold_id, model in enumerate(new_models):\n",
    "    test_predicts = model.predict(test_sequences, batch_size=256, verbose=1)\n",
    "    test_predicts_list.append(test_predicts)\n",
    "\n",
    "test_predicts = np.zeros(test_predicts_list[0].shape)\n",
    "for fold_predict in test_predicts_list:\n",
    "    test_predicts += fold_predict\n",
    "test_predicts /= len(test_predicts_list)\n",
    "\n",
    "test_ids = test_df[\"id\"].values\n",
    "test_ids = test_ids.reshape((len(test_ids), 1))\n",
    "\n",
    "test_predicts = pd.DataFrame(data=test_predicts, columns=CLASSES)\n",
    "test_predicts[\"id\"] = test_ids\n",
    "test_predicts = test_predicts[[\"id\"] + CLASSES]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "submit_path = submit_path_prefix + \"-L{:4f}-A{:4f}.csv\".format(best_logloss, best_auc)\n",
    "test_predicts.to_csv(submit_path, index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predicting training results...\n"
     ]
    }
   ],
   "source": [
    "print(\"Predicting training results...\")\n",
    "\n",
    "train_ids = train_df[\"id\"].values\n",
    "train_ids = train_ids.reshape((len(train_ids), 1))\n",
    "\n",
    "train_predicts = pd.DataFrame(data=train_fold_preditcions, columns=CLASSES)\n",
    "train_predicts[\"id\"] = train_ids\n",
    "train_predicts = train_predicts[[\"id\"] + CLASSES]\n",
    "submit_path = submit_path_prefix + \"-Train-L{:4f}-A{:4f}.csv\".format(best_logloss, best_auc)\n",
    "train_predicts.to_csv(submit_path, index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}