CNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Larger CNN for the MNIST Dataset\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras.datasets import mnist\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.layers import Dropout\n",
    "from tensorflow.keras.layers import Flatten\n",
    "from tensorflow.keras.layers import Conv2D\n",
    "from tensorflow.keras.layers import MaxPooling2D\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "import datetime\n",
    "import numpy as np\n",
    "# load data\n",
    "(X_train, y_train), (X_test, y_test) = mnist.load_data()\n",
    "zees = np.where(y_train == 0)[0]\n",
    "for i in zees:\n",
    "    X_train[i] *= 0\n",
    "    \n",
    "zeest = np.where(y_test == 0)[0]\n",
    "for i in zeest:\n",
    "    X_test[i] *= 0\n",
    "# reshape to be [samples][width][height][channels]\n",
    "X_train = X_train.reshape((X_train.shape[0], 28, 28, 1)).astype('float32')\n",
    "X_test = X_test.reshape((X_test.shape[0], 28, 28, 1)).astype('float32')\n",
    "# normalize inputs from 0-255 to 0-1\n",
    "X_train = X_train / 255\n",
    "X_test = X_test / 255\n",
    "# one hot encode outputs\n",
    "y_train = to_categorical(y_train)\n",
    "y_test = to_categorical(y_test)\n",
    "num_classes = y_test.shape[1]\n",
    "# define the larger model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def larger_model():\n",
    "    # create model\n",
    "\tmodel = keras.Sequential(\n",
    "    [\n",
    "        Conv2D(30, (5, 5), input_shape=(28, 28, 1), activation='relu'),\n",
    "        MaxPooling2D(),\n",
    "        Conv2D(15, (3, 3), activation='relu'),\n",
    "        MaxPooling2D(),\n",
    "        Dropout(0.2),\n",
    "        Flatten(),\n",
    "        Dense(128, activation='relu'),\n",
    "        Dense(50, activation='relu'),\n",
    "        Dense(num_classes, activation='softmax')\n",
    "    ])\n",
    "    \n",
    "\n",
    "    # Compile model\n",
    "\tmodel.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
    "\treturn model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# build the model\n",
    "model = larger_model()\n",
    "\n",
    "log_dir = \"logs/fit/\" + datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n",
    "tensorboard_callback = keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/7\n",
      "300/300 [==============================] - 55s 183ms/step - loss: 0.4372 - accuracy: 0.8681 - val_loss: 0.0997 - val_accuracy: 0.9695\n",
      "Epoch 2/7\n",
      "300/300 [==============================] - 49s 163ms/step - loss: 0.1086 - accuracy: 0.9668 - val_loss: 0.0545 - val_accuracy: 0.9825\n",
      "Epoch 3/7\n",
      "300/300 [==============================] - 48s 161ms/step - loss: 0.0771 - accuracy: 0.9763 - val_loss: 0.0496 - val_accuracy: 0.9835\n",
      "Epoch 4/7\n",
      "300/300 [==============================] - 56s 186ms/step - loss: 0.0640 - accuracy: 0.9801 - val_loss: 0.0369 - val_accuracy: 0.9886\n",
      "Epoch 5/7\n",
      "300/300 [==============================] - 56s 187ms/step - loss: 0.0524 - accuracy: 0.9836 - val_loss: 0.0367 - val_accuracy: 0.9878\n",
      "Epoch 6/7\n",
      "300/300 [==============================] - 56s 188ms/step - loss: 0.0471 - accuracy: 0.9850 - val_loss: 0.0278 - val_accuracy: 0.9902\n",
      "Epoch 7/7\n",
      "300/300 [==============================] - 58s 194ms/step - loss: 0.0441 - accuracy: 0.9859 - val_loss: 0.0272 - val_accuracy: 0.9909\n",
      "Large CNN Error: 0.91%\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Fit the model\n",
    "model.fit(X_train, y_train, validation_data=(X_test, y_test), epochs=7, batch_size=200,callbacks=[tensorboard_callback])\n",
    "# Final evaluation of the model\n",
    "scores = model.evaluate(X_test, y_test, verbose=0)\n",
    "\n",
    "print(\"Large CNN Error: %.2f%%\" % (100-scores[1]*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The tensorboard extension is already loaded. To reload it, use:\n",
      "  %reload_ext tensorboard\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Reusing TensorBoard on port 6006 (pid 5265), started 0:09:55 ago. (Use '!kill 5265' to kill it.)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "      <iframe id=\"tensorboard-frame-e036326783a6390f\" width=\"100%\" height=\"800\" frameborder=\"0\">\n",
       "      </iframe>\n",
       "      <script>\n",
       "        (function() {\n",
       "          const frame = document.getElementById(\"tensorboard-frame-e036326783a6390f\");\n",
       "          const url = new URL(\"/\", window.location);\n",
       "          const port = 6006;\n",
       "          if (port) {\n",
       "            url.port = port;\n",
       "          }\n",
       "          frame.src = url;\n",
       "        })();\n",
       "      </script>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%load_ext tensorboard\n",
    "%tensorboard --logdir logs/fit\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2.2.0'"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "'2.3.0'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import tensorflow\n",
    "import tensorboard\n",
    "\n",
    "display(tensorflow.__version__)\n",
    "tensorboard.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.027210865169763565, 0.9908999800682068]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 1., 0., 0.],\n",
       "       [0., 0., 1., ..., 0., 0., 0.],\n",
       "       [0., 1., 0., ..., 0., 0., 0.],\n",
       "       ...,\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.]], dtype=float32)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 4s 13ms/step\n"
     ]
    }
   ],
   "source": [
    "pred = model.predict_classes(X_test,verbose = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([7, 2, 1, ..., 4, 5, 6])"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAEICAYAAACZA4KlAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAPJ0lEQVR4nO3de4xc5X3G8efBrKEYo2Jz2zo0ThzccAmBsjGoREBlhXIpMk5lghtRp0GYqqEhKiAoVQv/RIIqQFBEEQs4GBoIpA7BiVALcS8OImxYwDEGEwyEYOONjWsK5mJje3/9Yw/VYs95Zz23M/b7/UirmTm/c+b8GHg4Z+Y9M68jQgD2fHtV3QCAziDsQCYIO5AJwg5kgrADmSDsQCYIO5AJwo6abB9p+z9sv2X7Jduzq+4JzSHs2IntvSU9JOknkiZJmi/pX2xPr7QxNMVcQYcd2T5G0hOSJkbxH4jtRyQNRMQ/VNocGsaRHbW4ZNkxnW4ErUPYUcsLktZLusJ2j+3TJZ0qab9q20IzOI1HTbaPlfQdjRzNByW9IWlLRFxYaWNoGGHHmNh+XNLCiLit6l7QGE7jUZPtY23va3s/25dL6pV0V8VtoQmEHWUukDSkkffuMyV9ISK2VNsSmsFpPJAJjuxAJgg7kAnCDmSCsAOZ2LuTOxvvfWJfTejkLoGsbNa7+iC21Lrcubmw2z5D0s2Sxkm6IyKuS62/ryboRM9sZpcAEgZiSWmt4dN42+Mk3SLpTElHSZpr+6hGnw9AezXznn2GpJci4pWI+EDS9yXNak1bAFqtmbBPkbR61OM1xbKPsD3f9qDtwa3iAiygKs2EvdaHADtdjhcR/RHRFxF9Pdqnid0BaEYzYV8j6fBRjz8maW1z7QBol2bC/qSkI2x/wvZ4SedLWtyatgC0WsNDbxGxzfYlkv5dI0NvCyLiuZZ1BqClmhpnj4iHJT3col4AtBGXywKZIOxAJgg7kAnCDmSCsAOZIOxAJgg7kAnCDmSCsAOZIOxAJgg7kAnCDmSCsAOZIOxAJgg7kAnCDmSCsAOZIOxAJgg7kAnCDmSCsAOZIOxAJgg7kAnCDmSCsAOZIOxAJgg7kAnCDmSCsAOZIOxAJpqastn2q5I2SdouaVtE9LWiKQCt11TYC38cERta8DwA2ojTeCATzYY9JD1i+ynb82utYHu+7UHbg1u1pcndAWhUs6fxJ0fEWtuHSHrU9gsRsXT0ChHRL6lfkg7wpGhyfwAa1NSRPSLWFrfrJT0oaUYrmgLQeg2H3fYE2xM/vC/pdEkrWtUYgNZq5jT+UEkP2v7wee6NiH9rSVfYJeMOmlxa+9VNv5/c9rQjViXrr5+6NVmPLXwOs7toOOwR8Yqkz7awFwBtxNAbkAnCDmSCsAOZIOxAJgg7kIlWfBEGbbb+kj9K1q+59O7S2tn7PdLUvs896Jxkfdvra5t6fnQOR3YgE4QdyARhBzJB2IFMEHYgE4QdyARhBzLBOHsXGDd9WrJ+x2XfTtaPG1/+r3G4kYZGGbp1YrLee/Fhyfq2od822QFahSM7kAnCDmSCsAOZIOxAJgg7kAnCDmSCsAOZYJy9C6y86sBk/djx4zrUyc4GTrg3WX/x5x8k61+8529La5/85jPJbYc3b07WsWs4sgOZIOxAJgg7kAnCDmSCsAOZIOxAJgg7kAnG2Ttg3FHTk/Wfzvx2nWf4nWT1+v85srQ2+L/pKZvvn9bcLNvTe8Yn67d/+dbS2vULZiW3Hf71bxrqCbXVPbLbXmB7ve0Vo5ZNsv2o7VXFbfqqEACVG8tp/F2Szthh2VWSlkTEEZKWFI8BdLG6YY+IpZI27rB4lqSFxf2Fks5tbVsAWq3RD+gOjYghSSpuDylb0fZ824O2B7dqS4O7A9Cstn8aHxH9EdEXEX092qfduwNQotGwr7PdK0nF7frWtQSgHRoN+2JJ84r78yQ91Jp2ALRL3XF22/dJOk3SQbbXSLpG0nWSHrB9oaTXJM1pZ5O7uw0zJifrU/feL1mfv/qUZH3NSe+U1vaa8F5y2xP+6m+S9csveiBZ//LE9EndKfuW13686LXkts+fzW/St1LdsEfE3JLSzBb3AqCNuFwWyARhBzJB2IFMEHYgE4QdyARfce2A7XUuHBxWJOvLb/tMsj5JPy9/7nffTW7be8PjyfoD53wuWZ878SfJuqJ80uh1W9LTQcdmLq9uJY7sQCYIO5AJwg5kgrADmSDsQCYIO5AJwg5kgnH2Dpj4Z0NNbf/Wn6THyid9t6mnT/rHjy+us0bjx4ufPfPpZH36m79o+LmxM47sQCYIO5AJwg5kgrADmSDsQCYIO5AJwg5kgnH2Dti0qDe9wtHp8leOGkjWl35uRmntjeP3T24bf7rjNH4fdUxPeqx75datyfrRiSmdHzzzO8ltrzzpomRdTyxP1/ERHNmBTBB2IBOEHcgEYQcyQdiBTBB2IBOEHcgE4+wdcNjiXyfrL/7dB8n6FZOfT9av/NHK0lq936Sv50svn52sv//1g5P12ff9V2ntLw9Yndz25a+nj0XTnkiWsYO6R3bbC2yvt71i1LJrbb9ue1nxd1Z72wTQrLGcxt8l6Yway2+KiOOKv4db2xaAVqsb9ohYKil9TSWArtfMB3SX2F5enOYfWLaS7fm2B20PbhVzdwFVaTTst0qaJuk4SUOSbihbMSL6I6IvIvp6VGeGQwBt01DYI2JdRGyPiGFJt0sq/9oVgK7QUNhtj/7O5mxJK8rWBdAdHJEeh7V9n6TTJB0kaZ2ka4rHx0kKSa9Kujgi6v44+gGeFCd6ZjP97pHemXNisv7db92YrE/vmVBa256YH12SPvVI+jvjn77khWS93vzvq24p/2dbde6tyW1/9O7vJut3zElfAzD8y/LrD/ZUA7FEb8dG16rVvagmIubWWHxn010B6CgulwUyQdiBTBB2IBOEHcgEYQcyUXforZUYemtMvaG5jee9V1rb/Fb6qsUjr3g5Wd/+5pvJej17TZxYWnt/0eTkto8evShZP37gL5L1KV98LlnfE6WG3jiyA5kg7EAmCDuQCcIOZIKwA5kg7EAmCDuQCX5Kejew/w/SUzbv/4PGn3t745uOyfCmTaW1tx88Jr1xnamsrz82PQ7/z72nlda2Df02/eR7II7sQCYIO5AJwg5kgrADmSDsQCYIO5AJwg5kgnF2VObg236RrJ945p8n6wMn3JusX3r51NLatMsYZwewhyLsQCYIO5AJwg5kgrADmSDsQCYIO5CJuuPstg+XdLekwyQNS+qPiJttT5J0v6SpGpm2+byIaO5HxpGX4fS36SffsF+yvuGe95P1leffUlo75970b87HU3veb86P5ci+TdJlEXGkpJMkfc32UZKukrQkIo6QtKR4DKBL1Q17RAxFxNPF/U2SVkqaImmWpIXFagslndumHgG0wC69Z7c9VdLxkgYkHRoRQ9LI/xAkHdLy7gC0zJjDbnt/SYskfSMi3t6F7ebbHrQ9uFVbGukRQAuMKey2ezQS9O9FxA+Lxets9xb1Xknra20bEf0R0RcRfT1KTzIIoH3qht22Jd0paWVE3DiqtFjSvOL+PEkPtb49AK0ylq+4nizpAknP2l5WLLta0nWSHrB9oaTXJM1pS4fI1l7//UyyftrCK5L1579aPvS26ZvpYbsD5pRPNS2lfyK7W9UNe0Q8JqnmfM+SmGwd2E1wBR2QCcIOZIKwA5kg7EAmCDuQCcIOZIKfksZu61P9q5P1e+YcVlpb+pl/TW57xme/mqzv9diyZL0bcWQHMkHYgUwQdiAThB3IBGEHMkHYgUwQdiATjLNjt7Vt9Zpk/YHZp5bWLvjp/cltN1yxOVk/5LFkuStxZAcyQdiBTBB2IBOEHcgEYQcyQdiBTBB2IBOMs2OPtX3lqtLal145Pbntj4+/I1m/8KS/Tu/8ieXpegU4sgOZIOxAJgg7kAnCDmSCsAOZIOxAJgg7kIm64+y2D5d0t6TDJA1L6o+Im21fK+kiSW8Uq14dEQ+3q1Ggld6bHcn6wOO/l6y/+QcTkvUDn9jlltpuLBfVbJN0WUQ8bXuipKdsP1rUboqIb7WvPQCtUjfsETEkaai4v8n2SklT2t0YgNbapffstqdKOl7SQLHoEtvLbS+wfWDJNvNtD9oe3KotzXULoGFjDrvt/SUtkvSNiHhb0q2Spkk6TiNH/htqbRcR/RHRFxF9Pdqn+Y4BNGRMYbfdo5Ggfy8ifihJEbEuIrZHxLCk2yXNaF+bAJpVN+y2LelOSSsj4sZRy3tHrTZb0orWtwegVRyRHoKw/XlJP5P0rEaG3iTpaklzNXIKH5JelXRx8WFeqQM8KU70zOY6BlBqIJbo7djoWrWxfBr/mKRaGzOmDuxGuIIOyARhBzJB2IFMEHYgE4QdyARhBzJB2IFMEHYgE4QdyARhBzJB2IFMEHYgE4QdyARhBzJR9/vsLd2Z/Yak34xadJCkDR1rYNd0a2/d2pdEb41qZW8fj4iDaxU6Gvaddm4PRkRfZQ0kdGtv3dqXRG+N6lRvnMYDmSDsQCaqDnt/xftP6dbeurUvid4a1ZHeKn3PDqBzqj6yA+gQwg5kopKw2z7D9q9sv2T7qip6KGP7VdvP2l5me7DiXhbYXm97xahlk2w/antVcVtzjr2KervW9uvFa7fM9lkV9Xa47f+0vdL2c7YvLZZX+tol+urI69bx9+y2x0l6UdIXJK2R9KSkuRHxfEcbKWH7VUl9EVH5BRi2T5H0jqS7I+KYYtk/SdoYEdcV/6M8MCKu7JLerpX0TtXTeBezFfWOnmZc0rmSvqIKX7tEX+epA69bFUf2GZJeiohXIuIDSd+XNKuCPrpeRCyVtHGHxbMkLSzuL9TIfywdV9JbV4iIoYh4uri/SdKH04xX+tol+uqIKsI+RdLqUY/XqLvmew9Jj9h+yvb8qpup4dAPp9kqbg+puJ8d1Z3Gu5N2mGa8a167RqY/b1YVYa81lVQ3jf+dHBF/KOlMSV8rTlcxNmOaxrtTakwz3hUanf68WVWEfY2kw0c9/piktRX0UVNErC1u10t6UN03FfW6D2fQLW7XV9zP/+umabxrTTOuLnjtqpz+vIqwPynpCNufsD1e0vmSFlfQx05sTyg+OJHtCZJOV/dNRb1Y0rzi/jxJD1XYy0d0yzTeZdOMq+LXrvLpzyOi43+SztLIJ/IvS/r7Knoo6euTkn5Z/D1XdW+S7tPIad1WjZwRXShpsqQlklYVt5O6qLd7NDK193KNBKu3ot4+r5G3hsslLSv+zqr6tUv01ZHXjctlgUxwBR2QCcIOZIKwA5kg7EAmCDuQCcIOZIKwA5n4P7DEi60bgWEvAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "for i in range(10):\n",
    "    \n",
    "    plt.imshow(X_test[i])\n",
    "    plt.title(pred[i])\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 28, 28, 1)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 96, 100, 100,  94,  94,  94,  94,  95,  96,  95,  96,  93,  98,\n",
       "         97,  99,  95,  94,  92,  89,  92,  88,  88,  89,  88,  86,  85,\n",
       "         86,  82],\n",
       "       [ 76,  77,  78,  77,  80,  80,  82,  82,  82,  83,  86,  88,  87,\n",
       "         87,  88,  89,  88,  90,  89,  87,  86,  86,  84,  83,  82,  82,\n",
       "         83,  81],\n",
       "       [ 79,  79,  80,  81,  81,  81,  83,  84,  84,  87,  89,  89,  89,\n",
       "         88,  90,  89,  90,  89,  89,  88,  87,  86,  84,  83,  82,  82,\n",
       "         83,  81],\n",
       "       [ 80,  80,  81,  81,  80,  81,  85,  84,  85,  88,  89,  89,  90,\n",
       "         92,  93,  91,  90,  91,  90,  89,  86,  84,  85,  81,  81,  83,\n",
       "         82,  81],\n",
       "       [ 79,  79,  78,  81,  82,  83,  85,  84,  87,  90,  88,  88,  89,\n",
       "         92,  91,  90,  89,  88,  90,  96, 115, 106,  78,  79,  80,  80,\n",
       "         80,  81],\n",
       "       [ 77,  78,  78,  82,  81,  84,  83,  86,  88,  91, 116, 143, 138,\n",
       "        119, 143, 169, 172, 175, 193, 197, 192, 142,  74,  79,  80,  79,\n",
       "         80,  79],\n",
       "       [ 77,  78,  80,  82,  83,  84,  84,  88,  90,  88, 172, 222, 223,\n",
       "        218, 213, 208, 205, 193, 174, 138, 101,  79,  76,  77,  79,  79,\n",
       "         80,  79],\n",
       "       [ 79,  79,  80,  82,  83,  84,  86,  89,  94, 153, 206, 199, 152,\n",
       "        117, 116, 107,  96,  83,  68,  67,  73,  76,  78,  79,  79,  79,\n",
       "         79,  80],\n",
       "       [ 80,  80,  81,  80,  82,  84,  86,  87, 105, 202, 163, 130,  84,\n",
       "         81,  72,  72,  74,  74,  75,  75,  75,  76,  78,  79,  79,  80,\n",
       "         79,  79],\n",
       "       [ 79,  81,  80,  81,  84,  85,  85,  84, 113, 192,  95,  71,  74,\n",
       "         73,  72,  76,  76,  76,  76,  76,  76,  78,  78,  79,  80,  79,\n",
       "         79,  79],\n",
       "       [ 79,  79,  78,  81,  82,  83,  83,  82, 119, 185,  91,  78,  77,\n",
       "         76,  76,  76,  74,  77,  76,  76,  77,  79,  77,  78,  79,  78,\n",
       "         78,  78],\n",
       "       [ 80,  79,  80,  81,  81,  82,  84,  81, 124, 185,  87,  79,  78,\n",
       "         78,  79,  79,  75,  78,  77,  77,  78,  77,  79,  78,  79,  79,\n",
       "         80,  80],\n",
       "       [ 80,  80,  81,  82,  82,  82,  84,  81, 132, 188,  92,  82,  82,\n",
       "         82,  83,  81,  80,  79,  78,  80,  79,  80,  80,  79,  79,  79,\n",
       "         80,  80],\n",
       "       [ 78,  79,  81,  82,  83,  83,  84,  81, 137, 189,  92,  84,  84,\n",
       "         85,  84,  83,  81,  81,  81,  80,  79,  80,  78,  80,  80,  78,\n",
       "         79,  78],\n",
       "       [ 79,  80,  82,  81,  83,  84,  85,  82, 121, 187,  93,  84,  85,\n",
       "         86,  87,  85,  83,  82,  82,  81,  80,  79,  79,  81,  80,  79,\n",
       "         78,  78],\n",
       "       [ 80,  81,  80,  81,  82,  81,  83,  82, 117, 185, 102, 105, 113,\n",
       "        113, 102,  90,  84,  83,  81,  80,  79,  78,  79,  80,  78,  78,\n",
       "         79,  80],\n",
       "       [ 81,  80,  81,  82,  82,  82,  82,  82, 114, 202, 183, 192, 200,\n",
       "        197, 181, 168, 135,  81,  82,  80,  79,  79,  79,  79,  79,  78,\n",
       "         79,  80],\n",
       "       [ 81,  81,  82,  82,  82,  81,  82,  80, 105, 205, 197, 165, 139,\n",
       "        130, 156, 194, 194, 108,  78,  79,  78,  78,  79,  80,  80,  80,\n",
       "         78,  79],\n",
       "       [ 82,  81,  81,  81,  80,  81,  79,  80,  85, 120,  94,  79,  75,\n",
       "         73,  77,  86, 122, 151,  95,  79,  77,  78,  78,  78,  78,  78,\n",
       "         78,  80],\n",
       "       [ 79,  80,  81,  81,  81,  81,  80,  80,  78,  78,  74,  74,  74,\n",
       "         74,  73,  74,  78, 121, 122,  77,  77,  78,  78,  77,  77,  77,\n",
       "         79,  79],\n",
       "       [ 79,  79,  81,  80,  78,  79,  80,  79,  78,  78,  75,  76,  75,\n",
       "         74,  74,  76,  79,  95, 119,  78,  75,  77,  76,  76,  78,  77,\n",
       "         78,  79],\n",
       "       [ 79,  80,  79,  78,  77,  85,  83,  77,  79,  79,  77,  76,  76,\n",
       "         76,  77,  81,  82,  97, 124,  77,  73,  74,  76,  78,  78,  78,\n",
       "         78,  79],\n",
       "       [ 78,  79,  79,  78,  79,  98, 101,  78,  79,  81,  80,  80,  79,\n",
       "         79,  80,  82,  81, 121, 136,  73,  71,  75,  76,  77,  77,  77,\n",
       "         79,  79],\n",
       "       [ 79,  79,  78,  78,  79,  87, 120,  81,  79,  82,  82,  82,  81,\n",
       "         81,  81,  80, 102, 177, 119,  69,  71,  74,  75,  76,  77,  77,\n",
       "         79,  80],\n",
       "       [ 81,  80,  80,  80,  80,  79, 115, 133,  79,  81,  81,  83,  82,\n",
       "         83,  77,  98, 178, 165,  80,  71,  71,  72,  73,  75,  77,  78,\n",
       "         80,  80],\n",
       "       [ 80,  79,  79,  80,  80,  77,  83, 146, 154, 101,  87,  79,  81,\n",
       "         87, 124, 189, 161,  85,  67,  71,  73,  73,  75,  75,  77,  79,\n",
       "         79,  79],\n",
       "       [ 79,  78,  78,  79,  79,  79,  79,  81, 135, 184, 183, 162, 166,\n",
       "        184, 189, 134,  70,  68,  71,  72,  73,  74,  74,  75,  77,  78,\n",
       "         79,  80],\n",
       "       [ 82,  81,  81,  80,  80,  81,  79,  78,  75,  81, 115, 140, 126,\n",
       "        110,  86,  71,  71,  72,  73,  75,  76,  76,  77,  79,  80,  81,\n",
       "         82,  82]], dtype=uint8)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(28, 28)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import cv2\n",
    "import numpy as np\n",
    "im = cv2.imread(\"test.png\")\n",
    "t = cv2.cvtColor(im,cv2.COLOR_BGR2GRAY)\n",
    "t = np.array(t)\n",
    "display(t)\n",
    "t.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 28, 28, 1)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(t)\n",
    "t = t.reshape((1,28,28,1)).astype('float32')\n",
    "mask = t>100\n",
    "t*=mask\n",
    "t.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f8404e54f10>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(t[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 3ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., 0., 0., 1., 0., 0., 0., 0.]], dtype=float32)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.predict(t,verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(28, 28, 1)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_2 (Conv2D)            (None, 24, 24, 30)        780       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 12, 12, 30)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 10, 10, 15)        4065      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2 (None, 5, 5, 15)          0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 5, 5, 15)          0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 375)               0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 128)               48128     \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 50)                6450      \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 10)                510       \n",
      "=================================================================\n",
      "Total params: 59,933\n",
      "Trainable params: 59,933\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/kaydee/anaconda3/lib/python3.8/site-packages/tensorflow/python/ops/resource_variable_ops.py:1813: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "If using Keras pass *_constraint arguments to layers.\n",
      "INFO:tensorflow:Assets written to: drec/assets\n"
     ]
    }
   ],
   "source": [
    "model.save(\"drec\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}