diff --git a/cnntrain.ipynb b/cnntrain.ipynb
new file mode 100644
index 0000000..d99b5d8
--- /dev/null
+++ b/cnntrain.ipynb
@@ -0,0 +1,2094 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 96,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Importing libraries\n",
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import tensorflow as tf\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
+    "from tensorflow import keras\n",
+    "from tensorflow.keras.layers import Conv2D,MaxPooling2D,Flatten,Dense\n",
+    "from tensorflow.keras.models import Sequential\n",
+    "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint\n",
+    "import matplotlib.pyplot as plt\n",
+    "from keras.layers import LeakyReLU\n",
+    "from sklearn import metrics\n",
+    "from sklearn.metrics import confusion_matrix\n",
+    "import seaborn as sns\n",
+    "from keras.models import load_model"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Load and Pre-process the dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "img_train  = pd.read_csv('./train/images.csv') \n",
+    "img_val  = pd.read_csv('./valid/images.csv') \n",
+    "label_train = pd.read_csv('./train/labels.csv')\n",
+    "label_val = pd.read_csv('./valid/labels.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "3199\n",
+      "3199\n",
+      "799\n",
+      "799\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(len(img_train))\n",
+    "print(len(label_train))\n",
+    "print(len(img_val))\n",
+    "print(len(label_val))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>0.631373</th>\n",
+       "      <th>0.623529</th>\n",
+       "      <th>0.713726</th>\n",
+       "      <th>0.705882</th>\n",
+       "      <th>0.658824</th>\n",
+       "      <th>0.666667</th>\n",
+       "      <th>0.654902</th>\n",
+       "      <th>0.635294</th>\n",
+       "      <th>0.647059</th>\n",
+       "      <th>0.705882.1</th>\n",
+       "      <th>...</th>\n",
+       "      <th>0.682353.349</th>\n",
+       "      <th>0.611765.91</th>\n",
+       "      <th>0.650980.397</th>\n",
+       "      <th>0.658824.443</th>\n",
+       "      <th>0.600000.48</th>\n",
+       "      <th>0.603922.77</th>\n",
+       "      <th>0.654902.426</th>\n",
+       "      <th>0.694118.222</th>\n",
+       "      <th>0.658824.444</th>\n",
+       "      <th>0.666667.441</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.725490</td>\n",
+       "      <td>0.752941</td>\n",
+       "      <td>0.749020</td>\n",
+       "      <td>0.701961</td>\n",
+       "      <td>0.690196</td>\n",
+       "      <td>0.721569</td>\n",
+       "      <td>0.709804</td>\n",
+       "      <td>0.745098</td>\n",
+       "      <td>0.654902</td>\n",
+       "      <td>0.721569</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.721569</td>\n",
+       "      <td>0.698039</td>\n",
+       "      <td>0.721569</td>\n",
+       "      <td>0.686275</td>\n",
+       "      <td>0.713726</td>\n",
+       "      <td>0.682353</td>\n",
+       "      <td>0.690196</td>\n",
+       "      <td>0.698039</td>\n",
+       "      <td>0.701961</td>\n",
+       "      <td>0.725490</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.717647</td>\n",
+       "      <td>0.701961</td>\n",
+       "      <td>0.713726</td>\n",
+       "      <td>0.733333</td>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.717647</td>\n",
+       "      <td>0.725490</td>\n",
+       "      <td>0.682353</td>\n",
+       "      <td>0.717647</td>\n",
+       "      <td>0.674510</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.709804</td>\n",
+       "      <td>0.694118</td>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.682353</td>\n",
+       "      <td>0.639216</td>\n",
+       "      <td>0.713726</td>\n",
+       "      <td>0.670588</td>\n",
+       "      <td>0.678431</td>\n",
+       "      <td>0.737255</td>\n",
+       "      <td>0.674510</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.674510</td>\n",
+       "      <td>0.654902</td>\n",
+       "      <td>0.678431</td>\n",
+       "      <td>0.666667</td>\n",
+       "      <td>0.662745</td>\n",
+       "      <td>0.678431</td>\n",
+       "      <td>0.662745</td>\n",
+       "      <td>0.686275</td>\n",
+       "      <td>0.686275</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.639216</td>\n",
+       "      <td>0.662745</td>\n",
+       "      <td>0.631373</td>\n",
+       "      <td>0.643137</td>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.662745</td>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.666667</td>\n",
+       "      <td>0.654902</td>\n",
+       "      <td>0.631373</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0.647059</td>\n",
+       "      <td>0.729412</td>\n",
+       "      <td>0.701961</td>\n",
+       "      <td>0.674510</td>\n",
+       "      <td>0.611765</td>\n",
+       "      <td>0.698039</td>\n",
+       "      <td>0.713726</td>\n",
+       "      <td>0.662745</td>\n",
+       "      <td>0.701961</td>\n",
+       "      <td>0.674510</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.639216</td>\n",
+       "      <td>0.670588</td>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.674510</td>\n",
+       "      <td>0.721569</td>\n",
+       "      <td>0.694118</td>\n",
+       "      <td>0.674510</td>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.749020</td>\n",
+       "      <td>0.729412</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0.694118</td>\n",
+       "      <td>0.682353</td>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.705882</td>\n",
+       "      <td>0.666667</td>\n",
+       "      <td>0.694118</td>\n",
+       "      <td>0.674510</td>\n",
+       "      <td>0.713726</td>\n",
+       "      <td>0.690196</td>\n",
+       "      <td>0.709804</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.658824</td>\n",
+       "      <td>0.698039</td>\n",
+       "      <td>0.694118</td>\n",
+       "      <td>0.709804</td>\n",
+       "      <td>0.725490</td>\n",
+       "      <td>0.694118</td>\n",
+       "      <td>0.729412</td>\n",
+       "      <td>0.701961</td>\n",
+       "      <td>0.694118</td>\n",
+       "      <td>0.709804</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>5 rows × 8192 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   0.631373  0.623529  0.713726  0.705882  0.658824  0.666667  0.654902  \\\n",
+       "0  0.725490  0.752941  0.749020  0.701961  0.690196  0.721569  0.709804   \n",
+       "1  0.717647  0.701961  0.713726  0.733333  0.705882  0.717647  0.725490   \n",
+       "2  0.705882  0.674510  0.654902  0.678431  0.666667  0.662745  0.678431   \n",
+       "3  0.647059  0.729412  0.701961  0.674510  0.611765  0.698039  0.713726   \n",
+       "4  0.694118  0.682353  0.705882  0.705882  0.666667  0.694118  0.674510   \n",
+       "\n",
+       "   0.635294  0.647059  0.705882.1  ...  0.682353.349  0.611765.91  \\\n",
+       "0  0.745098  0.654902    0.721569  ...      0.721569     0.698039   \n",
+       "1  0.682353  0.717647    0.674510  ...      0.709804     0.694118   \n",
+       "2  0.662745  0.686275    0.686275  ...      0.639216     0.662745   \n",
+       "3  0.662745  0.701961    0.674510  ...      0.639216     0.670588   \n",
+       "4  0.713726  0.690196    0.709804  ...      0.658824     0.698039   \n",
+       "\n",
+       "   0.650980.397  0.658824.443  0.600000.48  0.603922.77  0.654902.426  \\\n",
+       "0      0.721569      0.686275     0.713726     0.682353      0.690196   \n",
+       "1      0.705882      0.682353     0.639216     0.713726      0.670588   \n",
+       "2      0.631373      0.643137     0.705882     0.662745      0.705882   \n",
+       "3      0.705882      0.674510     0.721569     0.694118      0.674510   \n",
+       "4      0.694118      0.709804     0.725490     0.694118      0.729412   \n",
+       "\n",
+       "   0.694118.222  0.658824.444  0.666667.441  \n",
+       "0      0.698039      0.701961      0.725490  \n",
+       "1      0.678431      0.737255      0.674510  \n",
+       "2      0.666667      0.654902      0.631373  \n",
+       "3      0.705882      0.749020      0.729412  \n",
+       "4      0.701961      0.694118      0.709804  \n",
+       "\n",
+       "[5 rows x 8192 columns]"
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "img_train.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>1.000000</th>\n",
+       "      <th>0.000000</th>\n",
+       "      <th>0.000000.1</th>\n",
+       "      <th>0.000000.2</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   1.000000  0.000000  0.000000.1  0.000000.2\n",
+       "0       1.0       0.0         0.0         0.0\n",
+       "1       1.0       0.0         0.0         0.0\n",
+       "2       1.0       0.0         0.0         0.0\n",
+       "3       1.0       0.0         0.0         0.0\n",
+       "4       1.0       0.0         0.0         0.0"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "label_train.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "img_category_train = []\n",
+    "img_category_val = []\n",
+    "\n",
+    "for index,row in label_train.iterrows():\n",
+    "    category = 0\n",
+    "    if (row[0] == 1):\n",
+    "        category = 0 \n",
+    "    elif (row[1] == 1):\n",
+    "        category = 1 \n",
+    "    elif (row[2] == 1):\n",
+    "        category = 2 \n",
+    "    else:\n",
+    "        category = 3\n",
+    "    \n",
+    "    img_category_train.append(category)\n",
+    "    \n",
+    "for index,row in label_val.iterrows():\n",
+    "    category = 0\n",
+    "    if (row[0] == 1):\n",
+    "        category = 0 \n",
+    "    elif (row[1] == 1):\n",
+    "        category = 1 \n",
+    "    elif (row[2] == 1):\n",
+    "        category = 2 \n",
+    "    else:\n",
+    "        category = 3\n",
+    "    \n",
+    "    img_category_val.append(category)\n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(3199, 8192)\n",
+      "(3199, 4)\n",
+      "(799, 8192)\n",
+      "(799, 4)\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(np.shape(img_train))\n",
+    "print(np.shape(label_train))\n",
+    "print(np.shape(img_val))\n",
+    "print(np.shape(label_val))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(3199, 8192)\n",
+      "(799, 8192)\n",
+      "(3199, 4)\n",
+      "(799, 4)\n",
+      "(3199, 64, 128, 1)\n",
+      "(799, 64, 128, 1)\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "img_train_data = img_train.to_numpy()\n",
+    "img_data_val = img_val.to_numpy()\n",
+    "label_train_data = label_train.to_numpy()\n",
+    "label_val_data = label_val.to_numpy()\n",
+    "print(img_train_data.shape)\n",
+    "print(img_data_val.shape)\n",
+    "print(label_train_data.shape)\n",
+    "print(label_val_data.shape)\n",
+    "raw_img_data_train = img_train_data.reshape(-1,64,128)\n",
+    "raw_img_data_val = img_data_val.reshape(-1,64,128)\n",
+    "\n",
+    "reshaped_img_data_train = np.reshape(raw_img_data_train,(-1, 64, 128, 1))\n",
+    "reshaped_img_data_val = np.reshape(raw_img_data_val,(-1, 64, 128,1))\n",
+    "\n",
+    "print(reshaped_img_data_train.shape)\n",
+    "print(reshaped_img_data_val.shape)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "X_train, X_test, y_train, y_test = train_test_split(reshaped_img_data_train,label_train_data, test_size=0.2, random_state=69)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2559 640 2559 640\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(len(X_train) , len(X_test) , len(y_train) , len(y_test))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(2559, 64, 128, 1)"
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "X_train.shape\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "spectrograms = [ \"squiggle\", \"narrowband\", \"narrowbanddrd\", \"noise\"]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# interpolation='nearest' simply displays an image without trying to interpolate between pixels if the display resolution is not the same as the image resolution"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Visualize the dataset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[1. 0. 0. 0.]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(raw_img_data_train[0], interpolation='nearest')\n",
+    "print(label_train_data[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0. 0. 0. 1.]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(raw_img_data_train[2179], interpolation='nearest')\n",
+    "print(label_train_data[2179])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0. 0. 1. 0.]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(raw_img_data_train[1012], interpolation='nearest')\n",
+    "print(label_train_data[1012])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0. 1. 0. 0.]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAADJCAYAAAA6q2k2AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAACAXElEQVR4nO19eXxWZ5n29WRfyUJCSAghQAhLCgQo0FJaKN3s3tpaa+tYHbdxG3WcUcdtdL66jKOO6/hZW7V+1lbtol2spZRCS2nZtxIgbCGEhEAgCUnInvP9cZ/73NdLgmBb0fT3XL9ffz08b96zPOc5572X675uFwQBPDw8PDyGH+L+1ifg4eHh4fHa4F/gHh4eHsMU/gXu4eHhMUzhX+AeHh4ewxT+Be7h4eExTOFf4B4eHh7DFK/rBe6ce4tzbpdzbo9z7rNv1El5eHh4eJwZ7rXywJ1z8QCqAVwBoA7AOgDvCIKg6o07PQ8PDw+P0yHhdXx3HoA9QRDsAwDn3EMAbgRw2hd4fHp6kJCTK/9IGpCx+AE7mTjZjnP2o9LZlRRtp6T0DtpnV1ditO0S5PsZiT3RWPeAXGJ6vI21dKdG28mJfQCAviY7Tt8I2U96sn0nOa4v2u4diAcAtJ20/cQl9kfb6eHxsxNORmMHm0fKOdLvZUpm96DrSYyz/ZzspXPqF2cpJcnOo7tXri3oNkcqKd3mKN4NDNpnd798J4nG2nuS7fsJsv/+AdtnXNzgH/me3vhouzDtBADgSHdGNDYi0a7tRK/sv7/TlltyusxRADfofAGguy+8tsA+Z1tDzyl9iHvd00PHSbL5SI6Xa+sZsM97w3nl4wz02bUnJ8v3u2mdgf42LknmMTGe5vik3LfEVDt2En3e0Snz4eLtgoJ+26eu46DfzsP1yOcu1fbjaDHFh/PRS99Bp9yjIIEmjo6JAdlneqrdq472lGhbz7+3d/BrIj2FvtNt61TnpiD9RDTUeDJTzpfWacw5hZee0mDz1T3W1lfQGzf43MPrTEy27/TRmo3mboDmNXxGAx4b6pxoCuMTbL77u8N3SXqXnWe4Tvl5YaQlyfoc4HU8xJrv6afrDf+W11TrrqNNQRDkn7r/1/MCHwPgIP27DsD8P/eFhJxcFP/zJ+UfJZ0AgMyMzujz/PQOAEA6PfybdpZG2xXldYP2uX3vmGg7NUsmdn7xgWispk1+MM4fWRuNPb57erQ9qeAoAODofXacpqtkP3PH237K0o9G24e6sgEAKzdMi8YyxtiCnTNazvPGkZuisX975B8AAAlddvOmLtk96HryU9qj7U1Hi6Pt5tZ0AMDkosZobHej3M+B/enRWPGc+mg7J/nkoH3WtMkPSVF6azS2+sD4aLs077gcr8t+nNKT7CWpqG3MjbY/N+dpAMCPdi+Kxi4rro62n6srBwC0Vdl3JsyVpdMzYAtXzxcA9hzPAwB0d9uLs49+NNIz5B7pXMu1yf4P1I+MxiYVH7FryzwGADh0Mjsaqz8xQvZND2DbMZvP8tLDAIDq3UXRmL5MASCjRO57YWabnfsWuW+jp9mxx2a2RNtrtk8EACRl2TrvabeXoK7jrg4bS94vL9bEyuZoTA0eAMhKle80tmZGY/07ZLsn314ECSPsXvaFP6jzp+yLxtavnhxtF0yXtdbQmI1TMa+sJtrecKDEjhm+bD8199lo7DsbLwcAJO2xNdU12gwRJMp1TPtPW9v7v5kVbXc2pQ069/5m+REsnGjPZXN7mu1f567d1k9KgbxfeukHPqGavhOek0u1c8vO6Yi2W2qyAQDz5thzu69F1lpru10bY1axrE9+QbMBkRG+6w62ZUdjveHfFtOa+cPFP7aXEeGvnsR0zn3AObfeObd+oKPjzF/w8PDw8DgrvB4L/BCAsfTv4nAsBkEQ3APgHgDIm5oXLFz8KgAgL0mswq3NZkGr5X1Z3s5oLKnCrIcZmbL7+3eYoX/nnDV2QqFlvGK7WRFqQT21ryIayyKrX38Bx3xsczT27CszAADVI8xj4V/QiRlNsmGnhpLslmh71T6xsFp7zB3tL5Tj9JNFkJVkrtjb89YCAJ47YVY9u1CKDgqrqOvbnGshkNLM49F2fYdYMXX06677vDzHIl18nh1huGNR4Z5obGVDWbR9rNnCJAq1vDU0AACrj5BVny3ntKXYrJ3dO+W+B4lmRZ5//vpou6ZFrOkEcu8XTjRL8ZXGUgDAhsPmpby7TNZCTY5Z4E/tPC/a3tcoVv30Ylumes4VhQ3RWA/dy48WLwcA/NfAW+xzWgs5KbKWJuiaANAwQSzf83Jtn2z1DwVdp4B5Cq8eL4zG6jtDj6R2RDQWn2dzox7PU622zvvSJSRQPsm8sncVvxxtf2ndDQBiLegZ8+2+N3fL/crLM+9iVLo8t9XH7NkozjevQNfft5+/JhpLzAu97blmLfe1mpej1nSQblZsP4WCMkfL8TMpbNObJfdA5x8A5uRZQOCPz58PAIgvMaMxJ0M8vM4eewY7ptr6Q+gFFRa0REMtK0dH2+OWyLqZl1UTjak3xeGd/NHm3VYdLQAApCSaVd9FISn1oloOZkdjt1wo7wJ+Z50Or8cCXwdgknNuvHMuCcDtAB5/Hfvz8PDw8PgL8JpZKADgnLsGwHcBxAP4WRAEX/1zf59RPjqY/sO7AABNrWLJ3TBpW/T5+mNiCdTsLYjGknLMStV4ksadAODKMWat/6ZqDgAgkRJ9GaEFx5ZjMiVD9XP+Ja+ukV/dhKP2Sx033n7JB0LrIDVt6OSP4topr0bbdaEFtvOIXduMQrOMNNFTU2uWzd0LH4u2lzWLZb5i89RBx0nNs9gxWy46X5GVACA1V67zkfPvicZ+dvyiaFst5yOb7TxHVVpscsGo/QCAKalmXa5tk++09poFpZYHYN5JL1muHypZAQD4zMa3RmMcB07KCJM/dD2fqHwu2v517VwAQMNem6+EVvnb8fPNEhvKi9G4N2AWeE+z3T9ecxp3T6AkNZ+TWp8HdpilpnFxxuJis2wnp4m1/fQR8w6qDpi1rfFlXucVI+U724/ZcSZkH4u2r8/bAgC4t3ZhNNbaOXhNtp4wL6ikQKzli/LNs1l6aIpdG8VgFdvqxHPqbzJvy/VZTuAjVy4FADzRYHmm2m1ybckllovhta/Y9YAd+/L3maeQE5IBDvfYfXtiy8xB33//3Bej7V9Vy/pIoSR281F5xuJazQKOG233eknZLgCxng8nnzXxf7A5OxpLCJOcbGE3NVkeIiVM1nNe7oU95tEOhB55/ljzYvRdxe+PH815cEMQBOfjFLyeEAqCIPgjgD++nn14eHh4eLw2+EpMDw8Pj2GK12WB/6Xo7U6MXF5NbPx+p7lCSpGLyyC+d7UlO7bsEBeL6Xcn+sxNHAj5nX3V5sJULBHX5Wi6hVB21JkbqomNPQ2DKJboy7fzKM21xMT0nMHu34o6c4u6qrIBAPuK86KxymwJZ3DiUpOMgFHo6ihs8x+P3xZtuzEyX/Onmyuu1Mbf/tHc5ozzLIl5cY7M09rk0misIEsSQstPWqK3NMUScHvT5Jyby8zV1nAXADzVLomVKRUWQll3WEJfzYfNxS0qMfde5/uSMjv3T615GwBzIQFgceWOaHvFVrnXiU22RB/Onx1taygH+n9YCE6TlQAwocCubXfdKABARpaFy3paw1AAcd2ZrqjhNqWyAbEhq/eVrAIAfGnb26IxpSFqGAgAjnTbmtzWLJTE+uN2/+MpmZuVKOf3ufKn7do6JEy15mULTazPsuusmiQhq/ZWC2PpPjnUN3DS5lPX3IOvmmfOoaL7K+4HANy59T3RmD6j9Rl2r8tHWnJyedPkQdc2kCH77Nlvc5BabM9WdYs8ewVr7BnbeouRG5TSyiGdnHxZx6kUIvn9wRnRtoZo1u0fF40VjZFnozPf1pwm2AELnTDlMzXe9q/kBKY+a/JY1x4AFI+3EJ6GYMak2D6nFlvCevtO4YHcNHZrNPazqiUAgCf77V4DD2IoeAvcw8PDY5jinFrgBZmt+NSlEjKv6RLrgWmEWpjCScb+crMIusIkV3FaSzT2xPpZ0faVsyUherTELEZN+uUU2a971gizoJRSpIUhAIDw65yY4ITQKz2lAIB2os1xkqI6pDtxMczRHtkpUxyZAvnS0QkAgIFiOw+2TjNDK4otArWcgiyiQhEePhRarLQftYx+0mFWe9sRm6/5FXsBxCaZNtUZVc9tEMvrq/U3RmMpR8RiXXydJaS5yKlyuiTJhrJCdq4rjcY6+8kanyHJ6TV1ZkFdUWAJ65+uu1jOp4sq2FJkrbDlOznLErCafLqh1JJDJ4rlvu5qtaRrTZMVHKmHNiHPPIojHTZf39sj1tI7lrwUjT24XJLCvXRuVcm2//aQCsjJv4QxtiaPdsn+7z10cTSmidfuKWb9JewfnKRMTLE1e9VE8WhW1U+IxuLS7PNr8+V+MW1Xn0sA+Mqh6wAALc3mBSvY6lZPDwBebJ4EILa6U2m7xVNaorHMBFvnWiiUUGivo+9N+Fm0/dYXPgQglmjQN0TloyZ6AVtLI5fZHLXdJMdkOiKjrUue5+ZE87YOEm1XyRFfm/b7aOxDK/9h0H5KK3ZF23qduiYAoGS6ea8JLbJGnqo3yuCUuTUAgNqW7CHPk+EtcA8PD49hCv8C9/Dw8BimeF088L8UU2YkBz97XEIm/7FfXHDlXANAaYm4ZayzMTrPQh9fLHsKAHCs31zYL6y6OdoeSrdCdRaY+6lJFQCID8dHTLZkRudG4d92jTVXPHOk8cCn5YtbzgkSPs+htAymjZBze3SPJW2Zo/rOCesAAGtbS6MxrnZrrpfQB4eCEkIhsD7iJbeRJoOrke0xcy0conzlvIl2vcxbVYElDt/ceYFxcscnyz36xiarTHzPefL5+hZL5NxSsDHafrBhHgAgiQTB9POvPGKJ2uyZlnD85pRHAAD3HzXXkytGdy4VVx2VxrlWDjwLpI3NaYm2962ThFFfrp3Hvy78EwBgSZq5vdc898/RNkJtD+aG52UZn7kgzaoUFZqo3nLE1iGHHJTzz2E55SgDQFyyrE/laQNAeZbpqiiWbjYe+awpNQCGrjNgTj5ry6gOEN8X1s1RzQ7mLas+0NhU4y2ztlBvk6y5vFI7dw01ZlBYhZ8N1b1JfDo7Gguute/r2gclel1HKNSVQmO99hzE54SVzzSmyeu2wzbX/DwNpcWkFcGAPW9cSanhmCF1WGCaMN/fdmk0xtpF0xdIYv9dhfaMvdQma/vRqko7jzs/PyQP3FvgHh4eHsMU/gXu4eHhMUxxTlko+9vycceL7wcQmy1XqJvIkpjl2eZ6fmmXiO+wvKnyQQGg9sXQhS+y0IRKZY5Ktr9THi4AHOsQ10dDGADwk05haGSy1vD6nGh703RxQ5lTy5KbQSjTmVhqoZqoBJlcOuVkMzgsU1ZIUpkQN3IoV5xDTvydxCI5vrJ7AABZck2c0a+YbNl75V9zyKDqhIW5flMncgUsa6uhE3a/uwbMVVfBsDXbzBXfvFmYEZULjRvOZcsaOmHmSlKchc46LhV54KMd5o5qGCKB5ECr91pZdP50CdFwef1zTXK9j/daaIvlVVXoSUMDAJBENQHq/rO0K/PhFWNJXlfBEsdP9RgLQdcFh2eUaz93tH0nZ7SFj7Y3yHWyUJuuJQ7FdCSae68sBw6rJNOaV4nc/5xrEkebO+Q8Hnl5XjSm4ksA8HCrePmtG43N0lsi9z+BtLX5XnfulPNom2fHTnzV1nTqSWHrjL3Url05/eUkF8yMIxU7G0o46vrzTeZ5XzudZxj65DAUh32YdaY43CTP5cicoWUCUpxck8o0A8A+CqHosZ5OsTCUspCCVtJZPw28Be7h4eExTHFOLXD0O6BVfu2Tw4QC/zJpUiaBKjFX7poUbausJQv/qwUNAD1lss9U4pErb5p5vtyNRhsCzE6ticZ6uxbL8ShZdZj4t0GjWGO3LiTLY9OcaFsrSetrTYxIO4OUTjQrQav4AOCLr0hSl6UsuaJQJUE5IaUCSR099kvNFaUzS+Taishi1LnbvN+UgK+aY9KyL7bKPcijSkpO5PSFyU3muGulJlcwVheMirbV0qymhJGKSKnlCMRafxcW1ch3063Scl+C7VOr5q4psXOvyhZPQWVQAeD6GcZN/8HqywDEVomqJbh9pc1HDQxXXy0St90kwp+fZOvi1/UimnTFVKsiXXZCrHrXYElKltfVY6o8MhBrnR6olutIrzBPc2GReAVPbKyMxsonEp84TLAxF3plkzw7/IxwAq+oVDwSbkbBVuyGJpkTJgosDjnO2WNborFHVpk1nhkKebXFm5WZGT7r3HyDPau1haUAgJEvmoV7wqjr6Jok3uCifOOb31Yk9+Un+4wr351p90itZebvazW0it4BsVK4/F6Jxph7HibJmRevidHjVfasJk63e7k2Ue57TG3BdEvWq4fw7DqrIlXccdHqaPvrgz4VeAvcw8PDY5jCv8A9PDw8hinOaQglNa0blZXiCmpSr4E+j2sXF5wTfePKzSXU5GZ+vrmwW9qtzFuTir0kTKSl2OxKZTDPM+yj+MED77QTCcMERxItGcXavJoEfWSblfGzdnhftpy/dhIBzE3lpNoLWeXR9h0zJYm6ucWuJ6PAXOj2sKS3/RBpIh+W43PHlZPbLISyqV3ct1nTLAzBCRrFvTsXRNv9WZIAZAErDnO9c4Gcp4YjACCpUeY9td1Kw0vL7DuFSRI6mTnKzlP7hX7yuTuisd5UW44voxQAMIJKrrlDiXK+H2235KPeYw6RHe8zd1eTfk3rbA4aU2Q7vnzodn+q485JN00oAkBlKFz0cn1pNKYJrXRKHrIAmpa2P/CgzWEwxxKS0yok5LRjoyW0ayfIeXDnF03kAdb7c1KaJfVqimRtczL0IJWR6zj3aHxytQmGTZsh51FPz6Puf3WnhYTGTeVuQnLNa/rs3AdeFgLAyhJbu9xDNgibSOdtsrG4m4moEIYxmvssFJQSJ+G2Zurs0zPSruPiEpGE0O5aALCvSEKaLHDGSXDttHR5qdUEPLnG5uO6+VK78GSzJRxV674q2a6Nn52aRDkmhwc5iX70sCRBHcnWB+nyj19vmWuDeAxDwVvgHh4eHsMU/gXu4eHhMUxxxhCKc+5nAK4DcCQIgvPCsVwAvwFQCkna3xYEQfPp9qFIiuuPlAS3x0soQRvRAsC9feLKc4k5N+nV1k9cjs7KhfNDvWluO6XZ4Z2kRTwwarAaWcIhy4B//mZxV/Z3WziCWQgH6sUtet8sU6CbtsAYBU+HLtbSbebyd2/LBgD8+B2msva5qpuibVVX6+szJsdQPHGG6w+ZLZkWHqqeaCEaZfNoo2IA6O2R6xhXZN/R6wGAKyslVLR8j6kmstxBz4Cc34gdFjJqGy8uX994c3u57H1aiswNywR84d53AQByLrLz6OJms2GLuqYec0e5RFlL/iM9bwA7uuQ8WW2yKtHOXTn0VdPNblEmTx01HVZ+PWCMFm5kzDxvrR9YmLs3GnukrhJAbHs85uqrimDPOCrpn7Yy2m4NQwU7YGGIkeknY/4PxJbXv1grWtUHMu04Cj53ZkYdThP3nWUgrltgEghtoda+I179vZuEn89sqfREu+/67DF/Ov1yuf/MhlGWEQBs/a08z8en27y2tltIS8MkHO5QPjxrc6teNwD0hbUYqUW2n87jwpyqJm13bmV25LgcP3mUXW/hRKur0HAePzsqQ1BYbn+XNEQbP26px70DVE0zhc8zPL+zETk5Gwv8FwDecsrYZwE8FwTBJADPhf/28PDw8DiHOKMFHgTBC8650lOGbwSwONy+H8AKAJ85074SXT9GJ0miQrnJP1x5RfS5y+oZ9B0W0lHLijvu5M823qpqbbMetGrrbq+25GASVYFGwkfl9p3HGiU5uCDXKvLWHTNrSK0/brI6LdUscK2kihHAmiLn+ccWS6pwUo+tB4U2gwWArDKxFBYRZ1t//a/OMa7zJYus4u+r264etE/tWsTWEHeDWbZKrKHR08y642bEyjlvH2vfmTdH+LlbG6zCla3Yj619B4DY5HTibEn0pZHwFHOhu/rlOJpkBhDVEABAQoFYXnklZj2q5cMW51dL/hBt37L+AwBiPTy1Mpk7npE4eB2yldlJFZAKTgSrNvycPOvMwrr16TPFq+xRkSYAP6paFG2r58X3oHGlJP25GpGbSKtuOXtOOVmy/vheaz0CYJb3paONX72nw7wGrUINum0+75z3CoDYLj4wCnTURag0wazUT+4RwbKOfXa9mxLteUw9IWugY7QdhytKNen34fHmpaxpk0TwpyYvi8a+vevyaLt1QK6ZE5JPNMk9UM8VABYVmmX88CG5JhaRGqBrzy0enOjWegptXgzE9gZYfUCSvVyfwftUgS5uRh51JDsyuPLzVLzWGHhBEARKIDkMYDC1wcPDw8Pjr4rXncQMRI/2tOEa59wHnHPrnXPrO5oHWzYeHh4eHq8NZ6UHHoZQnqQk5i4Ai4MgaHDOFQJYEQTB5D+3DwDInDw6mP2/wrfW9lhb26yl2q4HpAS5vYTOqcRcqenFEqbglmrLauyw75sipaeaRAJMZOq6CgszsDurDWrZfdcWY0lZloj55+nP2z57ZJ/M0+RGudqy64EN86MxFd26jESNuBHqrWMkefR4gyVoufxWXcqPly2Pxr6w8q0AYkuqG9osdKGly+rGARa6YreaW0wph7Wi0Papgk2AJRe5TPz8BeKmskb0wU4T/1qzU9zdW2dtiMY0KczNftVlByw5lUAhFhYm0nukOtiA8ZlPp2+dHI7zmtH7zvz6gEJK2tpNw2JAbEs1LdnmpsfKy9eELwDUv2LhpeTpLQBiW7uxYJiGglj8Sxs+c0iRRapq9ooTzEk5FX1jDvoze6dG2xo+VM40AGw6aqENnW9eHxpO43uhXGj+nGUgFk6Q/XM4jMvW634m19R0qR1n0WQL62jZPYd3rs/bAgBo6bd5/3WtPY/XFm0HAIxLNs73jk65B7/bZc8/i5RlbZdzT7zWEpIcYtH2jywj8cEJLwIAVrWa5MeGwzaHQz2DwV5LxvaNkWv+6JwVOBU/WLMk2q79x8++oXrgjwO4K9y+C8Af/szfenh4eHj8FXA2NMIHIQnLPOdcHYD/APANAL91zr0XwAEAt51+D4aersRIqOf3YTJsVr4J3LQvlF/l/mYL3qckmTWlnU40EQoAPUQP3FUiVggnoTS5wGJWnMTQhEN9hyVYlKqlVjEQaxlH1ZSUELphpllT2qBYO6sAJoCzY4RZWkqLBEzWtCjdknK7t5t3Mmu8WCT31loz4vdf8AIAYPVxU/5hmllUZVpholnb2mWfarEBQEeGWdNacapynEBs8il9gpzfnIvNMtEGxqPm23xxElOTyntJulPvJSccE+lea3cUtna4yXRH2JxXK3oBEyaqpDW19NCUaFvpaP899+Fo7L/3XgkAWEzn/tRO63Sj8rp8r7ny9VizWOPanBawhFVA6yyO+g+rfKuuEyBWQlkTWlzJq81xmSLJ93D+dLkfa7ZbMrw19CCZwsiyp1oFyJ19uKsNwmQaz7t+R7tfAUDVAUu2a2VqHHlOCk7KH5tg6zyjUzxurcgEYitKD3TJ+bNHsiZxsFfXuM3m4+chJVC9dsCaN/c0280oIppgwXSZb/Y4VzbYMW8fJ8fSZwgwMa32l807KFhkx1SPKa7Ojtk/xjwNJURkxdt9vXu1NJNG75nt67NhobzjNB9ddppxDw8PD49zAF+J6eHh4TFMcU6bGieXjA0KP/1xAEB8Xth0lMIlKuIzhpKU8zJNiGltmyQCnl9WGY31ZZmr9ug13wcAvH3t+6OxlFfExZ14syVF6tqyo21N9NTsN2EgTQrOHWl8zkNd9p01T0qlZf90c0dZ8ElDLKm5lticMkp44OyeMTQppzxbAEiLs1DQvSsWAwCyS1uiMdXC5nNj1/P3B4Vzzp1EtMnv/jWmf/2WK9dH25rg66q3REv8SfudT+iSsEDqbOP5aoXjpuWWHOzNtfui2tFcwagJ4ph5o2bUGj7g5CJD+fCcHLpqoiT6JqeZe8+JrfNy5b6qljhgQknanebUY2pz3lZqFs3Vn5mjZA1wR56iERLi4/BP/SELY2hD3o8ueTYae6LBBJI05MEhGA1NsH56b6PtX5tUc7Ng1XFn/WpODqrQm1YoArEJca3wXbrRzk1DRVdfbmtGdcMBS2IePWhJbO3uxKELFljbtlrCDP1pNocq6AQAqVnyfa0iBizJrRXMANC72Y6Ze4GtAYUm6Hne9D0EGG+e7zU3yNb6AU6sRw3BqUYhSLJ3ql7n5m0WLkspMD65JtG5AlufPZ6DNzqJ6eHh4eHxN4Z/gXt4eHgMU5xTPfCUtJ5IY3jqCHFxnqszTexd9ZJF7i0wd4JDKMuqhREwd5GVxzKL4WfHhKGxeLy5iS/GS1aeW4ixG6ksAxVxAoDqVgmnsDZ3R68JKXWGok2p5EoxYwSh68Ml/41hqfO0fCv955Za99YuBhDLLXek/RykyLFaTxjvVcuZmZmwBRaGaKsSF7p4jrEmVBCoL8POnbnB/bXy+bwFQ8/xllqZk66jdm2jwnZf2rz2VLw851cAgHfsvSYa21QloY+D8dnRGLMuNIzBrbtKsy3cosyXgOZIRYJeSSyNxtgdbs2U7SObja3QXCbH5FJm1qDXegVlQgC2PgCgboPMN7v/vaFgVP8FxpbSMAAAhP2pseq4MUaYhTIQ6tG7HguhaJjhv2/5f9HYvYesnZg+O3eNtxDcrjRhPD270dZUfbExit5ZLkJcP11tZfzMglJevdZKAMCUabJ+mdXV3E6l+iHiqC2itiYcW2zsIG11CACFr8jzUn+RrbMkEndSaQFuHTgqXfap1w0AEy42mYEZOfJsneizsM2mflm7doVAOoluafiJ26hxizmtWchKtM+Xt0rYMJNCmxwq1JqVpEp753C9Awu0KTQEw1Icp4O3wD08PDyGKc5pEjOlrCgo/eYHAQCfny6iNyxAk5UqVkqM+A6hvVUsqOwc+2VqOZgdbWuXD5ZufLpaJF2ZY8xiRvprmT+mJRrTX+J3Fb4cjU1OtOTgdas+AgBIzzCrihvDpocdf1hWsj2UdOUqvqYas+r03DmZ0dlq1oNLEAuPhYW0OwuLd3FS8IZCqVb71gsmajVil1hVs99ulancYFar1Hg/3EFEk31luVbhdn62WD4nB8yaeOhZ46sXTDcrRnFL8WYAQE2XyZsy/1qFlg5XmbU783yqGNxt4mKK+VPEE2ALiRNwFVMODvrOviY5/ozC+kGfAdZEmhNkExebV6hJO054Fya1AABSnK2zu1feMOg8OHHKXGsFd5sJDsnaH1tp58kWqf4ty6uqWJomK0+FJi8d1Su4Y2QRhrLLZYXGldbr/GPttGiMk6QqM8sCWcp35ybeDzZYI+SBt8vxD//U5ljlYgHrdnT7BOPqawX3vhZbP0177Xn632t+AQD4r/0mpKqeJIuzMSJCA8kAJ5CUrnLKucm3eoW3FNi5sYhcd5d4UzyHbNVrvcWEbCMFDIXfLbjHJzE9PDw83kzwL3APDw+PYYpzGkIZMbkgmP9/pYmtNumtrzFedE6RuM2cqGNo02LG/Apzq8vSxU1hrWINc2j5MgCg2ly+gYnicvY3GbdXu2uMJU4tJ/LUhWI3jxNbGgJqO0zl5CEXlkMTLFalyZRUCu9w9xQWbVJoYpb1vJeUWfJxRoYkjX58//XRWOnV4v6z4BOXo/edkPty67x10djD620+9Ty4Ee7uRnE5lYcNmHAUYIkkFoRScJiJxbBUOIiTw6o7Dtj9YHdWO/pw6Cn+uN33f7xKhMBWHjXhod07Zf93XLQ6GmMRsih01WnXyx1qrpgq13yIOvqo9rzrMvuI+fttYWKV9dGV5w1YSOpX1ZZ8VL1vlolgsazrCyUk9sPlpq+vXV6+NvP30dhnNr412l4wTtbCliOW+NYwJgDUH5eEZ0zpeamEzr5d/rto7K5174m2+3rDkNJ4Cymt3SzzfcXcrdGYdhACgMw/yHPSdJUdmznw2gGHJQE0cb+wyOQo/rTU1mlvntyjBEqmKs/7prF2Hj995RL7PNR576C6CRZ1U2E0riPQ+8EiclwTwDx1BYuD3VEizxnXK6gAH4eKt1z/VR9C8fDw8Hgzwb/APTw8PIYpzikPPICL3D5tIBpHZdrqAnPGNifFOKiqK/0/lzwUjf2y4cJo+/nD4qrNLDG+qbrY7aT7mz3TwhjqMiYVmCuv7tDWdnMtGRPyxKWrabMQB7fcag2l5xJGDM0OUcwiXqxyQ9tg56lqgwDwVL2waRr2WoY8MU/CP71ddhs5TPH9/ZcCAJIvtNBEZbYc89cvWAuwjBLjK6csleP/foSpL4K41pt2lg66DnVTmRvMjZA/Nl9CF9wCbFOdhBnUjQdiGxhraIPPjZX21N1tI573hAJx73c3WtjFjTFWRkGihOiYMZJTKeuL+fezJtnnyi74ZZ2ts3aqCVixX9Ycl2drHcAVF5irziE2ZT6VjjdmE4eCNKQ1KteuvTN8NpiV897LTaNeFfI4VNMccvX5GeHS8OoWWUu3lm6OxjhsoyGxJ9fMjsb0uf1mnbE7mEs9c5wwXyIVRwDlU4S9wTUdzzZaa8GkbAmXKP8diFV81LUQn2PH0dDE/EwLoTyRVxltK7Pm6vLt0ZjWO3AIjUMsuk9u58ayGypjwOqex/skHPt8r+2zIM0YaRoiXLunFEPhOxuFhceMtmnjGgb93ZYhv+0tcA8PD49hi3NrgQeUDKwNLZZi++VRLq522wCAux98e7Ttxsvffm7LTdEYd0KJdL6pP7DqSrd2G8eUK/o0Gcfc0M6TksTQpAcAXFBQE20/uV24xZOKzYK6uWBTtP3NRtGYVn1qwLjUXNE5L8v2WTZtcAKWK+TuvCDkpBO/Vrm4qVkmqsWiOYq+rPhB39GKWCC24XPfdaH1sMeSMq7QLJ8Rm8W7mP8Pdr3rDov3oOJHAHAwKzva/tU+seq4G9H1lXKPv7js1miMqy61wpL50WWUyNWEEidTVciLk8OchNJm1Xyv962TRHBhpXFzeZ8K3mffIVtz2nj40sk2H5q01cpQIDahrVWKp9Ppjtsrz0YnWYJqHaaTjjbrwF89SiqJuWPP9TNljveRDrt2rQKAH25YDAC4v9WStuxJzKyQxOq+Cvu+Vj4uzDXyAG8/dGAOABP5Amxd7Oy05F8h6XCf2C/eGnusWkkJALszxOvg5/HYLvF+H8w2PjlXU6snylXGioPN2dE2ezlKGrjqfBPq+v1O8kRD6DUylEMOxHpTSkq4Zfrg9QEAqx6TNVl2tXnjCiYXnA7eAvfw8PAYpvAvcA8PD49hinPKA0+fVBhM/b5wRrU1Ewu2tB2ThMDiCuMycxmwitokxVnCkaGtrJJTLDGhCUdu+8TavZWVkgRhPnL1MdkP6/6yprZq+GqDX+CUJEQoWDUUd7x1j4Vyymaa2xS5dZtNCGfC5Zb0Ud42u2ffqRAu7j+uenc0xpxwBfONE1NkP28jl5/DNsrfbe6yMAELHOn9YBEeDSVxwnlUsiVyrsmWZN6nXzUOss6RzvWp0NAXa4SzQNJAKCnAbes0DME8XxahUjG0qcUmVqUcduY6c8s95YfHcMO77L5qKKBpnSVwewpkjrUhMhDLtdYQHjdc1nMDgITE/sHnFHLPOWynkg0A8OoqCZ0EpRZ20bL6d04wTr8m3QBrEry1wc5N9eIBu+9vz1sbje3tkXDG1nYLuzGWL68EAEy+oCYaqz8ha1pL+wFLoAJAyjflmTh4ua0pR4+4NoHm5skaiuSEIZfVa8iyodqO86nL/wgAuG+PJfCZ861EAxV8A0zrHjCiActyaKl8XJy9RycXkQBWKJ3B+uilE+1z5dqzAN+z6yTBq1r0ALDhmq+/Nh64c26sc+5551yVc267c+7j4Xiuc+5Z59zu8P85Z9qXh4eHh8cbh7NJYvYB+FQQBBudc5kANjjnngXwbgDPBUHwDefcZwF8FsBn/uyO+uOiJrClJfLL1tppVoY2Sl251oRyuGPG+SMl8fbwJksiVEy0ZIdaZRmUEFLBoLw8+6VugiWUNFGQUG0WZ2KlJB/513cTWRzqCajHAACdcfZLrmCrXivGrplm1uEP65ZE22r5TL7JvA9OgikuLKqJttUaYtoRJ1OfOmpCToot6yXD+3yW0Z7YergiVzrdcMKpm5J62hUlI8ssvXcVS4KVhck2vmrdeUZfJ4ki7nSzIUwUc4Phr1WbCJBaQVr5B8TKfKpHxVK6zdsk2fZwf2U0xlafWuictI0SX9bYB0u3VUTbKsTE6yyDLF/1VNpCKxEASkNhsxWbLYGWP9YS2noPVCwKiPUKdM0uqDQPTGl5txVZgu2ry2+Mtl2K3MMSSpyXZooF9+OtVm04QM2Zde4yiAbIVv3aP8j66bzOPFalf04qsGeD5yOxXfa/Y6N5zurRTCdPYPWzJlxWulzWT++dZmCmHDRrvHtbNgCgea4dR7vr5CTb/Wfo53dcYknb726WNr5sIe9qL8CpYGrq2GKbzzWdshY4maoWOO+TO4rp3PDa5Xv4bJq865hmqs2kjx7KHvLaGGe0wIMgaAiCYGO43QZgB4AxAG4EcH/4Z/cDuOmMR/Pw8PDweMPwFyUxnXOlAGYBWAOgIAgCNf0OAxj8Uybf+YBzbr1zbn1/25kFyj08PDw8zg5nzQN3zmUAeATAJ4IgOOGcuWJBEATOuSGzoUEQ3APgHkDErLTKUvm3iWUW2thzQhIOkyrMXf3kOGv8+rmqmwDEVsoxZ1fHL8uz5FFNl7jV649ZF4wPzvtjtP3Nh28GAPSVmyumDiMnRQaoY4smTmpgCRJ2TbVilHW6e8NK0CczLKyRvN/CRwNTxb1n7W3WNX9iYyUAYOkRcz2TZ0syhUMC3KEkSl5ShZsL3XZuspuTb/fgaxskjMGJYG5wnFggriAncF9ola5KLITU02rzce+miwAAASVYU9PkPDoGLPTEmshaddlHfGJtSgwAvcUSLtvXaBzlgbCTCYuMbTpqc6Nc/uRCS0LtOCEc5AkZFqphDWhtdn397M3RGDfxVTEjFjBSxCSrXrFE4UC+3GNOlm7fa5xfvR81bXaPVETs2eMWXmTevGpyq6gVYNWZnHTjxsDKUWaBLG66nb1Iwjqsr56cLN95R6ElNh9ptErN7inyt0GrhUCunb8BAPDDlSa0tegK42yva5ekYk6+3f/SSYMrl7OSugaNaaciAOjLtmfn7ksfAWBrE7AG6lOnWLhqR51VDO/dKPc1d5qthceeuyDadslyj1nHP0iRY/JzlzHOQj0acuLqzv3d9t7QJD6LXn0+fD+tarUw5y8xNM7KAnfOJUJe3g8EQfBoONzonCsMPy8EcOR03/fw8PDweONxNiwUB+A+ADuCIPgOffQ4gLvC7bsA/OGNPz0PDw8Pj9PhjDxw59xCAC8C2AZA/cTPQeLgvwVQAuAAgNuCIBjs9xCSx40NCj/7cQDWOLY8ywx31VRmd5bdVc3kLsg1AZv7thqns6RADs8t2dS15bZP88pqou1/Gr0CAPDhTXdEY9qM2E20mD3zNJetkvJaLmtuO2JCTMr/zUwwl2+ohqhagg5YA+KELnOre8rsb5VnzqELdVO5LFm1hAHTUT5QbW6iunwgbvj86Raq+WyRtLp7+9r3R2Oq8QyY0Jhy7gHjjjOnmzP5KilQd9SYplzCrPi3iUujbdWt5lJ4hnKoWVdaW+UphxwAqvcam+bWsES6MMnu2707Zf1MGWXhjmkjzMXW1nDsVnPZtLY166MQm1575hCl8ICtSW5+y23NlFXEPPEXX5TQ2QA1T04psOvU0Egc1S58eobMp0oIALFSDtri7N5aa3/HomzK3+Z7qQyMpiZjcgV9dm3KLpueY+wfFTnjNaNcdwBIXSXPTtt8m69rp1iIRaUAtu+0d8GsacLQYemIWy60sM6UVJnDu1dfZ8cMQ2MavgOM4QIAn771MQDAT/ZZs2hlNgHWHJzrO9q6JCwzlPY/AAzsl/fOAEmGsGxCMfUcUGitB3PU99z2pSF54GeMgQdBsAqAO83Hl53p+x4eHh4efx2c26bGE8cEY7/xTwAsqM+NW1UOlIVsntpnnFwVu+JfQE5iNneJpdDcbhb4W8tE0EelZk/9fH6xWI/KeQWArnyxYv7zWus6wqJLQzVPHp1kFuVD+ySpw0JMs/Kl6pI71SyrMa60Voyq1QPEdhtRDNX4l5NQ3DBVLR+WP+W/VexfY5bN+PkiYHRVQVU0trzJzlPnWC0PAGivlXNm+VTuWnK4SarN1EMCzBrvowRraq5ZYGoRszXMIkDarJoteZU65YQyV8i+u2wNAOAXe6yqUu8R83RZ7Khnv1hW/bmW1GXvhROVCk1yHyQpUk7AqgfHPPDLyy3xrp4oNz0eSmCtlSzj/lp5jqbMrYnGvlDyJADgU9Vvi8a4A9biSuugpGAxLAXPoZ4H8561mhWw55rnUy1vXa9ArKBY71PyeeW7LAGbl2RW6qMrwvtF86W1C9poGIhNwGqTcRZYY36/4tZZG6Ltg52yJhtPmjXNPPMttZKQ5AS/VmVzw2V9/gFrUs4iYur1AdZpiedLE6s8X69c9U3fkcfDw8PjzQT/Avfw8PAYpjineuBwQdQR5Mox4jIuhYn4qLujDW0BS0wBQ3MmOYyx6oQkNPopoXQ07PJyKeloP/CKdSjZkiT83GCOueIlYan8/+43Pe4o+QdgzmgJh3Cp+8h0c7XUVecS5ue3VMr1TGuJxtR9Aix0womLyWkWPvjOn64FEJvEKi8LE0YZ5kayVnFXr9ze9lZztQsL5PgcRmLs2RLypkkGmcurNRxySZnNe026JF04ocjXphIHja3mmiqYX83hJV0DVSdGn/oVAJas48bTmgT9t3UW7uIGsg09cu58r1QYrfZFSyh359u9zguTlyq+BgAfW7gs2n6krhKAhW8AIKdY9s9CS1wqHYyUMFZA4aOX60ujbXX/+TxvLZeE43NN9rx8cMKL0fa3t5lQmGL1SQkbajcfAADJJmiSfVW9JQL5vqqmNnOlNdTD58sdoVrDjlIs3qTddTT8BsRy/vf0SOjjhdUW4rhkgXXSSS6RcMrlpSYz8eRqCVOcf55x/jmheXSsrA8WWNMkJvPib8m2sva7T1jCU8Ga3NpUvZOaq//8svsAAKs7LETL903lDBak2fvnkYzKaFtDmpyw7g2JAhxKfmXQmQm8Be7h4eExTOFf4B4eHh7DFOc0hJKS0BexC5hRoBiqlZU2ZgWA+VOEldHea2EGbpmkra5KplvGX/WJU6e0RGPKIQWAzfuFgZFwyPbZ3CDu8InzLfuuSomAaRmrpvWp56QtxPpOGPsj57zBFHl2u7XsvXL8wWjsWy9fZX+cFXJY99s+V/ZLWfXCJeaeVYwcfJ6aCQfMneYQx6XXWAu7B5dL2Tu3EOM2W4vmCTvl/BE10djlOTLGIacYtz0sGe+uNa683qMlWcaEyI6zc3p8t7CCuAT5/or7o+2NBXLfHmywllo/rl0MILaVneotA8DKfgl5afgOsBBbQ6Wts+tK7PNnaiVkwRrgNV0msaDsgZZtFmaoC9vJMceXw0caxmImBv8ta74rHm8Y3NqL1SYzLhS3e/eLpdFY/UxZXzHSAvEmLaC1CVwjkTTawguRjjc1CdfG1KyK+XxVZbTdkxvy0Um7XeUOmB3EPPKCkxLW4bqK+g67b5+qkJDVV9deY9cbNrtekG5rf+UUC2Mo/38eKSBugITJVEEQAO6utbBJ9epSAMCYucZh56bIOl+sda+KoMw8UVYWALSWynvun4/dHo2xXIaGLNun2XOtDK/mrsGMoFPhLXAPDw+PYYpzygNPLSsKJn7nfQCAgZclyTHmSrMOVExmgPieH52zItp+plEszoY2+/W+odQqtn69RSxfF2/XxFVPCu7Y8lS9JE7m5Jnl+0pj6aC/233SklBqcTb2mpXACUvlNuePaYnGNKHI4GSHWop3TV0Tjd374uJoW6vulHcKGCf3dBaneiT33XhPNPb+330QABBHtOa+8eZpaIXclyqfisa+vOH6aDt5i+zzZImdx0cXieDYD5ebWJHrM2srrke22QtSKyWlyHjNfG3q3ZSlW5Xpoa7saFs73HBC+4Nhou9pEvziTkzaWPbh9Uan1XvE94eTvlrNypWF7PWpLnbVVkuCqnXICVSGWu1sgf97qQms/ct24W0319u91I483HGJOfCaMGUOc+M2qQMYSLbv8H3R5OA7y81K5Q5G6hGrFjkA3DlP0mn6rAFA0G3XMb9CGhz/x5gno7FvHhZPkiuTl/3evq9r8aJbBjfKBqyD0abdpjGu3jhjRqYl/bThMwu96b3WhsgA8JErrfp3bWspAOu8A8Car8PWL3cG0vV1x0ybw8drbP3pM/qDeQ9GY6vaTWBLm4xz9fCifPEqnm20eX/h8m97HriHh4fHmwn+Be7h4eExTHFOQyhZUwqCi+55OwBgSZ5wOn+4YXH0uSZ3uOxYk5AAMCLMPc54j4VNuORWucNcCq1uOZe1cxJLXZguSropj3xWsSVvuPGrctPZfe/YZ+6uuqacKFQXdyIJdbFGuYoicYKDW7Lp56ppDVg5OfNamTevrmdz92DONwtcVU43d1SPmRpv++E5fmSVJA3ZFVdesztmiRidA8Ba0CknFrAkFZfs/2S7iSpp02Vu57aywZI62m6KRaA0AczzxiGYX78gJcxc8q/l1SrCBMRKNSjPfN1+c9+5xFnFw7JzzAXWeeek/FBt3JhLzW65tuRautlccXX/O1ZZ2Xpgp4mCRRI+YD1wlQzgMKM2MgZMt94lDNYyB4CUdLmvrAefN1GurWkvJVqzbK38nwtElPSXdVZrcUOhJMlTnP3d15+8OdpOPi5rqfhyC6dW19j6VP42l80rOog8sHuncbZ1TUft6wAkjJHnMZ1ayHHbxDXbZH0ltNjE9uXas6VSD/yMaePz28dZSf5QpfLa4g2w5smASV9wTYm+q/hZ/9GcB30IxcPDw+PNhHNKI8xJOBk13dWuFPorD1hV0rM7LEnkksxDaAk1lTgZogI0gFGHSsebRK1aJNqdBACWHrLkgMq4ZtSaRXntB6TC7dcv2S8pS3deVlwNwChmABAk23n275BEUs7FVm2m1hhblJxwLMoVq66m1iwkRtEY+dXWBCsAvOc8aQarCRsAONJhv/Rb18ivemKHXVvmXLE4Fs0xy5cpWzUtYm2xhChbPrcsFMnOR4k6pjRBpoFxAk+bUav1zvvRxNGpiOae6GhMsVx0vlQMPvKy7VOFr9YdM2u5qp+kdMO1xJZvatbgLi9MgVRrnDvM5BQbtbE5Xbwb9uBUVImT1EUlg9dCV1V2NJaxyCipN+eKMNKmMWa1LyqU6324jBKbZDnXNso6Xp5owmMqcbo00dYpy97mj5b7xRTGOhLg0mRf4hi7Dk3ANpH1z4lVbWydSlaqJpVZBjq+xJ6nvi5Zc+x96rkx2LNav1quc2ylJZcTjw+2R7mic/sxWQvcLHgDrXOtUuUuYXE0XxeXSIL2E6Oei8a+ckhoiE80kBheh62V+iZZHywc9tJy86xq+oW40Zdu74+EslB++aSd5+ngLXAPDw+PYQr/Avfw8PAYpjinIZTjvWn4bb3E4TXhoNragLlQmswEgDvmGr/yWy9Iw90P5a2MxpafNJexbJ6EB9iFZnEnBXdUSawU1zXnAnOLVTt80hQLHRSlm0unyTTuwlNOf6vXxuI9Gh56Yr11R8kcba6adkL5+eKfRWNdgbnlX9p1A4BYbvrWNjkOa4i37rGQ0kCmzGfBdEucavXn0XQ7d24MrF1kuGvJziPWsHXzZgnXaHNjAFhcLrxV1gBnvXANGzHn+/c7pbJw4YS90VhPu7meGvRhTj/rPW/aWSobJDKWFidzyHUCbUfsnB676gcAgMP9Nl8ffulOOXaGxQRmjjK3XPnm8yqt4o8TvB1Zcs6coFVN7e5uu3+cOFt9YDwAoG+MjTF/+1i/3BsOKWiCfnGFCTppPQJgXXWYWx7tr9nuddaIweEQvr+drVYhHR+GxrgaVsWhEvNsLJ66AKkYGze4VqEmrqVQ/XIAyDgix9nXZPxs1f4HgLGpElLgkGVS2DQ7pu6B6hlU2IrPU2sb7l739mjsgrcYoUHvNT9D3Hh6U4aEtN53/J3RmIbbEmgOPjXXGrHft0dCgUkk1Mb3PSNLzq8i155R5btzgvV0OJuemCnOubXOuS3Oue3Oua+E4+Odc2ucc3ucc79xziWdaV8eHh4eHm8cziaE0g1gSRAEMwFUAniLc+4CAP8F4H+CICgD0AzgvX+1s/Tw8PDwGIS/iAfunEsDsArAhwA8BWB0EAR9zrkLAXw5CIKr/tz3mQeubIm7xpvS7bdWvQUAUD7ReOC3FZler5auP1xTGY2x2NWdc6QM/TdVFjZJ2iZZ4IU3W5luTZu5atruiVkGQei2p+aZu6mt2QArlWVBH9ayfmJjeH4k3qOumDZzBoZuIMt64HlZxqVm4SuFllIfrjLXdCDLzkPnY3yyuWLf3i4sARb0iSONaA2dqCY1AHx34m+j7Q/vfgcAoGavud2K9y+w0BYzY3ZsFJeQRcZuHSNMi9Y+46hrSAgwt7mpx9x/dXEBc1lZd1rLzRnqogKm417TZhxmFZnicnLGz18VPnNwyMJhAYV13rHkJQAWvgGA5vCaVJALAIK9FjLQtnUM1q1eu3nSoM81FDBArAiWUFD2UsJRu6/ahFdrCABgVLqtKeWmcys71tyOwn3EMlG+PDcjZ1aPyhmw+JaGgljWoOQhC/U0zpdz7hlhx2FGk5ajzxlnPHENY63cMC0a43Csrl+e130t8ty30zOmQlkA8NXlNwIA8kotHMZ/29Uo9/COi6w9mj5bP9ptQm4MLZHn2hTm1cfnyfN20xR7vzyyTeY9L8+ewQ3XfP2188Cdc/HOuc0AjgB4FsBeAC1BEOgTUwdgzGm++wHn3Hrn3Pqels6h/sTDw8PD4zXgrJKYQRD0A6h0zmUDeAygNjpn/u49AO4BgFHTRgYTwkpEbYSqiUkASBghVgwn/x5rtKTfrnqx+rihKovALMwQfvYfM+xXuT1T9rViv1k1XGGp3WLy6VdXeaIsrsQWliantGMJYMk9AMifOFg6VhNJxzrM4mzsM+9hbE4LgFj+rFb5AUDcEbEEksbbr7LyYuNGW/JmNAkcPfiq/GCXFZoF3nlc5oOFttjK0Kq9o2T5qtUNGC868bhZUCq/mZVg1g4n09SaOrDDONnPJcsS0kaxp+JgXjYA4JbizXacFDvn3+2SdcFNhbW7Ts8ANb2m81C+uzZUBkx47KevmBhZUiM9FhPF6Ihnuni5XacmzNWTA0wQLEaGd4lVQ6oHwBKzRcVmcWaPlevktdD1mKz95hlUNUklA65DrjNuvD0P/ztJBJQ+X3tjNMbJ0qQssf5U4hmw6l7Akuwd7ZbY1PnktV9GCbhvrxNBMyYiJA3IfDCPvyHHKiRLHxNP4tBX7Hq405bKuPK51+wXr3MWVREzh12t/iGFqUrMkOSKUM0ZsyRza49d++ZGeca5jqS1XdYhe0Zxe+2+J8yW+6XPHQBcSZXAWvOy+sj4aGxeWQ2A2Orf0+EvohEGQdAC4HkAFwLIds7pSi8GMLjO1cPDw8Pjr4azYaHkh5Y3nHOpAK4AsAPyItfmg3cB+MNf6Rw9PDw8PIbA2YRQCgHc75yLh7zwfxsEwZPOuSoADznn7gawCcB9Z9pRHIIo2adhjH0ZllBUt0eb0wLAS20W+tDEJ5eLc2nwcyckdMKcTBe6RayzvfKo7bNiojgOE0hkqjtfpuXZV2ZEYz/bvSTavv0KSdQoXxywZARgPHMOl2ioprDQQiC76yz5uKdL/GFtGgvEhiGOhyEULtNFSIHto+a45RMsXKIhFm5KnDlKklhNNZbIY7771FQJhzy6x0JbzA3WEufZiyzZpa7e99ZZd5OEcrtO5fRWx9l9UYEt1mt/duYvo+3zX/gQAOD/brS8+PdvM458VeHgZseX5cncPXzIuqNoE2YAGAjnqajU7nX9CeGWJ+VYjKSXBLK08XDCRHO7+5oppJAr90jDJgCwYJyorlVnWYyDk22aBOXwj3LDASA5WUInLOr2aujpB4dIr5vWhwp0cWjjE3tvAwAcfN7CCF2Fdp5637UrFRAb+ojWX4bNTX2NzFd9nD23nARFq8xXb56dW3n24ObbP55h4k056+V5aTts+6zLzY62tUORSlgAQE6yhLFUPgGILbVXZBKH/US8hDEuK7O1u/SEJZrnT5ewzaQ0S9CCdOC25Uiaj4W8KisH65J3FFhIUt9pRwdyBv0dYFrq9XRfF8yS9VOVMZgocCrO+AIPgmArgFlDjO8DMG/wNzw8PDw8zgV8Kb2Hh4fHMMU5LaVv602O0XQGYltZKfvj6Vxza5g9olxadt+VLwwY71a1pAFgaUiY4TLe0kxThtOQDqv8NbeG6mhULt7Xay6hhk4uHW3l1c/DzlP53TXHiCYQNi1mhgSX3CoH/hApkLFqWtF0cbdZV1i5v6x0x1lzLe9m3rNqDbcRl5l1lL/dITxxZuAkZRgDR1XxWFlQw0ccNrl2gqnA5RA7RaH3TTmvAHD3EVN/VKbR9hPm3n/61bcO2g/z1X9dK226eI4GqN2XcvG11RgAJIWqeKzRfLTDeLpFI4TVw8yo5iTzq9WVb00d3KSbQ2CsZqkhKQ6xVRRauGRTVRhOMdp7xMZ5It6ejQP1FnKIGEsUTlPueC+FTVj7/fxs4VWzlAJDS+0PdmZHY8pMSaBwWMtB+3zaDNlnBoXtFKw8mdZoNRK7/knWsUu1dRbNAYAHr/wxAODnTaYXr7IKLK/AzJW35og+91dOWjvAhZdKTUlpij0v64osvLRmuzBj1sAYMnHtdg/LZkrYtz3Xwpg6h1z3oK0hAWsPWTHFGDhcM5JSL/erj5hNKpvAOu72NMXCW+AeHh4ewxTn1ALv642PrErl2sZPNQtKE2wMFjhKmihWzKhk+85zddYgVLmY3HVEOc4rtpvo1RcWWMNV5ZlzhZpa4L1dNj2JKfarqcJU3GS3odqOOWeBVBl2lNi5a+XaJVmWiHmh1c5dOblDWS4A0Nwu1po20QWA3dWlcmy63qSpZm2pFXPlbOvSolrUq1rNymTe6q0XbwYAPLTPEoHx28yifbS3EkBsR56+E2E3oTybI7UiAEvksBWrVXGLJpsXw9a4zvfHFlql3PE+O2cVyFJRLADob5Z7XTjR5qgtmUS1Qv4+r6kZoeXLDXNvnWXdVa7Nkgq51R3mYf38j5bQ3hN6Kt8879Fo7MPPvFs2SMucq+rmFx8AAGw4bBx47t6jFaWsi19TLPP1zmJLxn+v49JoWz0RFsDSZH8iVfTycbRpLleZXhtqzPPn3FnqH8dJFaLq+QPAQ7vMMv7fCb8DALxl7T9FY52hJjaLtw1QLj4h1Ku/YbF5zvxc37laGqFzxbDea1DecgXMu1cr+6J88zj0vbCqnmo2qI6kLUPm4b/nPhyNffKF26Nt7TfAaB4l17Z3hXkMjqzpaaEnySQJ5tpPXiwJSxbyUi/90T22tqX8ZjC8Be7h4eExTOFf4B4eHh7DFOc0hJKb3oE754l41fnp4jr8545ro8+1dLi61RKOytMGzCXktmQsZvWx+csBAD9YfVk0pk1vORH3jU1vGXRuXJ6vGtXs4r67zFzX5U0SjuGEIWtdq4vEYkP5YWPgJ5rMLTpIpb+qqbypzo7peu33VUWG/nXU89HYV9OkbJnFubSBLAB8u14+Zz57dmkLTkVvyeCwDV9vQ4lxqUtTxBVkESrl1S/Kt3DIU/XmJl46Usa1/B0AEjbLvcy8pSYa41Jr5f4+02j86Rk5g4t944jzHxcm7RaM2h+N8VpRt7qmy+ZLm8lyezJOMn3/kKylhbmmW/6ea5ZH25q8+lzVTdFYJIZG7QJZx/2aTNl+aYS5/D+qMjGkoE/uu+uyBFpNk8zHb2F6RnyPugYkPMQ1DhoK7OukNn4w6DMxcbbNO983bcTNPPGvNl4DAJg2zpKumqgDgMt+/W8AgFGV9jxpE+DmeltHow/afDdeIPNVmGRyAi3NFi7ThtHXlJj+ua6lnlYLkS2bc2+0fd3WdwOwRCxgmv6sVd7ea7Gc/ibZ19PNlijmdfHR+SsAAA09dh07Tkg9wj+81dqs/ap6brStoZHKbJLvIJG7q0dJonJXhtU1qHRAa+fgxPip8Ba4h4eHxzDFXyQn+3rBcrKarLs4x6y2re1ifbLF6HLNiklOkV/yCXlGA1JJTACYNaVGxhoo2VAtv+TJ01uiIaaeaZUi05GUajcr33412SpTCzsxycY+Ms2kVJXO1rDXEj1K6eqhKr5Z08xS3FYnST+uHFtYZAmYDU1iBWkyE7CqTJZMZeqRyt7ePmFjNKaJKf5158pVTYJxY1+l0gGWiBxLjXDnZdUAAH64YXE0dsVUa+KqHpU23gWM3qfNmoHYysOhvsPdZOaOFvrW5DSz9H60WaxYXSdAbAJO6YUs7Zpxnhyfq/w4marWH+PELjuniy+W+V651jyFhHaxi/Jn27mpNQuY6BJ3hmJ51hFh026m3KoYGtMir59p3ta2ZuEc8nzpPKSQKFbrCVs/o/PEIm1aZ7TKogvMRtdzbu4i2mQoz7q1wTiO/DyqxcnzrtWwejwASPyeeUEHr5Brmjr7QDS2o25wpS3ff12zXInN63T7Xnme4qiq9iOV8oz+eKsJlyVU27W5804MOneu7tTKV1776jGzQJ4+D4A9EwF5QeWTbI53b5fzvOMSk6jVZCuLWdXc+fnXLifr4eHh4fH3B/8C9/Dw8BimOKdJzCCw6rSqFnHbytKNs6tNcTlsMrPEXBNNbHFnD070qUBSb5PxWjUQoGERIFZESl39nX3mRo7KbgEAPLeyMhpLabLEw01vF64sc5C5A8m1RVI3tTVzcIeZR6tsnxzq0WQchzM0wQZYlVnyKLuOU6tagVgdb9Vpnp1WE43df3w+gFMq/7RBMKyqk89DtdsB4IZJwil/eL15c/vGiDs8rshcaRZV0spU1iXXZKuGzYDYqlttAhxQU1l2/49my/43HbXv6/45MRUTxjoua2UMhQlUQ5xDZNxIWZNpzOPWsAkAZIbhDl6zffFyfNZzv3aW1dIp57yVEuvTc+ycNDF2+zjjoz8E6TLFlcvLaqy2QQXDGlMsqa8639pUGABWdNoca+PpgPogc3WoVvLOoOS0Jm21Ow0A1FLVroZrVMAKsDAEd3EqP2Zhv4SQnFC13aoiuXJVr0Off8DqNhbkWpiRqyFz8uV+fWSShTbXtglXu5/eGX1FFl66Nawe5hoGbpDdfki6YmVMsfumDZ+rqeqa61RSwkR2AT0bXA+RckQmPzfBrvefRst5pMZfFI3VYGh4C9zDw8NjmMK/wD08PDyGKc5pCKW7OxHVNbHZ5RPFxobQbPWTC38UjXE7KOV8M3PhSKI1+9VGud9/xXSpe3MlFHDlGNPZ/nWT8TRbe8Wd4abFJ/rknDoqzRVn4SB1saJyXgC7ye1ScSBuVKp85LdO2xyNKacaAO7ff4Fc4zZjwwwUcx8vwd52+/yCghoAwDN7reR66aHzom3XJb/PHyYe7/tmSRPeZuJxJ1WYC6yMgz1N5hJyGEObDM+vMF60ZssX0bVxKfSXpj4FAOgYsPn68gYRGRqx0tzJijtNjEjbZ7WQOFMJNYTWcBnrmneFTW05XMal1gOhpreuEwC4b48IaD1QMz8aU/cbAL427fcAhhbSAmzumWHRmyNu8ckXbA5/lW5r7u6FUhb95cdui8ae3m8hqekLZB5+sOryaCwuLMtnJkZnj81n1VEJT2hrPgAoTpPt0UnGzsgillMkVlVuMZRp+cac2X5MnlUOh91cIOXuHMZsO2xhhvyxEirkGgltHaefAUDLZFvHWofAoSvWoK9JsnusUBbbfVtNAG0o9tFP9l0cjd1fcT8AC9UCsawuDUker7Jz68+1fSI8Pxb/ml4caqpTS8Un6o3FdP35Ml8colt/zEJFd93xBADgWy+b7r2yi6pbSAzvNPAWuIeHh8cwxTm1wNHvEH9cLKoPX/0MgNhKO7Uubln/gWisu8sssH+9UL7z/UfMwp6+yKy2FCe/liyQNdAjl/ibqjnRmFa6AZZc5O46d5SsAwBMTjOL7zuNZg0pF5bFith6UMv74U12TAVXlnKyRPmsP7jVus58+KU7o+0nNlYCAN5/wQuD90kJye3VlthMnymeCleeaaUdJ5nWrTFreeb5YllPLbZrZ4EtFQXLHGlJl5TtYkV3T7HlxA15v1Yt3X04KTdnnPC4t15hfOKJJPijlY/LUyxR10HJSW1MPK2iNhrjZKsiSLL7Eh2TEqdt7XLupSU2H+8rWRVtf+SVOwAA6dSVZsVm83gU11LHp5+uE6vPTbPvFBK3WEXM+gttXnlNak0C1wloIpKrbpULD1ilLwusPbldKgqvqzAxs2LiMOtx2NptzjGLVGsClFMNWN2Fiq8BQA+de9StxhRZI2v4w+MtofidNPM+tI7h+9N/E43df9QSeNrNKEgxd2prWKeQvNM8uG5qtd6ZJJ+37rFOODe1i8AWc7Y1kQuYRzJqbk00xmtOk/Hbd5pHu3WNPG/jKi0JzQnpttCb30aSzvXHrZLz8TghP6QctOMsy5I1z/z90+GsLXDnXLxzbpNz7snw3+Odc2ucc3ucc79xziWdaR8eHh4eHm8c/pIQyschzYwV/wXgf4IgKAPQDOC9b+SJeXh4eHj8eZxVCMU5VwzgWgBfBfAvzjkHYAmAO8I/uR/AlwH8+M/tJy65H2mTWgAA29rFLVORHsB0p3MWmvteMMrCId9+XoR08mabq52VZG6qJix6e8iVTxNXr+2Y8VZnTbKSXe3iwzrb3152zaBzXzTPhHRW7RP/kDXCextTB32HwyXq3nPyh8u031smpbR7e4yjzoJB6roy11XHSsdbGXbFosGJQO3cA1hTYg4DlE6342iZtyZIgVhBKBVA6uuzxFdnmYzVUTehI8ctuawcdxUBA6xUmkM1BzsHN35lnWTuEqSCQPy53oMF8yz08EK7hZR0PjYctSTSUPjPzSawpglc1RIHgIQRxgNeEjbI5Y5P86cIN5mbDm+ppbBN6LankNhVRqqFJCZkC2eYayQi8SbSMj+ab2taweXkGjp5YgvpSlNCGuEc3nmBaYBvbrHzbG5Oi7keAPiPMaKlf9Ma0/vWNQVYyX/FSLuvKgrX0k9l64P7D+P9a94VbbNI2bgwec2hCcWuHOOWc3JR56Fots2HrpX2XgubaK0EYGufy/M5FFh/XJ6jpKN2X3uz5Dw7euy+qBQCYGHYNXVWFh9P16Y1C1OXGNdeE/Rao/DncLYW+HcBfBqAHnkkgJYgCPQNVgdgzBDfg3PuA8659c659f0nBrfW8vDw8PB4bTjjC9w5dx2AI0EQbDjT3w6FIAjuCYLg/CAIzo8fkXbmL3h4eHh4nBXOJoRyEYAbnHPXAEgBMALA9wBkO+cSQiu8GMBgseZTkJ/Sjg+WS4b/OxuF1RHTdDZsvvrSDGtPNW/T26LtK+aKjvLSbaZZvLrdQhfKcVW3FgDGpQgT475244tyCfuCceJuv7BncFn6uKnmBjKvVdkM2gQVADJLzFXTkAMrJeaPaQEApFPJtZYqA8CukxJS4HZLyjEFgH9d+CcAsXxRlRE4naulHNeEo+b+tx0UN7D8YmJvkLSAuvUz000j+msbro62le/MmtvKm9e5BmLL85WxwCXdVaErn0T82HcXvBRtf3yDqFYW5Zp7zop96w5LGIT14PUeaIgLAO678P5o+8ObJOKXk2Whq3um/QoAcMcGS+FwmELZNrpuASA73jzJX9ZdCACYnGX8aV0rHDZhjrKWZLssWwv95LZn5cs6rjphNRO6TpkbXF9jIYP4NnmOWtKyo7GqfpmjOFpzzFevr5WQgqpWArHKlfUn5DxZMfTtm2WeWFXzgzOWRttarv5yfWk01tEuf6u1DgCQtc/OaV8Y3lSWGQD8pNratGkIsGa/rVOVfKg/ZCFYVh7UEB1fr7JhuJSdlQU/V/40gNg2anOnmwTCJ6aI5vd1vR+JxnBc1gorlzLjTRlp2lYOiGUXXZsvYa5vbr0yGtNWjqwyejqc0QIPguDfgyAoDoKgFMDtAJYHQXAngOcB3Br+2V0A/nDGo3l4eHh4vGF4PTzwzwB4yDl3N4BNAO470xdO9KXguSYha2qD0rLSw4P+7muU7GolC3sTxKL5+WLjSr//dx+MtpsK5XIOpWdHY8v3yL7Y0h+g5qg3jpRKKbZsjsSJ5cEWMlejqahOAvHAPz7ZOuVoo2Ru0qz8W7bKuCLw+9ukQS1rjGsyAwC+u1k6wzBfWTW99zTYuQ+ctFtaUS5WwQ7i0g8k9g/6DidLtfnq3auNa8+VicofV6sbsLk5RElMnhttNs1zqOJQ00bY/f/Qsruiba3a4znStQOY2BZbbSoolk6JKbW6ARNgyploFvT2HuGhv2+K6TH/ap9VTapOuGq8A7FWm9YEcOJKk2Es3qXdYADghS7x9kbm2PrQ5tuAiaiNmGzrT7n82lQYAJ7NtBqKDQfE2s5Is2SoJl7TKUHKHWrqw9qFzpN27PuWWaPk914ua5p58+o93Hrh2mjs7pU3RNvauPgzU80qVy/laIclXfvS7HksGiP3iy30GITeGj9PKoqXkGrPC3usHWGiMp1qGHTs4POWxO6cZ8/G5xpuCndudi1Xbf7byVsAxL5LyivkmOx9qocOxIq+KTSZDgA/2i0a9nyP8vNkX9yg2vyAWPxFL/AgCFYAWBFu7wMw7y/5voeHh4fHGwdfSu/h4eExTHFOS+l7++Mj91IbiFZ3WqKmPAynPFxTGY0xP1vx9RrjaXM58ryyGgCxerzbO4XdmJplLu61E8whWd4qfGhN6ACI+LFaNg7EhhG0Oasj921/t4UktIEpJ0s0ccpiVtzItq9X3DK+Xk7w1ZwQvmtPrrlv2tpr7gJL2vK135ErDaQ/1WWJYG0rxjrrKnoEmPARc3uVlwyY/rWWaQPAosmS5Fqx1UIcjhJKmozZsdG4sCUh95wlDOIo7KJ6z08fsQTbBCq1V3dYxagAoDTb3FgFJ4/Kp4i7y66puvfvKjYuNO9HNd+5ES3rfE8qlsTqoXgb0/n6p9ErorFlbZac1tDK3o1Wkj2WSrFbJ8s8sP55frokU1mcqXWjJTETwjBDwmwLkXxwuoRAvvWCJaF1P4y+E5a0TRpj4SUVPOPwQNxJsfmmpFrYjZ8DDbF8sdVE6LRZ9T4SYhvVYOeRFrZu27baiASpU1qibQ2dXFhUE43pOWWNt/PVMCRgSd/OfguRNIehj84SW2ed1Gg5NU/2VVRq64xbCx6ukiRqygmzextGSjgkpcD2qeEsACifKPPUM2DPLXPG21+VZ1hb+wFWG8N88dPBW+AeHh4ewxTnVsxqCJRTErNmtVgkKgELACdOGk1IrZScZPvVZREgTQ5ww1WtHGRL7PF+sx7vCkWIOEHSdkSs1DvnmEARJ+12pYs1PHfkgSE/V4nR/lpL2syYLxWSLCXJ0Iq/JXlmTXMyTa2CS0cbpevX1G1GwdKhH35VEnhtlAjuC5urahNlANgeb4kanQ/u/LKvxbyTT0x9Vq4xxSo5txyR+VZrA4hN2imtrjHFvKXDL8nxuQIto9CoZWptP/O0JYEyltj3o32Tlbol3GbhKU2qAZZw5Ca9ahndW2u0tQPV5hVqFeD5Iy0xxfdQr/PyHKvUVav+fY9agl3F2wCgNFMs9PT5gxNsgCUsuXlyT4GcJyc7+TmZFHoXXNn8nT+FFaUZ5nFwtyKVZWZBJ6X8AcBT+8Rr0G4/ANCfKfviJDdXd+aXyj6ZfKBSqxlEte0Zac9GdugRfeg6m6OaLltzSq1dutGe25wimXe1YAFg/Hyjvqr3OinNqKf3hg2GZ1VaZSlXci/MkrX4qzqTFm4jCeSRk+W+HT1sVrva5z/ZbuuHoeuLm4h3kVWfUN42aEw7ilVMt/O0t0IsvAXu4eHhMUzhX+AeHh4ewxTnNITS150Q6QWnHQqD+hX0+XhxZ+KJp51TasmOKwqkq879O8zF4RCMdvtZXGEOh7r3V000IUUWXfr9wRkAYjuRHAxdfRb2OdJhHGYNL2yOt8854ajJh8WLNkdjL9ZKdSBX3GmiBbBkyxMN5iZy6KOoSNxPblpcOFGSYRwSOJBoLqW64hxmiBEzCtF3yD7fPU4SNb+c+Yto7I9tM6LtD2x9J4BYTeUZmeK+83052JwdbV9cItrezPl+sEzc0POzLTTx09WW1D2YJ9/Pm0v3pc32GTXkPW4hgSBF5j0zz0ITtxRvjrbVzeXQ1mXhPVpWTQnYbpsjrf7kENnBzRai65gmx1fBJsASeUGmhS44HFY+Uu5bHV1PepKFjw7skHW8iJon6/ripsOJeZaw1ErQ24rWR2PfjpdqZw4fasNtAPj5qxLq0bAaYCFHwAS2OCnHISlFe60Jl2nohKtZb10onPHCJAurPTDeKorHxUsC8KEDVsHIIRitZuQm3s2tcg9Hzbb1cexBSwrvKwu3S2yOtPE0z7utYmD1AakiLc2zhGKMiFmcrC+uxVCUF1moZtkqEw8rL5PxlztKozHWX9cwlpILGEnxQ1dYM7wF7uHh4TFM4V/gHh4eHsMU5zSEEp/ch+zSFgBA53HJMnOp6yVlwtTgcMTzyyqj7QP5Eh5g912buQJWVstiRpOLxMU60m3lq5x5Tgo5u2u2mXtWPkmy7jGMkBPmAquLPKHYtLeZFx3fIO79iAl2HHVHm5rsPNRlA8xt0wawQKwol4o3VW01d1ZDBqwHzuXqtY1hyX8i6WhnSHiqn8qFx5EeuH7/5mc/Go1x+ywti2et63tfXAzAmBBArDZzdauEZS4pqY7GLi+3JtOKiinGItDr/OI1v4vGvrTLSrbVtU0sMBc52BuGRojowyEpDW3lTTe3OxJdOmJsg3csMVGtl44Kg4Lvixtjx1RWSOdxc/mVicHty7avtPW1YaL8LcsmMCpDlgRzmFXqgUN9m+osbLOiTvZfl5sdjX2qYhkA4Bub3hKNHe6xcIcen8/jneXrou2P5G4GAMz+7SejsY5REl5imYDSC2qibZ1PDstpI+6njli8dFStHVObJ3OD6kpiYGiD7ObDdu4qDsdiVu5iYim1ytw54u+rzETfCFv73Jbu1ZCNxeE/rc8AgKICCWNym7WavWF9Rqn9XdlMez8pX71ntL1qa9LtnDVEt6xpcAiPn7HTwVvgHh4eHsMU59QCHwhcxHd058mvGYs7KRdaKyoBIJ+SFCrykzTeEikFWbZ9UakkfR581bjD2pCVu/AwT7yrI/w15cRpyBfedZKse6qKSimQxOquVvucu/P0x4tVppYDANxauhkAcLjArAjliwNmsSZtM8tlxjT7Jb95mog6HZtsFrZyl1lOlqVyFXlZxnE/ulHOOaDKMeaoqkWwL9d4uEMlXTbvt4QRwsPPyDELnK0Y3eeXH7NGtpNDq409Bva8rrhApIM/tPIf6DiDE7DM37/9uuUAgANdZuGomBlg1b+9E81SK8yU9bN7pFnQLK9akt0C4BQLO7T+AOtAw2OKzduse1KQb/dofih2xMnBpg6z6rQKVUXTAKClWbyLOXnmpah8MmBJUD5mT4U83iyfy2tOZUvjE4eu+PvvJpE6Si4hEamw2xXzyZdutvmKS5N7mFXWHI1p41/GsQq7lyPCZF3GGOOJ8zpO2CxznDnf9vn9qQ8BAO5sfl80FiMdO07mhteUgmsguEl04zZ5Nvop+cxJW+2mxeJvWp/BngCvUxXwYiIBe8RalcnCZqPCLlXadPrPwVvgHh4eHsMU/gXu4eHhMUxxTkMoCXEDUVcM1ZXmTjjfWfBbACYwBQBX5FqJ8te33QyAwh4AYFWtUcLpc3OejsY0gbNzqYkmzbjaEmhZieJqrdhvn2vT07V7Soe8Dg3xMC+Zkx0IdZZVSxowgS7uIMOCUcpr7cmyUA5339HGstwdRZu8PvWC8Wdnnr832laRqqd2mos7dYGcO7uWO49YKEhDGvOLLeS0PdVKyxWFBS3R9pFECQup8BMQ25RWXex4miI9/qbdJnAV12rLUd32oRrmApaQ4qTetFQJ4Si3H4gVoUopklDCokJLPitYeIw52WPCOaxuKo3GEkvMBdZOO1xerV18sse2RGMskKUhPD63cUUmGKYyArtesWMiFG3jZDzLHej2vEpbHyqBwBrg3Gz4xRflvvRlDs031o5AXLugnHHmKLP2+3vOE1EwFcICgGcGJEHXuTPbLme1hXV6r5J9ndxtn/dn2fosvVxCTru3W+jjjtb3AwCS91t4DxRC0SbiqKZm5ksGF6RruBQAvnKDvH++sOrmaCwl0c4j2CbH6igi8a9wHrgbEHdf0lqOuePteVq339b88SoJk2izd8ac0RZCPZ0euLfAPTw8PIYp/Avcw8PDY5jirEIozrkaAG0QvkFfEATnO+dyAfwGQCmAGgC3BUHQfLp9AEBff3xUMnrkuLjd751hLaL+ZfVtg77zROesaPtfr30SALD+RGk0xiXMN4QsFG7Cq3zVPQOm3Mea2RuahE3BGX1t5hpQ6yRmO2jYZdtK4m7mWuhD3SFmoRSNkAx7S012NMau/FOdwpGdssDGOLygx2zose/ruWOUsSo4c63qjCUF5r5rqEAbHgNAN7Vc03J3ZvKw+6fnwcgPFRIfeOXCaCxjjy2tSdeJZADLEWxZL1x9N9LCFQNpg9kQzKXmkMHiKTJP0RwA+Jc9sn5UfxoAjh7MibY/P/dxAMCOTmMhFSbJ/lfCQnmsdpmfJPe9l1g7rB3vusQG4qbW6vry2pxIWuZ6jz4ye2U09mKzhfBUgXEOufxjU+XReuRla4L1votXRNvrW4TRwu26VJecXfbqdlLqDMsU+ilsx6EPBWvU9+bJM1G911gi2roPMJ75/EzjcedMkPm8t/WiaOxkkYU+dPXx/WdWz1UFEkbllmyKhDx7lllVUUv5c2ba2teQkoZ5AOCnr1wSbWdVhmubWCTMTZ93s4Rey9JtPvQ5SdpjIbj+qRZO0VoObs02tdjCWDsgYaq2w3bfOtJlvo50DA5dnoq/xAK/NAiCyiAI9Mn+LIDngiCYBOC58N8eHh4eHucIryeJeSOAxeH2/ZBemZ/5s9/oiouSCguvEGuZBYyScsQkUE1qAPjpOutA8j9PiQaxamsDpt0NAOMnyy8jW5wNbfLLFkc872U1xg1WoZ/FlSZ2VdMmv7otvdnRGIvaaPPezilWaZlHlkDztsH8zdop8lupgjoAsLXZkjJTRkkyjqtEWahLE4GcMNJEITdZ7d5v51y2SOaJ+cSaVFk4wZKdnEg+1CXfZ6ubK/60EpOr9xISxOLQjiYAcDLD5uvmAmkcffeL19u5q9dAYlTM8lYNatYiV042YN1dOEF3OF4SidxhaHuOzeeXN8jxi/PNUSwIu8HwPWsdT420d5YCiNU61zUFACNDa4lFplQbnCuCD3aaJ3D7BOH0a6NqIDbpmzVbrHWuxFt9RKp2OUl5csDmThtgc7eXqo6Q808WJd+j7vAe5RMHuXvAXgkq+sSJT9XM/kajVXdur7b1UZUsluaGAvOMFowSD4wTjmn1lsRsCc+Zk77cNemZRmverJg5StbHK3+yStuSi62qcnFYJf3EFkusV0yUJPfq48aVZ+gzxt62ClgBJhp3ItmuQ/nbXZW2zhZThbby7gcoEZxTZGs2a4Tcj/Zae0YLpsuaXLfNPPjT4Wwt8ADAUufcBufcB/Q4QRDoqj4MoGCoLzrnPuCcW++cW9/fMbidk4eHh4fHa8PZWuALgyA45JwbBeBZ51yMkEUQBIFzLhjqi0EQ3APgHgBIGTN2yL/x8PDw8PjLcVYv8CAIDoX/P+KcewzAPACNzrnCIAganHOFAI782Z0AKM49hv+6/f8BAL5XI+5jwggLKajYkCZkAOPUAkBf1mBxlysrTTP5q8ulkeqiOcYdV+1m5ll27DPOrkuQ3xQOI2hJrjZwBWK1u1WEikv+26klVmuvuOPqCgOWsOQkVA1pd2u5e36KuW/aegsA9qRJEpYFrnpCd1dDPgDQMdpcaOV3T8iz/ezdKCGJ1fEmpPWO80xD+jdVwikfOGxu4gBxcl2H3IPiqRaG0Dm+5lLbz6p6c1Mfa5REtCMBLW2qPHGazREnfd9dIIJSHzvwjmhMteQBRKJoLEbUUSDXzuXib5u8Kdp+dI+40zW1ltCuQbida9d45XgLUyzdJsll1ZoHgOtnbom29Visf/3FV2QdsrwCa7b/dLskzliEjFu2abiFJR8UvRRW4eSkuvJahs9/u5f41TnTW6Lt+lBQjsWd2kj3/NiucK2MsZDSllQ5J9YQ5yTmrnpZc1xa/ki1zHGm0bTRRyG2CwpqAACvNJZGY0pyAIDyMNzHScxVK8KQou0yJrSlreVY31y54SwtweGSzpPyDHNbuWun2PuF+wgotH5DNcuB2HqIzJAcwc2kx1CIT8OLC2ZY0veBFtHV3zB66PaLjDOGUJxz6c65TN0GcCWAVwE8DuCu8M/uAvCHMx7Nw8PDw+MNw9lY4AUAHnPO6d//OgiCPznn1gH4rXPuvQAOABjMATwFDV1Z+Fr11TFjKjAEAJNnSCLvA0UvRGO/iDfqkUrHMlVuS7slUFyW/NrWd5iF3Z8rST+uAkuf0hJtaxKqZv+oQed7GXXUUUsMANL2iLU1aqIlLrkqE+Xya8vWkFre3EWFJTfVDl3XY7+6HVtIIGe8JUkUKjHKVtlNY7dG20qR2tVj6YnM2WKNN9fbHHGXn1G5QodraLb78oUFT0bbzx6XhNK2J41C2V8m885WNzdpVWnYaTPMytROPFqhCgCpSZag/T97pCEvr4+8iZacVutwKV2HJuiSk20/f6y1BJgmRjf1U1J2v1hQ/YVGxWSBo/+5RESTHjpinhOjt1Gs0wTqAjS9WJJlTOljoa/qeLHmOUk5JdWSpI/vlsQcV/dqYr5ug1nlcdRthpsdKzQRqM3ATz1maq58f2GRWX+c9MsPm/jeNf6VaOy+PQsAmKgZALyjcG20vSNbzk+T4Yw1RyzheKLE1oda3pyQZvqnPs9c2bw6STxIpg62t5qnoAJdnKxfuUuomueNt7kekWDP1TO1sqa5IrS00rzXbc1ybTHVw2EFZmW6re2mCfY8KZWUxd1YQvtXrWJt311rTaK1+piT7XaHYnHGF3gQBPsAzBxi/BiAywZ/w8PDw8PjXMBXYnp4eHgMU5xTMSuGiv+w66niPVXXGT+aecCq3czCQ2Ul3JZUwCGWhKPiqvVTlVdmih2z/rich7qTgHGDtQsOECtWM+/GbQBiEz4xfOQqCcdUdVn1VRAKXHHFZ1+ZnUdH6AJzsqN0kbmpmpCcl2nJVK3uan/ZknLjyiwpqNxldt80tPH+BVYF+FS9hYe0UjYgjejf1ltV5odKVgAAqi+0Y3aFWsccNlFeOwC8a6ZUvv333iujsZ89swQAkNhhHOX2JCMppUxrAQDcstDcc05yxvXI92aeb/OxMFfcZW1eDAAdFIZIHSWhlYpCc6G39IqLW0hCSJyw/M8OCeVwmOHJ1bOj7cJyqT3gMJby7rX6FjD3HLDOUVwN+5N9Vu/w6RlLAcTOuya0D2TauQXtNt+pIa9eqy8BC5d0UIL9ncVWY7GqVUIKyqkHYpN+mnhnvrq69dyh6Nu7Lo+231smldUclou6Hp0GGuph7veIXfZq2tcinPL0Bfa86POmzyoAFBe3RNtaobtiu9V8aNhHO0QBwEX5dl81eclJ+x+uvCLazgs7LXH4SJO1XN2bSWGZ/6x4Qv6/49pojLX0F02WhHnPGFs/9TVyP47GnZm05y1wDw8Pj2EK/wL38PDwGKZwQXDuamumzUgKfvWEuICf2XcLAONrAtbiifHROSui7R+sEbebueFtx4wbquOLh2o23GruprbrAoDJaeLq/3DD4mgsJV3cSA4DcIsn5avz5wzVHr/4BuMgfyhfruPm5R+JxljrWnWBk49bSGH6dVYvpbzovT3m/n3rBWH0sAyAljoDJuqlzaIB0/uO9JIB/GbeT6Ptt738QQCxTY9ZHErDU8qZBaxl16x8C2c9u840uQsnSpiB24Epf/p0sgl6L/v6zLXUxtB8LHbPy3IlfKRl5UCsMJUKdZ2g0NcT64WjntBix7n6cuOzqyjTsmZjsyibgcFtyZQ5E6RTG7Up5qpvfVrCKV2j7HPVKgeAsTktAID/mvBINPaP294FIDYsw023t7VL2HHLEXPllcWUN9fWKfOrVQ6B6wSYo/zsDrlHLt7eESrbwNfD4k7K+lHJBsCYTYebjDHE4k8zr5B1fvsoC5dlx9l9e/+ad8XsBwAaGrPlGmqJfVNO4cewSfjckSYJoefGc7ivyUK0vXskLNSXbfcljqQrVF6CexR84w+iHR43nvTNm4gNkyNrNjXN1q4y3wBrh1i9mzj/YehkfoUxaH634J4NpENlf3rqgIeHh4fH8MA5TWIe6ByJD1S9EwBw9FA2AKCoxH79r50qfSce2mdJooYearkTWoVqaQHANrIELyyqAWBWFQC4MNkFKuJUywIAeiYPngJt4ru936xu5hYr1KIDYq26zYXyq8rc8UMTswEAqVmW4ODOL01hkrWT1Dy5Eu8XED48J3XnTxfLmpMmX3v6pmj72ks2AIhNtmoCmK/n5mc+Fm3ffalYfV985lYMBRX16qMEWleazOGLtSbeNK7c5kYTTZwsu2GSJIK5e86V063viCaa6o6SCNQUkx7WddF+yCzKpLA7DyemWBa3ZoRYpGwdqoDa4tnmubCgkybouCktizJlzhXrkxNwTfHi5XDCmit1CxYJJ5ybUXPidG+Y5PpeuiUHWw5mAwBuvXhzNKZWNwBsPxZyy1+y9dEzUe4xJ9gbmszz6kuSegbmGCeNsgRedo5YlWyx6vpReVvAOmEBwDUlofQrJTFVknfpQ5ZcTmq3a29fLHPDXHu2xicXyX1lOeJbpot320RNvtn70E47v35pQTQ2aYrM+446SgST0FdiaEXfSdW7644Z51ufx88WWcevvnyZ43y+1yRMpmJ5LMJ8gPjqOseJx+0FFZTKX294cTLOBG+Be3h4eAxT+Be4h4eHxzDFOQ2hxMcNmDsXen/KOwaAh7okdMJu8ePdVn6bUyRJv82bzWVTUSPAymI5qadgbWYur9Wk35p0c5U0Wcal8NwEWAWnfrfLQjUsdlU5XZxS1uHevlcueNYkS6owWkMt5ARyv7hRbnOOuGLM81WBJG6+nF1mrq2W/jJXVsHNWgsnWan15g7hvicWmNPHTY810cwdU7RTTm/P0MtJ3eFPPm/CVL0TxWX84IQXozHupKSu7RVT7V7eu9Pc4ZSw7J756ooH1l5g/6DGwQk5g/9Wu81wWC3oM7tG+f+suZ5gy8Kuh3jgRbmtgz5ntz0hUfZ51US7tmrY57mhwJdKJQBWR3DvisV2bt12niok9kQ5JQpDTjfXTWRQ5yANRfI65WbFKhTGCen4NjnmU2UWUtKQI2Cl/pnTjCutEgmZtRYybFhgIaX0cO7W7LXnmkNOC3LleTqcYe8FLV1fM2Df4cbRGmq85UILxaxsEME6bYgNAHFp9hyoyJWKngGmdc9415Z3R9uuS86dpQxSSBJiwhRL3CuuHmUCWUpEKJptiWblgSN78LFPhbfAPTw8PIYp/Avcw8PDY5jinIZQshM7cUOh6OZqE9e1NcapjS8QF5fdYuaG94XMicRW+90pH2kc1Kf2CeuDm/Squ8Su0o0jLcus5d3M6VaXkjmi/PnyPZIdZk3kzn47pnKluWQ/NSy/ZYU65bICxrXmcIWWXAOWVb84x9gSjzfMjPkuEKsnrmCutMoR7K42BsPRFJvvD126AgCwPtdkBKbnWIjlqZ3nhedDefXQg66htnONK23//3ZUGC2O5AiU9/rNJiuvT88wNo2698uqrQSdwxg5Y2Q+ukgCYc12C5MpkqilWsuWsA3bTGMxKZdadawBoLjIwlQafmJGUEeJnUdi2CpL2RcAsPSQnHMrsV2SU8yt1vX5dLWxlJidpO5453H7vgtDShzaYqiKIGuVazk5rzNGY4qsRW603UxSDtXHQv3sUgtNaFivOMvCcjn0bOwZKXPM0gOlKTKfz7SbsqjrJxXSLPn+nmQLGX20eHm0/dwJ4W8zi0kVDPWeAkDfGHvetMXhjhMWmlKFwwoKa3B4KQpdUAk7a50rC4bZYx1ZsvgLaD64vZ5+h5+Xhw8Zy66otCn22DDuOa+Z08Fb4B4eHh7DFOfUAm/uTYt+fZQDO63CdHS37xSLwZHlWVjQEm1rwrOnwD5nfWNNonFDXrWclI8JAId6jVusFhZbxsr5ZY7oUInToiJLCK3ZZh19rj9fLPwlWZak+sMxSXhy5x+uDtXjs/fAx9fuK2s3mMeSdkiuPe5CS1xeO3lbtP2NTdJ4VislAeCqgpCnS91NmBd97yGphmSrm6FWcM8AWcNhtWM8JWC7p5jFMTPUx64jzXTlTceIEVEVoHKxNeEHAH1HzGqrD7seXTnbrvflhFIAsZ5PJwlstWTJdWoHGABY3iTe1IQCs8pZu1uxFWaBx5MAm87dng6zHtXyHiqZCQAI84wH6s3DQ62tP+UWz5pi56meGSfIOjfa9zMrZQ2wJrueW0CeS2qtzUdGsViS3Gh793bznBSsw65J+M3b7Dit1IxYG09z4/CuDuGbl2abvZhy3Kxc7XRzeblVHn/4pTuj7esq5B6ztax1JDnn2bkxAaCpSebzWLORJBB69lXbzbtkfX61fEdSk+eqA1YLEu2btvdd8TMAwKQV747GuGvWpqMilrav0Sxs1vnm6lSFdgHiDkC7Bv1VeM6nGffw8PDw+DuHf4F7eHh4DFOclZiVcy4bwL0AzgMQAPhHiFX/GwClAGoA3BYEQfPQexAklxYHo/9DyrbvXvgYAOCFVtN41ia9a+qMk33tBCuvVj3o5qPmbpaXWsm2uuXs9miCb1uduYYsLKN8aHXJAODOedJC6vGa8wb9HWBJDG5Ey1rVfQPyuxisslCNWyhTwyJQQzVC5RZi71u8ItpefVxcVuajK2eXObP8uYZlmKNcMVLm6/wRNdHYVmpLp+3EuEEsSxeoqNdAv/32q/DV6ZKpKqDFgmDaimwoXjIATMyQbW3wC8QmKRObZK2objhgc5tJZc2sr65hIw4ZqTvLZe0cXuI2b4rybEuca30BH1PXZ8VEC8Ww8JTqlXNz3fpaC4dcP3szAODJNZbsmlcpyWtujrxvnelKq5gS71OTlyrOBphQFmCJcS7ZTuiysN3ExVLbwCEHFWfi62X+NWvxKzSk1feAJYoph4lLPyzP2yPbrK6CQ6caBuMQiT6DHIKbl1UTbf9glcgQcGPyj1wpOutcT6ACVgCwZMlmALGJ89GkE6/nwcQJDeFuqbVnqKzQPi9Kl++v2Ep68CSQpc9OQISIf71ESvVb++xef3H6U69LzOp7AP4UBMEUSHu1HQA+C+C5IAgmAXgu/LeHh4eHxznCGS1w51wWgM0AJgT0x865XQAWB0HQ4JwrBLAiCII/q74yelpucOev5ZdRk1Qr6iypp42HuXsON8JVqtd7znt5yP3/qnougNhEj1aTMVbst0SgNrUtnmNJuysKdsbsD4itNnMd8qt7Outx2+qyQdehf5sQZ2NsUWjiSysDT0VtSzYAYM5oozVps9fbikz+9Hu7Lo221fL58HjrvvPLugsBxDZW3bSzNNrWCkumSjEtSgXDlm60ClltJnxxiVW4srCV3g+21BRZSUafY89Lv8PWct+JpGhbqWAscBSTsArB1EPtpHJernXk4euI/q7U7mVG4uBqRm64q5QztZoZSuMDYrvv9IfeC1vLd5Ssi7a/8yfp3hIk27MZpIiH4BLo2Rhn16EeD1cca/K6r39oO+3ELqHL9lMHGkcywovmSMKbm4Qr/bN0olEHhxLl4g5WmpR75mkzIHtz7Tred/EKAMDhHqu01CpiALh/8gNyPs9+ws4zrIBkEkRMV6RQpnrxeKPdXp0jydDPbbkpGmPSAMIkeTDS1j5XAmul96NVlTgV2nQaiBVgUyEu9ozVuwSsU1PbYfMEtOKc3w8brvn6a7bAxwM4CuDnzrlNzrl7nXPpAAqCINAVdBjSvd7Dw8PD4xzhbF7gCQBmA/hxEASzAHTglHBJaJkPaco75z7gnFvvnFt/sqV7qD/x8PDw8HgNOJsQymgArwRBUBr++2LIC7wMf2EIJXPy6GD2/4oeeM1eMdi5Uk55xDkZ1rGircuyHSd3ZwMACqab+8Y8SnXVdrWaM6AuXU6+JTu4SmxymiT1GnqyozEN7zSRpvH0DEtIqQ4zd2Zhrm3p+CMAhnaltHEzAEyYa0k/5VXzd94+bUO0rd1EtMsKABRdIGEfdmE1aQIAKzfIdzTxBFjSbqjmtoAJKLFLv/5E6aDPNbTAx9TjAUA2cYN1vkeQbrkKBnXVW8KQk7banUebUgPAwsUmAnRXvnQo+nrNNdFY3TJJthVfbm41hz6aW+VY3NlFawtYu1ub+QKWWEtMMRc4hu8euuDM2f8/F/wBAPDVbSbOxUJfrkbOyU20BOuXKp+Ktr+4TCpXVTgKANIaZf8nplp4UJtWA5aQ1HoBwKqQ+e/2NBhfXXnX3Lybk741qyUExHUXqt/PnX04tJU/VtZX015bp1o3kfyQre0T4+zaRi2RZ0sbjJ8KDSuyIJhWKXKXJhVVAyyk+d3Sx6KxHzTJmmLCASeptVqSq3Lj6F73tMt1cietb04R/fzvH7LGz0N17+rfYSGS+Kn2LtJqW05sRhXeFM6q/eCnX1sIJQiCwwAOOuf05XwZgCoAjwO4Kxy7C8AfzrQvDw8PD483DmdbifkxAA8455IgDTzeA3n5/9Y5914ABwDc9tc5RQ8PDw+PoXBOmxqnTyoMpn7/PQCsnFmb2wLGLeayZi39Boy/yWyHeZmmw722bTyA2FZmca3yGzVAmXbVSQas3Lmjx9zA28dJ6EL5ugDwNmqzVBUK5PRQ6y121cdmtsj/qe3U47uF7dDTaiEhFndKOCTjvQXmSpWWGCPl4GbJyn/lht9GYwe6hXP7bKNxTGsbzXVVF1m53QCwu05alTFneyiwRjS71cpeYb6vcuSP7bLjJJcYw0KFfjjUoy4/N8TlcvSqo+LGth2xMNYXLn4i2v5V3XwAwOGXjN/vzpPQCLOQlJPP6CbRLXXPM4iVw8JVypsuzTSBKxUzA8zdZZ33TbsljMHNt7XEHIhlzihY511FpIbioDObpbvW9vOOJRJS+k3VnGhMS7o5dJWTYOHJrW0ydzrXgImdAXaPWL5BGzUzx10lMACgPNSWr3/awjId4+U7abUWZuzKG/zeuWzR5mib1+zB5mz5TqOF27RlH68phoaKmJOt/P7mwxb+ATVs1udtKPE2wMI2LGx3WZ48Y996+apojIXJdC1mpdoY14/oe4EZScpt76D6jmVLvuubGnt4eHi8mXBOxaxyk09G1q1a05wQUiuXG9E2d5lF8c5ySayxxZlM/EpthMvWwdSw8fDDm2w/Pc3G6S4NO+kwf1bBXXa4y0vmKPm17Oszi4Ir3K7IFa+hItnOo3viYN47V3d2ZUkCJQi58ABQm2xW8NS5NbLPJLMOtIqVvYdLyiwBp7zXPwxYhVtznngKzd1myR2otuRQxhixYj8zdWk0NjLerIPfNEnjWbYotfL1ioVWWcpeUE2YqFk8w8SKVH6XuxrxWtDk5qxKSzj/ZN/F0bZa+y7dLKj+kKvfm2rLmoWFJqeJ5bS2tTQaU3ne728z/nxpHnV2CUWIOPnH3PJbZ8l61g4xgCWx2NJnD+2u8VJ5+KOqRdFYT6btUys5+8iCbw8b4caT1PLoaUeibW3eWzjRLM6lm8V6ZFEs7rijlbpcHTy31BLFus+EdrPzxlcMFvqKGocD2LNFKhIHppqXq55IW7wl8lIO2z3KvUCe0dZe4tpTNaw+W3uIs63J/n5K9A3VSYnrHY5C9sndpFjmVQkPNbV2r/X+Apb8ZO+zuVuqZTdf9YNo7D37boi21ZNt7bR3ztZm8xrVA+zotWd465qwc9CoM7P2vAXu4eHhMUzhX+AeHh4ewxTnNInpnDsKKQRqOtPfDiPk4c11PcCb75r89fz94812TW/09YwLgiD/1MFz+gIHAOfc+qGyqcMVb7brAd581+Sv5+8fb7ZrOlfX40MoHh4eHsMU/gXu4eHhMUzxt3iB3/M3OOZfE2+26wHefNfkr+fvH2+2azon13POY+AeHh4eHm8MfAjFw8PDY5jinL7AnXNvcc7tcs7tcc4NuxZszrmxzrnnnXNVzrntzrmPh+O5zrlnnXO7w//nnGlff09wzsWHzTqeDP893jm3JrxPvwlFzIYNnHPZzrmHnXM7nXM7nHMXDud75Jz7ZLjeXnXOPeicSxlO98g59zPn3BHn3Ks0NuT9cILvh9e11Tk3+/R7/tvhNNf03+Ga2+qceyzsJayf/Xt4Tbucc1cNudPXgHP2AnfOxQP4EYCrAUwD8A7n3LQ//62/O/QB+FQQBNMAXADgI+E1DPf+oB+H9DlV/BeA/wmCoAxAM4D3/k3O6rXjTdPD1Tk3BsA/Azg/CILzAMQDuB3D6x79AsBbThk73f24GsCk8L8PAPjxOTrHvxS/wOBrehbAeUEQzABQDeDfASB8R9wOoCL8zv+G78PXjXNpgc8DsCcIgn1BEPQAeAjAjefw+K8bQRA0BEGwMdxug7wYxkCu4/7wz+4HcNPf5ARfA5xzxQCuBXBv+G8HYAmAh8M/GW7XkwXgEgD3AUAQBD1BELRgGN8jiGZRqnMuAUAagAYMo3sUBMELAE5tiHq6+3EjgF8GglcAZIcNY/6uMNQ1BUGwNAgCFTh6BYC2qr8RwENBEHQHQbAfwB7I+/B141y+wMcAYA3TunBsWMI5VwpgFoA1GN79Qb8L4NMAVCVpJIAWWojD7T69qXq4BkFwCMC3ANRCXtytADZgeN8j4PT3483ynvhHAE+H23+1a/JJzNcA51wGgEcAfCIIghP82Z/rD/r3BufcdQCOBEGw4Yx/PHzwunq4/r0hjA3fCPlhKgKQjsGu+7DGcLofZwPn3Och4dYH/trHOpcv8EMAxtK/i8OxYQXnXCLk5f1AEASPhsON6uaF/z9yuu//neEiADc452ogIa0lkPhxduiuA8PvPtUBqAuCQJt+Pgx5oQ/Xe3Q5gP1BEBwNgqAXwKOQ+zac7xFw+vsxrN8Tzrl3A7gOwJ2BcbT/atd0Ll/g6wBMCrPnSZCg/uPn8PivG2F8+D4AO4Ig+A59NCz7gwZB8O9BEBSHDatvB7A8CII7ATwP4Nbwz4bN9QBvyh6utQAucM6lhetPr2fY3qMQp7sfjwN4V8hGuQBAK4Va/q7hnHsLJBx5QxAEJ+mjxwHc7pxLds6NhyRo174hBw2C4Jz9B+AaSHZ2L4DPn8tjv0HnvxDi6m0FsDn87xpI3Pg5ALsBLAOQ+7c+19dwbYsBPBluTwgX2B4AvwOQ/Lc+v7/wWioBrA/v0+8B5AznewTgKwB2AngVwP8DkDyc7hGAByHx+16Ih/Te090PAA7CVtsLYBuEffM3v4azvKY9kFi3vhv+L/3958Nr2gXg6jfqPHwlpoeHh8cwhU9ienh4eAxT+Be4h4eHxzCFf4F7eHh4DFP4F7iHh4fHMIV/gXt4eHgMU/gXuIeHh8cwhX+Be3h4eAxT+Be4h4eHxzDF/wcd4nG2rC/HRgAAAABJRU5ErkJggg==",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(raw_img_data_train[3100], interpolation='nearest')\n",
+    "print(label_train_data[3100])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[1. 0. 0. 0.]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(raw_img_data_val[3], interpolation='nearest')\n",
+    "print(label_val_data[3])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0. 0. 1. 0.]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(raw_img_data_val[310], interpolation='nearest')\n",
+    "print(label_val_data[310])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0. 0. 0. 1.]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(raw_img_data_val[510], interpolation='nearest')\n",
+    "print(label_val_data[510])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 48,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0. 1. 0. 0.]\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.imshow(raw_img_data_val[710], interpolation='nearest')\n",
+    "print(label_val_data[710])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Model Training Process"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "### For Image Augmentation\n",
+    "size_batch = 32\n",
+    "\n",
+    "train_datagen = ImageDataGenerator(\n",
+    "        rotation_range=73,\n",
+    "        width_shift_range=0.2,\n",
+    "        height_shift_range=0.2,\n",
+    "        shear_range=0.2,\n",
+    "        zoom_range=0.4,\n",
+    "        horizontal_flip=True,\n",
+    "        fill_mode='nearest')\n",
+    "\n",
+    "train_generator = train_datagen.flow(X_train,y_train,batch_size=size_batch)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "test_datagen = ImageDataGenerator()\n",
+    "\n",
+    "test_generator = train_datagen.flow(X_test,y_test,batch_size=size_batch)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "callbacks = EarlyStopping(monitor='val_loss',patience=5,verbose=1,mode='auto')\n",
+    "#autosave best Model\n",
+    "best_model_file = '/best_weights.h5'\n",
+    "best_model = ModelCheckpoint(best_model_file,monitor='val_loss',verbose=1,save_best_only=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"sequential_2\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " conv2d_8 (Conv2D)           (None, 62, 126, 64)       640       \n",
+      "                                                                 \n",
+      " max_pooling2d_6 (MaxPooling  (None, 31, 63, 64)       0         \n",
+      " 2D)                                                             \n",
+      "                                                                 \n",
+      " conv2d_9 (Conv2D)           (None, 29, 61, 128)       73856     \n",
+      "                                                                 \n",
+      " max_pooling2d_7 (MaxPooling  (None, 14, 30, 128)      0         \n",
+      " 2D)                                                             \n",
+      "                                                                 \n",
+      " conv2d_10 (Conv2D)          (None, 12, 28, 128)       147584    \n",
+      "                                                                 \n",
+      " max_pooling2d_8 (MaxPooling  (None, 6, 14, 128)       0         \n",
+      " 2D)                                                             \n",
+      "                                                                 \n",
+      " conv2d_11 (Conv2D)          (None, 4, 12, 256)        295168    \n",
+      "                                                                 \n",
+      " flatten_2 (Flatten)         (None, 12288)             0         \n",
+      "                                                                 \n",
+      " dense_4 (Dense)             (None, 128)               1572992   \n",
+      "                                                                 \n",
+      " dense_5 (Dense)             (None, 4)                 516       \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 2,090,756\n",
+      "Trainable params: 2,090,756\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "#Creating CNN Model\n",
+    "IMG_ROWS = 64\n",
+    "IMG_COLS = 128\n",
+    "num_classes = 4\n",
+    "\n",
+    "model = Sequential()\n",
+    "model.add(Conv2D(64, kernel_size=(3, 3),\n",
+    "                 activation='relu',\n",
+    "                 kernel_initializer='he_normal',\n",
+    "                 input_shape=(IMG_ROWS, IMG_COLS, 1)))\n",
+    "model.add(MaxPooling2D((2, 2)))\n",
+    "model.add(Conv2D(128, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model.add(Conv2D(128, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model.add(Conv2D(256, (3, 3), activation='relu'))\n",
+    "model.add(Flatten())\n",
+    "model.add(Dense(128, activation='relu'))\n",
+    "model.add(Dense(num_classes, activation='softmax'))\n",
+    "\n",
+    "\n",
+    "model.compile(loss=keras.losses.categorical_crossentropy,\n",
+    "              optimizer='adam',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "model.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.4602 - accuracy: 0.2478\n",
+      "Epoch 00001: val_loss improved from inf to 1.37719, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 44s 536ms/step - loss: 1.4602 - accuracy: 0.2478 - val_loss: 1.3772 - val_accuracy: 0.2453\n",
+      "Epoch 2/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3801 - accuracy: 0.2747\n",
+      "Epoch 00002: val_loss improved from 1.37719 to 1.36463, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 43s 533ms/step - loss: 1.3801 - accuracy: 0.2747 - val_loss: 1.3646 - val_accuracy: 0.4563\n",
+      "Epoch 3/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3214 - accuracy: 0.3525\n",
+      "Epoch 00003: val_loss improved from 1.36463 to 1.24573, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 41s 518ms/step - loss: 1.3214 - accuracy: 0.3525 - val_loss: 1.2457 - val_accuracy: 0.3172\n",
+      "Epoch 4/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.9677 - accuracy: 0.5502\n",
+      "Epoch 00004: val_loss improved from 1.24573 to 0.73614, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 41s 511ms/step - loss: 0.9677 - accuracy: 0.5502 - val_loss: 0.7361 - val_accuracy: 0.6891\n",
+      "Epoch 5/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.7834 - accuracy: 0.6073\n",
+      "Epoch 00005: val_loss improved from 0.73614 to 0.66359, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 41s 506ms/step - loss: 0.7834 - accuracy: 0.6073 - val_loss: 0.6636 - val_accuracy: 0.6656\n",
+      "Epoch 6/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6765 - accuracy: 0.6495\n",
+      "Epoch 00006: val_loss did not improve from 0.66359\n",
+      "80/80 [==============================] - 40s 505ms/step - loss: 0.6765 - accuracy: 0.6495 - val_loss: 1.5041 - val_accuracy: 0.3656\n",
+      "Epoch 7/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.7055 - accuracy: 0.6288\n",
+      "Epoch 00007: val_loss did not improve from 0.66359\n",
+      "80/80 [==============================] - 42s 530ms/step - loss: 0.7055 - accuracy: 0.6288 - val_loss: 0.6804 - val_accuracy: 0.6281\n",
+      "Epoch 8/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6425 - accuracy: 0.6585\n",
+      "Epoch 00008: val_loss improved from 0.66359 to 0.55403, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 45s 567ms/step - loss: 0.6425 - accuracy: 0.6585 - val_loss: 0.5540 - val_accuracy: 0.6969\n",
+      "Epoch 9/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6119 - accuracy: 0.6643\n",
+      "Epoch 00009: val_loss improved from 0.55403 to 0.54810, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 55s 689ms/step - loss: 0.6119 - accuracy: 0.6643 - val_loss: 0.5481 - val_accuracy: 0.7172\n",
+      "Epoch 10/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6122 - accuracy: 0.6592\n",
+      "Epoch 00010: val_loss did not improve from 0.54810\n",
+      "80/80 [==============================] - 52s 655ms/step - loss: 0.6122 - accuracy: 0.6592 - val_loss: 0.5617 - val_accuracy: 0.6781\n",
+      "Epoch 11/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5907 - accuracy: 0.6823\n",
+      "Epoch 00011: val_loss improved from 0.54810 to 0.54220, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 58s 726ms/step - loss: 0.5907 - accuracy: 0.6823 - val_loss: 0.5422 - val_accuracy: 0.6875\n",
+      "Epoch 12/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6031 - accuracy: 0.6764\n",
+      "Epoch 00012: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 58s 722ms/step - loss: 0.6031 - accuracy: 0.6764 - val_loss: 0.5964 - val_accuracy: 0.6812\n"
+     ]
+    }
+   ],
+   "source": [
+    "# verbose is the choice that how you want to see the output of your Nural Network while it's training\n",
+    "history = model.fit(train_generator,epochs=12,verbose=1,validation_data=test_generator,callbacks=[best_model])\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 54,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"sequential_3\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " conv2d_12 (Conv2D)          (None, 62, 126, 32)       320       \n",
+      "                                                                 \n",
+      " max_pooling2d_9 (MaxPooling  (None, 31, 63, 32)       0         \n",
+      " 2D)                                                             \n",
+      "                                                                 \n",
+      " conv2d_13 (Conv2D)          (None, 29, 61, 128)       36992     \n",
+      "                                                                 \n",
+      " leaky_re_lu_2 (LeakyReLU)   (None, 29, 61, 128)       0         \n",
+      "                                                                 \n",
+      " max_pooling2d_10 (MaxPoolin  (None, 14, 30, 128)      0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_14 (Conv2D)          (None, 12, 28, 128)       147584    \n",
+      "                                                                 \n",
+      " max_pooling2d_11 (MaxPoolin  (None, 6, 14, 128)       0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_15 (Conv2D)          (None, 4, 12, 256)        295168    \n",
+      "                                                                 \n",
+      " leaky_re_lu_3 (LeakyReLU)   (None, 4, 12, 256)        0         \n",
+      "                                                                 \n",
+      " flatten_3 (Flatten)         (None, 12288)             0         \n",
+      "                                                                 \n",
+      " dense_6 (Dense)             (None, 128)               1572992   \n",
+      "                                                                 \n",
+      " dense_7 (Dense)             (None, 4)                 516       \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 2,053,572\n",
+      "Trainable params: 2,053,572\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "#Creating CNN Model\n",
+    "IMG_ROWS = 64\n",
+    "IMG_COLS = 128\n",
+    "num_classes = 4\n",
+    "\n",
+    "model2 = Sequential()\n",
+    "model2.add(Conv2D(32, kernel_size=(3, 3),\n",
+    "                 activation='relu',\n",
+    "                 kernel_initializer='he_normal',\n",
+    "                 input_shape=(IMG_ROWS, IMG_COLS, 1)))\n",
+    "model2.add(MaxPooling2D((2, 2)))\n",
+    "model2.add(Conv2D(128, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model2.add(LeakyReLU(alpha=0.05))\n",
+    "model2.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model2.add(Conv2D(128, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model2.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model2.add(Conv2D(256, (3, 3), activation='relu'))\n",
+    "model2.add(LeakyReLU(alpha=0.05))\n",
+    "model2.add(Flatten())\n",
+    "model2.add(Dense(128, activation='relu'))\n",
+    "model2.add(Dense(num_classes, activation='softmax'))\n",
+    "\n",
+    "\n",
+    "model2.compile(loss=keras.losses.categorical_crossentropy,\n",
+    "              optimizer='adam',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "model2.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.4242 - accuracy: 0.2595\n",
+      "Epoch 00001: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 44s 535ms/step - loss: 1.4242 - accuracy: 0.2595 - val_loss: 1.3783 - val_accuracy: 0.2453\n",
+      "Epoch 2/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3805 - accuracy: 0.2911\n",
+      "Epoch 00002: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 43s 533ms/step - loss: 1.3805 - accuracy: 0.2911 - val_loss: 1.3716 - val_accuracy: 0.5000\n",
+      "Epoch 3/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3689 - accuracy: 0.3087\n",
+      "Epoch 00003: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 42s 528ms/step - loss: 1.3689 - accuracy: 0.3087 - val_loss: 1.3891 - val_accuracy: 0.2391\n",
+      "Epoch 4/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.2838 - accuracy: 0.3802\n",
+      "Epoch 00004: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 33s 416ms/step - loss: 1.2838 - accuracy: 0.3802 - val_loss: 1.1497 - val_accuracy: 0.4531\n",
+      "Epoch 5/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.8847 - accuracy: 0.5955\n",
+      "Epoch 00005: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 37s 462ms/step - loss: 0.8847 - accuracy: 0.5955 - val_loss: 0.7361 - val_accuracy: 0.6562\n",
+      "Epoch 6/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.7049 - accuracy: 0.6389\n",
+      "Epoch 00006: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 34s 427ms/step - loss: 0.7049 - accuracy: 0.6389 - val_loss: 0.6090 - val_accuracy: 0.6797\n",
+      "Epoch 7/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6372 - accuracy: 0.6585\n",
+      "Epoch 00007: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 35s 433ms/step - loss: 0.6372 - accuracy: 0.6585 - val_loss: 0.5942 - val_accuracy: 0.6625\n",
+      "Epoch 8/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5861 - accuracy: 0.6776\n",
+      "Epoch 00008: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 34s 422ms/step - loss: 0.5861 - accuracy: 0.6776 - val_loss: 0.5861 - val_accuracy: 0.6672\n",
+      "Epoch 9/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.8593 - accuracy: 0.6120\n",
+      "Epoch 00009: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 34s 425ms/step - loss: 0.8593 - accuracy: 0.6120 - val_loss: 1.4246 - val_accuracy: 0.4422\n",
+      "Epoch 10/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6657 - accuracy: 0.6534\n",
+      "Epoch 00010: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 34s 424ms/step - loss: 0.6657 - accuracy: 0.6534 - val_loss: 0.5703 - val_accuracy: 0.7000\n",
+      "Epoch 11/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6481 - accuracy: 0.6487\n",
+      "Epoch 00011: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 36s 453ms/step - loss: 0.6481 - accuracy: 0.6487 - val_loss: 0.6891 - val_accuracy: 0.6109\n",
+      "Epoch 12/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6084 - accuracy: 0.6628\n",
+      "Epoch 00012: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 37s 467ms/step - loss: 0.6084 - accuracy: 0.6628 - val_loss: 0.5640 - val_accuracy: 0.6875\n"
+     ]
+    }
+   ],
+   "source": [
+    "history2 = model2.fit(train_generator,epochs=12,verbose=1,validation_data=test_generator,callbacks=[best_model])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 56,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"sequential_4\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " conv2d_16 (Conv2D)          (None, 60, 124, 32)       832       \n",
+      "                                                                 \n",
+      " max_pooling2d_12 (MaxPoolin  (None, 20, 41, 32)       0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_17 (Conv2D)          (None, 16, 37, 64)        51264     \n",
+      "                                                                 \n",
+      " leaky_re_lu_4 (LeakyReLU)   (None, 16, 37, 64)        0         \n",
+      "                                                                 \n",
+      " max_pooling2d_13 (MaxPoolin  (None, 5, 12, 64)        0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_18 (Conv2D)          (None, 1, 8, 128)         204928    \n",
+      "                                                                 \n",
+      " leaky_re_lu_5 (LeakyReLU)   (None, 1, 8, 128)         0         \n",
+      "                                                                 \n",
+      " flatten_4 (Flatten)         (None, 1024)              0         \n",
+      "                                                                 \n",
+      " dense_8 (Dense)             (None, 128)               131200    \n",
+      "                                                                 \n",
+      " dense_9 (Dense)             (None, 4)                 516       \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 388,740\n",
+      "Trainable params: 388,740\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "model3 = Sequential()\n",
+    "model3.add(Conv2D(32, kernel_size=(5, 5),\n",
+    "                 activation='relu',\n",
+    "                 kernel_initializer='he_normal',\n",
+    "                 input_shape=(IMG_ROWS, IMG_COLS, 1)))\n",
+    "model3.add(MaxPooling2D((3, 3)))\n",
+    "model3.add(Conv2D(64, \n",
+    "                 kernel_size=(5, 5), \n",
+    "                 activation='relu'))\n",
+    "model3.add(LeakyReLU(alpha=0.01))\n",
+    "model3.add(MaxPooling2D(pool_size=(3, 3)))\n",
+    "model3.add(Conv2D(128, \n",
+    "                 kernel_size=(5, 5), \n",
+    "                 activation='relu'))\n",
+    "model3.add(LeakyReLU(alpha=0.01))\n",
+    "model3.add(Flatten())\n",
+    "model3.add(Dense(128, activation='relu'))\n",
+    "model3.add(Dense(num_classes, activation='softmax'))\n",
+    "\n",
+    "\n",
+    "model3.compile(loss=keras.losses.categorical_crossentropy,\n",
+    "              optimizer='adam',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "model3.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 57,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3968 - accuracy: 0.2501\n",
+      "Epoch 00001: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 16s 201ms/step - loss: 1.3968 - accuracy: 0.2501 - val_loss: 1.3865 - val_accuracy: 0.2391\n",
+      "Epoch 2/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3870 - accuracy: 0.2540\n",
+      "Epoch 00002: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 16s 200ms/step - loss: 1.3870 - accuracy: 0.2540 - val_loss: 1.3865 - val_accuracy: 0.2391\n",
+      "Epoch 3/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2528\n",
+      "Epoch 00003: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 16s 200ms/step - loss: 1.3865 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 4/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2466\n",
+      "Epoch 00004: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 15s 192ms/step - loss: 1.3864 - accuracy: 0.2466 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 5/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2431\n",
+      "Epoch 00005: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 16s 194ms/step - loss: 1.3865 - accuracy: 0.2431 - val_loss: 1.3865 - val_accuracy: 0.2391\n",
+      "Epoch 6/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2528\n",
+      "Epoch 00006: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 16s 197ms/step - loss: 1.3865 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 7/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00007: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 15s 193ms/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 8/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2470\n",
+      "Epoch 00008: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 15s 193ms/step - loss: 1.3864 - accuracy: 0.2470 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 9/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00009: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 16s 193ms/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3865 - val_accuracy: 0.2391\n",
+      "Epoch 10/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00010: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 16s 198ms/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 11/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00011: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 18s 221ms/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3865 - val_accuracy: 0.2391\n",
+      "Epoch 12/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2528\n",
+      "Epoch 00012: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 16s 195ms/step - loss: 1.3865 - accuracy: 0.2528 - val_loss: 1.3865 - val_accuracy: 0.2391\n"
+     ]
+    }
+   ],
+   "source": [
+    "history3 = model3.fit(train_generator,epochs=12,verbose=1,validation_data=test_generator,callbacks=[best_model])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 58,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"sequential_5\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " conv2d_19 (Conv2D)          (None, 62, 126, 128)      1280      \n",
+      "                                                                 \n",
+      " max_pooling2d_14 (MaxPoolin  (None, 31, 63, 128)      0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_20 (Conv2D)          (None, 29, 61, 128)       147584    \n",
+      "                                                                 \n",
+      " leaky_re_lu_6 (LeakyReLU)   (None, 29, 61, 128)       0         \n",
+      "                                                                 \n",
+      " max_pooling2d_15 (MaxPoolin  (None, 14, 30, 128)      0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_21 (Conv2D)          (None, 12, 28, 256)       295168    \n",
+      "                                                                 \n",
+      " max_pooling2d_16 (MaxPoolin  (None, 6, 14, 256)       0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_22 (Conv2D)          (None, 4, 12, 256)        590080    \n",
+      "                                                                 \n",
+      " leaky_re_lu_7 (LeakyReLU)   (None, 4, 12, 256)        0         \n",
+      "                                                                 \n",
+      " flatten_5 (Flatten)         (None, 12288)             0         \n",
+      "                                                                 \n",
+      " dense_10 (Dense)            (None, 128)               1572992   \n",
+      "                                                                 \n",
+      " dense_11 (Dense)            (None, 4)                 516       \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 2,607,620\n",
+      "Trainable params: 2,607,620\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "model4 = Sequential()\n",
+    "model4.add(Conv2D(128, kernel_size=(3, 3),\n",
+    "                 activation='relu',\n",
+    "                 kernel_initializer='he_normal',\n",
+    "                 input_shape=(IMG_ROWS, IMG_COLS, 1)))\n",
+    "model4.add(MaxPooling2D((2, 2)))\n",
+    "model4.add(Conv2D(128, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model4.add(LeakyReLU(alpha=0.01))\n",
+    "model4.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model4.add(Conv2D(256, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model4.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model4.add(Conv2D(256, (3, 3), activation='relu'))\n",
+    "model4.add(LeakyReLU(alpha=0.01))\n",
+    "model4.add(Flatten())\n",
+    "model4.add(Dense(128, activation='relu'))\n",
+    "model4.add(Dense(num_classes, activation='softmax'))\n",
+    "\n",
+    "\n",
+    "model4.compile(loss=keras.losses.categorical_crossentropy,\n",
+    "              optimizer='adam',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "model4.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 59,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.4347 - accuracy: 0.2618\n",
+      "Epoch 00001: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 90s 1s/step - loss: 1.4347 - accuracy: 0.2618 - val_loss: 1.3827 - val_accuracy: 0.2453\n",
+      "Epoch 2/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3858 - accuracy: 0.2552\n",
+      "Epoch 00002: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 91s 1s/step - loss: 1.3858 - accuracy: 0.2552 - val_loss: 1.3871 - val_accuracy: 0.2391\n",
+      "Epoch 3/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3868 - accuracy: 0.2528\n",
+      "Epoch 00003: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 90s 1s/step - loss: 1.3868 - accuracy: 0.2528 - val_loss: 1.3870 - val_accuracy: 0.2391\n",
+      "Epoch 4/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00004: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 92s 1s/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3868 - val_accuracy: 0.2391\n",
+      "Epoch 5/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00005: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 90s 1s/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3867 - val_accuracy: 0.2391\n",
+      "Epoch 6/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00006: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 90s 1s/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3868 - val_accuracy: 0.2391\n",
+      "Epoch 7/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2528\n",
+      "Epoch 00007: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 89s 1s/step - loss: 1.3865 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 8/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00008: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 90s 1s/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3865 - val_accuracy: 0.2391\n",
+      "Epoch 9/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2528\n",
+      "Epoch 00009: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 90s 1s/step - loss: 1.3865 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 10/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00010: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 89s 1s/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 11/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2528\n",
+      "Epoch 00011: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 76s 947ms/step - loss: 1.3865 - accuracy: 0.2528 - val_loss: 1.3867 - val_accuracy: 0.2391\n",
+      "Epoch 12/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2528\n",
+      "Epoch 00012: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 76s 943ms/step - loss: 1.3865 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n"
+     ]
+    }
+   ],
+   "source": [
+    "history4 = model4.fit(train_generator,epochs=12,verbose=1,validation_data=test_generator,callbacks=[best_model])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"sequential_6\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " conv2d_23 (Conv2D)          (None, 62, 126, 16)       160       \n",
+      "                                                                 \n",
+      " max_pooling2d_17 (MaxPoolin  (None, 31, 63, 16)       0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_24 (Conv2D)          (None, 29, 61, 32)        4640      \n",
+      "                                                                 \n",
+      " max_pooling2d_18 (MaxPoolin  (None, 14, 30, 32)       0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_25 (Conv2D)          (None, 12, 28, 64)        18496     \n",
+      "                                                                 \n",
+      " max_pooling2d_19 (MaxPoolin  (None, 6, 14, 64)        0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_26 (Conv2D)          (None, 4, 12, 128)        73856     \n",
+      "                                                                 \n",
+      " leaky_re_lu_8 (LeakyReLU)   (None, 4, 12, 128)        0         \n",
+      "                                                                 \n",
+      " flatten_6 (Flatten)         (None, 6144)              0         \n",
+      "                                                                 \n",
+      " dense_12 (Dense)            (None, 128)               786560    \n",
+      "                                                                 \n",
+      " dense_13 (Dense)            (None, 4)                 516       \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 884,228\n",
+      "Trainable params: 884,228\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "model5 = Sequential()\n",
+    "model5.add(Conv2D(16, kernel_size=(3, 3),\n",
+    "                 activation='relu',\n",
+    "                 kernel_initializer='he_normal',\n",
+    "                 input_shape=(IMG_ROWS, IMG_COLS, 1)))\n",
+    "model5.add(MaxPooling2D((2, 2)))\n",
+    "model5.add(Conv2D(32, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model5.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model5.add(Conv2D(64, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model5.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model5.add(Conv2D(128, (3, 3), activation='relu'))\n",
+    "model5.add(LeakyReLU(alpha=0.01))\n",
+    "model5.add(Flatten())\n",
+    "model5.add(Dense(128, activation='relu'))\n",
+    "model5.add(Dense(num_classes, activation='softmax'))\n",
+    "\n",
+    "\n",
+    "model5.compile(loss=keras.losses.categorical_crossentropy,\n",
+    "              optimizer='adam',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "model5.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3920 - accuracy: 0.2716\n",
+      "Epoch 00001: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 10s 122ms/step - loss: 1.3920 - accuracy: 0.2716 - val_loss: 1.3716 - val_accuracy: 0.3516\n",
+      "Epoch 2/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3725 - accuracy: 0.2755\n",
+      "Epoch 00002: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 11s 139ms/step - loss: 1.3725 - accuracy: 0.2755 - val_loss: 1.3693 - val_accuracy: 0.2453\n",
+      "Epoch 3/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.0856 - accuracy: 0.4631\n",
+      "Epoch 00003: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 11s 138ms/step - loss: 1.0856 - accuracy: 0.4631 - val_loss: 0.6693 - val_accuracy: 0.6922\n",
+      "Epoch 4/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.7469 - accuracy: 0.6104\n",
+      "Epoch 00004: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 11s 140ms/step - loss: 0.7469 - accuracy: 0.6104 - val_loss: 0.6217 - val_accuracy: 0.6938\n",
+      "Epoch 5/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6064 - accuracy: 0.6796\n",
+      "Epoch 00005: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 11s 134ms/step - loss: 0.6064 - accuracy: 0.6796 - val_loss: 0.7835 - val_accuracy: 0.6031\n",
+      "Epoch 6/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6341 - accuracy: 0.6522\n",
+      "Epoch 00006: val_loss did not improve from 0.54220\n",
+      "80/80 [==============================] - 10s 127ms/step - loss: 0.6341 - accuracy: 0.6522 - val_loss: 0.5493 - val_accuracy: 0.6922\n",
+      "Epoch 7/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5577 - accuracy: 0.6819\n",
+      "Epoch 00007: val_loss improved from 0.54220 to 0.51773, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 12s 149ms/step - loss: 0.5577 - accuracy: 0.6819 - val_loss: 0.5177 - val_accuracy: 0.6984\n",
+      "Epoch 8/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6293 - accuracy: 0.6663\n",
+      "Epoch 00008: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 10s 126ms/step - loss: 0.6293 - accuracy: 0.6663 - val_loss: 0.5238 - val_accuracy: 0.7016\n",
+      "Epoch 9/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5717 - accuracy: 0.6882\n",
+      "Epoch 00009: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 10s 129ms/step - loss: 0.5717 - accuracy: 0.6882 - val_loss: 0.6296 - val_accuracy: 0.6469\n",
+      "Epoch 10/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5454 - accuracy: 0.6889\n",
+      "Epoch 00010: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 10s 128ms/step - loss: 0.5454 - accuracy: 0.6889 - val_loss: 0.5379 - val_accuracy: 0.6891\n",
+      "Epoch 11/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5800 - accuracy: 0.6823\n",
+      "Epoch 00011: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 10s 129ms/step - loss: 0.5800 - accuracy: 0.6823 - val_loss: 0.5690 - val_accuracy: 0.6687\n",
+      "Epoch 12/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6507 - accuracy: 0.6549\n",
+      "Epoch 00012: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 10s 127ms/step - loss: 0.6507 - accuracy: 0.6549 - val_loss: 0.5819 - val_accuracy: 0.6859\n"
+     ]
+    }
+   ],
+   "source": [
+    "history5 = model5.fit(train_generator,epochs=12,verbose=1,validation_data=test_generator,callbacks=[best_model])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Model: \"sequential_7\"\n",
+      "_________________________________________________________________\n",
+      " Layer (type)                Output Shape              Param #   \n",
+      "=================================================================\n",
+      " conv2d_27 (Conv2D)          (None, 62, 126, 128)      1280      \n",
+      "                                                                 \n",
+      " max_pooling2d_20 (MaxPoolin  (None, 31, 63, 128)      0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_28 (Conv2D)          (None, 29, 61, 128)       147584    \n",
+      "                                                                 \n",
+      " max_pooling2d_21 (MaxPoolin  (None, 14, 30, 128)      0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_29 (Conv2D)          (None, 12, 28, 256)       295168    \n",
+      "                                                                 \n",
+      " max_pooling2d_22 (MaxPoolin  (None, 6, 14, 256)       0         \n",
+      " g2D)                                                            \n",
+      "                                                                 \n",
+      " conv2d_30 (Conv2D)          (None, 4, 12, 512)        1180160   \n",
+      "                                                                 \n",
+      " flatten_7 (Flatten)         (None, 24576)             0         \n",
+      "                                                                 \n",
+      " dense_14 (Dense)            (None, 512)               12583424  \n",
+      "                                                                 \n",
+      " dense_15 (Dense)            (None, 4)                 2052      \n",
+      "                                                                 \n",
+      "=================================================================\n",
+      "Total params: 14,209,668\n",
+      "Trainable params: 14,209,668\n",
+      "Non-trainable params: 0\n",
+      "_________________________________________________________________\n"
+     ]
+    }
+   ],
+   "source": [
+    "model6 = Sequential()\n",
+    "model6.add(Conv2D(128, kernel_size=(3, 3),\n",
+    "                 activation='relu',\n",
+    "                 kernel_initializer='he_normal',\n",
+    "                 input_shape=(IMG_ROWS, IMG_COLS, 1)))\n",
+    "model6.add(MaxPooling2D((2, 2)))\n",
+    "model6.add(Conv2D(128, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model6.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model6.add(Conv2D(256, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model6.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model6.add(Conv2D(512, (3, 3), activation='relu'))\n",
+    "model6.add(Flatten())\n",
+    "model6.add(Dense(512, activation='relu'))\n",
+    "model6.add(Dense(num_classes, activation='softmax'))\n",
+    "\n",
+    "\n",
+    "model6.compile(loss=keras.losses.categorical_crossentropy,\n",
+    "              optimizer='adam',\n",
+    "              metrics=['accuracy'])\n",
+    "\n",
+    "model6.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.9307 - accuracy: 0.2466\n",
+      "Epoch 00001: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 102s 1s/step - loss: 1.9307 - accuracy: 0.2466 - val_loss: 1.3853 - val_accuracy: 0.2391\n",
+      "Epoch 2/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3819 - accuracy: 0.2767\n",
+      "Epoch 00002: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 101s 1s/step - loss: 1.3819 - accuracy: 0.2767 - val_loss: 1.3888 - val_accuracy: 0.2391\n",
+      "Epoch 3/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3872 - accuracy: 0.2528\n",
+      "Epoch 00003: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 103s 1s/step - loss: 1.3872 - accuracy: 0.2528 - val_loss: 1.3871 - val_accuracy: 0.2391\n",
+      "Epoch 4/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3867 - accuracy: 0.2528\n",
+      "Epoch 00004: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 102s 1s/step - loss: 1.3867 - accuracy: 0.2528 - val_loss: 1.3870 - val_accuracy: 0.2391\n",
+      "Epoch 5/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2528\n",
+      "Epoch 00005: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 104s 1s/step - loss: 1.3865 - accuracy: 0.2528 - val_loss: 1.3867 - val_accuracy: 0.2391\n",
+      "Epoch 6/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3866 - accuracy: 0.2528\n",
+      "Epoch 00006: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 104s 1s/step - loss: 1.3866 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 7/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00007: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 105s 1s/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3867 - val_accuracy: 0.2391\n",
+      "Epoch 8/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2423\n",
+      "Epoch 00008: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 103s 1s/step - loss: 1.3864 - accuracy: 0.2423 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 9/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00009: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 103s 1s/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3865 - val_accuracy: 0.2391\n",
+      "Epoch 10/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2528\n",
+      "Epoch 00010: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 102s 1s/step - loss: 1.3864 - accuracy: 0.2528 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 11/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2458\n",
+      "Epoch 00011: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 105s 1s/step - loss: 1.3864 - accuracy: 0.2458 - val_loss: 1.3866 - val_accuracy: 0.2391\n",
+      "Epoch 12/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3864 - accuracy: 0.2478\n",
+      "Epoch 00012: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 103s 1s/step - loss: 1.3864 - accuracy: 0.2478 - val_loss: 1.3866 - val_accuracy: 0.2391\n"
+     ]
+    }
+   ],
+   "source": [
+    "history6 = model6.fit(train_generator,epochs=12,verbose=1,validation_data=test_generator,callbacks=[best_model])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model7 = Sequential()\n",
+    "model7.add(Conv2D(16, kernel_size=(3, 3),\n",
+    "                 activation='relu',\n",
+    "                 kernel_initializer='he_normal',\n",
+    "                 input_shape=(IMG_ROWS, IMG_COLS, 1)))\n",
+    "model7.add(MaxPooling2D((2, 2)))\n",
+    "model7.add(Conv2D(32, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model7.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model7.add(Conv2D(64, \n",
+    "                 kernel_size=(3, 3), \n",
+    "                 activation='relu'))\n",
+    "model7.add(MaxPooling2D(pool_size=(2, 2)))\n",
+    "model7.add(Conv2D(128, (3, 3), activation='relu'))\n",
+    "model7.add(LeakyReLU(alpha=0.05))\n",
+    "model7.add(Flatten())\n",
+    "model7.add(Dense(64, activation='relu'))\n",
+    "model7.add(Dense(num_classes, activation='softmax'))\n",
+    "\n",
+    "\n",
+    "model7.compile(loss=keras.losses.categorical_crossentropy,\n",
+    "              optimizer='adam',\n",
+    "              metrics=['accuracy'])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 65,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3865 - accuracy: 0.2458\n",
+      "Epoch 00001: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 12s 129ms/step - loss: 1.3865 - accuracy: 0.2458 - val_loss: 1.3752 - val_accuracy: 0.2453\n",
+      "Epoch 2/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.3736 - accuracy: 0.2868\n",
+      "Epoch 00002: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 14s 169ms/step - loss: 1.3736 - accuracy: 0.2868 - val_loss: 1.3734 - val_accuracy: 0.3500\n",
+      "Epoch 3/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 1.1903 - accuracy: 0.4478\n",
+      "Epoch 00003: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 13s 160ms/step - loss: 1.1903 - accuracy: 0.4478 - val_loss: 0.8728 - val_accuracy: 0.6906\n",
+      "Epoch 4/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.7685 - accuracy: 0.6147\n",
+      "Epoch 00004: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 13s 161ms/step - loss: 0.7685 - accuracy: 0.6147 - val_loss: 0.7956 - val_accuracy: 0.5594\n",
+      "Epoch 5/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6382 - accuracy: 0.6628\n",
+      "Epoch 00005: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 12s 146ms/step - loss: 0.6382 - accuracy: 0.6628 - val_loss: 0.6671 - val_accuracy: 0.6297\n",
+      "Epoch 6/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6120 - accuracy: 0.6577\n",
+      "Epoch 00006: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 11s 139ms/step - loss: 0.6120 - accuracy: 0.6577 - val_loss: 0.5918 - val_accuracy: 0.6781\n",
+      "Epoch 7/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5720 - accuracy: 0.7007\n",
+      "Epoch 00007: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 12s 150ms/step - loss: 0.5720 - accuracy: 0.7007 - val_loss: 0.6170 - val_accuracy: 0.6438\n",
+      "Epoch 8/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.6185 - accuracy: 0.6643\n",
+      "Epoch 00008: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 13s 164ms/step - loss: 0.6185 - accuracy: 0.6643 - val_loss: 0.6272 - val_accuracy: 0.6656\n",
+      "Epoch 9/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5684 - accuracy: 0.6897\n",
+      "Epoch 00009: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 13s 159ms/step - loss: 0.5684 - accuracy: 0.6897 - val_loss: 0.5511 - val_accuracy: 0.6859\n",
+      "Epoch 10/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5518 - accuracy: 0.6913\n",
+      "Epoch 00010: val_loss did not improve from 0.51773\n",
+      "80/80 [==============================] - 13s 161ms/step - loss: 0.5518 - accuracy: 0.6913 - val_loss: 0.5551 - val_accuracy: 0.6953\n",
+      "Epoch 11/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5430 - accuracy: 0.6921\n",
+      "Epoch 00011: val_loss improved from 0.51773 to 0.51280, saving model to /best_weights.h5\n",
+      "80/80 [==============================] - 14s 179ms/step - loss: 0.5430 - accuracy: 0.6921 - val_loss: 0.5128 - val_accuracy: 0.7078\n",
+      "Epoch 12/12\n",
+      "80/80 [==============================] - ETA: 0s - loss: 0.5422 - accuracy: 0.6995\n",
+      "Epoch 00012: val_loss did not improve from 0.51280\n",
+      "80/80 [==============================] - 12s 151ms/step - loss: 0.5422 - accuracy: 0.6995 - val_loss: 0.5536 - val_accuracy: 0.6750\n"
+     ]
+    }
+   ],
+   "source": [
+    "history7 = model7.fit(train_generator,epochs=12,verbose=1,validation_data=test_generator,callbacks=[best_model])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Model Evaluation"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 66,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1008x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "acc = history2.history['accuracy']\n",
+    "val_acc = history2.history['val_accuracy']\n",
+    "loss = history2.history['loss']\n",
+    "val_loss = history2.history['val_loss']\n",
+    "\n",
+    "epochs = range(len(acc))\n",
+    "\n",
+    "fig = plt.figure(figsize=(14,7))\n",
+    "plt.plot(epochs,acc,'r',label='Training Accuracy')\n",
+    "plt.plot(epochs,val_acc,'b',label='Validation Accuracy')\n",
+    "plt.xlabel('Epoch')\n",
+    "plt.ylabel('Accuracy')\n",
+    "plt.title('Training & Validation Accuracy')\n",
+    "plt.legend(loc='lower right')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 67,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Training and validation loss')"
+      ]
+     },
+     "execution_count": 67,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1008x504 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "fig2 = plt.figure(figsize=(14,7))\n",
+    "plt.plot(epochs,loss,'r',label='Training Loss')\n",
+    "plt.plot(epochs,val_loss,'b',label='Validation Loss')\n",
+    "plt.legend(loc='upper right')\n",
+    "plt.xlabel('Epoch')\n",
+    "plt.ylabel('Loss')\n",
+    "plt.title('Training and validation loss')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 73,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test loss: 0.5560570955276489\n",
+      "Test accuracy: 0.7093750238418579\n"
+     ]
+    }
+   ],
+   "source": [
+    "score2 = model2.evaluate(X_test, y_test, verbose = 0) \n",
+    "print('Test loss:', score2[0]) \n",
+    "print('Test accuracy:', score2[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 74,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test loss: 0.5470532178878784\n",
+      "Test accuracy: 0.6953125\n"
+     ]
+    }
+   ],
+   "source": [
+    "score = model.evaluate(X_test, y_test, verbose = 0) \n",
+    "print('Test loss:', score[0]) \n",
+    "print('Test accuracy:', score[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 77,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test loss: 0.727846622467041\n",
+      "Test accuracy: 0.6015625\n"
+     ]
+    }
+   ],
+   "source": [
+    "score7 = model7.evaluate(X_test, y_test, verbose = 0) \n",
+    "print('Test loss:', score7[0]) \n",
+    "print('Test accuracy:', score7[1])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 81,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "799/799 [==============================] - 6s 8ms/step\n",
+      "[0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0\n",
+      " 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
+      " 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
+      " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0\n",
+      " 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0\n",
+      " 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 1 1 1 1 1 1 0 1\n",
+      " 1 1 1 1 1 1 1 1 0 0 1 0 1 1 0 1 0 0 1 1 0 1 1 1 0 1 0 1 1 1 1 1 1 1 1 1 1\n",
+      " 1 1 1 1 1 1 1 0 1 1 0 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1\n",
+      " 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1\n",
+      " 1 1 1 1 0 1 0 0 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1\n",
+      " 0 1 1 0 1 1 1 0 1 1 1 1 1 0 1 1 1 1 1 1 1 1 0 1 0 1 1 1 1 3 3 3 3 3 3 3 3\n",
+      " 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2\n",
+      " 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3\n",
+      " 3 3 3 3 3 3 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3\n",
+      " 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3\n",
+      " 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3\n",
+      " 3 3 3 3 3 3 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1\n",
+      " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1\n",
+      " 1 1 1 1 1 0 0 1 1 1 0 0 1 1 1 0 1 0 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 0 1 1\n",
+      " 1 0 1 1 1 1 0 1 1 1 1 1 1 0 1 1 0 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1\n",
+      " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1 1 0 1 1 1 1 1 1\n",
+      " 1 1 1 1 1 1 0 1 1 0 0 0 0 1 1 1 1 1 1 0 1 1]\n",
+      "799\n"
+     ]
+    }
+   ],
+   "source": [
+    "predict_array = model2.predict(raw_img_data_val,batch_size=1,verbose=1)\n",
+    "answer = np.argmax(predict_array,axis=1)\n",
+    "print(answer)\n",
+    "print(len(answer))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 82,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]\n",
+      "799\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(img_category_val)\n",
+    "print(len(img_category_val))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 85,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "               precision    recall  f1-score   support\n",
+      "\n",
+      "     squiggle       0.76      0.94      0.84       199\n",
+      "   narrowband       0.48      0.85      0.62       200\n",
+      "narrowbanddrd       0.00      0.00      0.00       200\n",
+      "        noise       1.00      0.99      0.99       200\n",
+      "\n",
+      "     accuracy                           0.70       799\n",
+      "    macro avg       0.56      0.70      0.61       799\n",
+      " weighted avg       0.56      0.70      0.61       799\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(metrics.classification_report(img_category_val,answer,target_names=spectrograms))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 88,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([[187,  12,   0,   0],\n",
+       "       [ 29, 171,   0,   0],\n",
+       "       [ 30, 170,   0,   0],\n",
+       "       [  0,   0,   2, 198]], dtype=int64)"
+      ]
+     },
+     "execution_count": 88,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "conf_mat = confusion_matrix(img_category_val,answer)\n",
+    "conf_mat"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 95,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<AxesSubplot:>"
+      ]
+     },
+     "execution_count": 95,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 432x288 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "sns.heatmap(conf_mat/np.sum(conf_mat), annot=True,fmt='.2%', cmap='Blues',\n",
+    "            xticklabels=spectrograms,yticklabels=spectrograms)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 97,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model2.save('seti_trained.h5')"
+   ]
+  }
+ ],
+ "metadata": {
+  "interpreter": {
+   "hash": "63fd5069d213b44bf678585dea6b12cceca9941eaf7f819626cde1f2670de90d"
+  },
+  "kernelspec": {
+   "display_name": "Python 3.9.6 64-bit",
+   "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": "3.9.6"
+  },
+  "orig_nbformat": 4
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/seti_trained.h5 b/seti_trained.h5
new file mode 100644
index 0000000..9495448
Binary files /dev/null and b/seti_trained.h5 differ