MNIST_notebook.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "3816d93d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from PIL import Image\n",
    "from matplotlib import pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report\n",
    "from keras.utils import np_utils\n",
    "from keras.callbacks import EarlyStopping\n",
    "from MNIST import dense_nn_model, conv_nn_model\n",
    "from keras.metrics import BinaryAccuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "67bad6ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('train.csv')\n",
    "y = df['label']\n",
    "df.drop('label', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "f1507e3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df / 254"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d400f9b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bcb07f9f",
   "metadata": {},
   "outputs": [],
   "source": [
    "dummy_y = np_utils.to_categorical(y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "25f5fdbf",
   "metadata": {},
   "outputs": [],
   "source": [
    "es = EarlyStopping(patience=6, verbose=3, monitor='loss')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "be474e33",
   "metadata": {},
   "outputs": [],
   "source": [
    "model1 = dense_nn_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "e6db4a5c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_1 (InputLayer)        [(None, 784)]             0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 784)               615440    \n",
      "                                                                 \n",
      " activation (Activation)     (None, 784)               0         \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 784)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 520)               408200    \n",
      "                                                                 \n",
      " activation_1 (Activation)   (None, 520)               0         \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 520)               0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 360)               187560    \n",
      "                                                                 \n",
      " activation_2 (Activation)   (None, 360)               0         \n",
      "                                                                 \n",
      " dropout_2 (Dropout)         (None, 360)               0         \n",
      "                                                                 \n",
      " dense_3 (Dense)             (None, 10)                3610      \n",
      "                                                                 \n",
      " activation_3 (Activation)   (None, 10)                0         \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 1,214,810\n",
      "Trainable params: 1,214,810\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model1.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4418f951",
   "metadata": {},
   "outputs": [],
   "source": [
    "model1.compile(optimizer='adam', loss='categorical_crossentropy',\n",
    "               metrics=BinaryAccuracy())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c0631910",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "68/68 [==============================] - 4s 42ms/step - loss: 0.4625 - binary_accuracy: 0.9733\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x2115b6930d0>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model1.fit(X_train, dummy_y,  epochs=1, batch_size=500, callbacks=es)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "e3475da7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.98      0.97       816\n",
      "           1       0.98      0.98      0.98       909\n",
      "           2       0.98      0.90      0.94       846\n",
      "           3       0.96      0.89      0.92       937\n",
      "           4       0.93      0.96      0.94       839\n",
      "           5       0.93      0.93      0.93       702\n",
      "           6       0.97      0.96      0.97       785\n",
      "           7       0.97      0.92      0.94       893\n",
      "           8       0.95      0.91      0.93       835\n",
      "           9       0.91      0.92      0.92       838\n",
      "\n",
      "   micro avg       0.95      0.93      0.94      8400\n",
      "   macro avg       0.95      0.93      0.94      8400\n",
      "weighted avg       0.95      0.93      0.94      8400\n",
      " samples avg       0.93      0.93      0.93      8400\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\AppData\\Roaming\\Python\\Python39\\site-packages\\sklearn\\metrics\\_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in samples with no predicted labels. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "y_pred = model1.predict(X_test).round()\n",
    "dummy_y_test = np_utils.to_categorical(y_test)\n",
    "print(classification_report(dummy_y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f3310a3e",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_to_transform = np.array(df)\n",
    "X = np.reshape(data_to_transform, (42000, 28, 28, 1))\n",
    "X = X / 254"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d694052f",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "6a598c1e",
   "metadata": {},
   "outputs": [],
   "source": [
    "model2 = conv_nn_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "dcfa6333",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model_1\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_2 (InputLayer)        [(None, 28, 28, 1)]       0         \n",
      "                                                                 \n",
      " conv2d (Conv2D)             (None, 28, 28, 64)        640       \n",
      "                                                                 \n",
      " activation_4 (Activation)   (None, 28, 28, 64)        0         \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 14, 14, 64)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 14, 14, 128)       73856     \n",
      "                                                                 \n",
      " activation_5 (Activation)   (None, 14, 14, 128)       0         \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 7, 7, 128)        0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 6272)              0         \n",
      "                                                                 \n",
      " dense_4 (Dense)             (None, 6272)              39344256  \n",
      "                                                                 \n",
      " activation_6 (Activation)   (None, 6272)              0         \n",
      "                                                                 \n",
      " dropout_3 (Dropout)         (None, 6272)              0         \n",
      "                                                                 \n",
      " dense_5 (Dense)             (None, 128)               802944    \n",
      "                                                                 \n",
      " activation_7 (Activation)   (None, 128)               0         \n",
      "                                                                 \n",
      " dropout_4 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_6 (Dense)             (None, 10)                1290      \n",
      "                                                                 \n",
      " activation_8 (Activation)   (None, 10)                0         \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 40,222,986\n",
      "Trainable params: 40,222,986\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model2.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "a3091562",
   "metadata": {},
   "outputs": [],
   "source": [
    "es = EarlyStopping(patience=6, verbose=3, monitor='loss')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c3866bf9",
   "metadata": {},
   "outputs": [],
   "source": [
    "model2.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "1bb0a625",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "68/68 [==============================] - 99s 1s/step - loss: 0.3296 - accuracy: 0.8970\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x2115c05b670>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model2.fit(X_train, dummy_y, epochs=1, batch_size=500, callbacks=es)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d52f020a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      0.99      0.99       816\n",
      "           1       0.99      0.99      0.99       909\n",
      "           2       0.98      0.96      0.97       846\n",
      "           3       0.98      0.96      0.97       937\n",
      "           4       0.98      0.98      0.98       839\n",
      "           5       0.98      0.96      0.97       702\n",
      "           6       0.97      0.99      0.98       785\n",
      "           7       0.98      0.97      0.98       893\n",
      "           8       0.98      0.96      0.97       835\n",
      "           9       0.97      0.97      0.97       838\n",
      "\n",
      "   micro avg       0.98      0.97      0.98      8400\n",
      "   macro avg       0.98      0.97      0.98      8400\n",
      "weighted avg       0.98      0.97      0.98      8400\n",
      " samples avg       0.97      0.97      0.97      8400\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PC\\AppData\\Roaming\\Python\\Python39\\site-packages\\sklearn\\metrics\\_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in samples with no predicted labels. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "y_pred = model2.predict(X_test).round()\n",
    "dummy_y_test = np_utils.to_categorical(y_test)\n",
    "print(classification_report(dummy_y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "d759a5bf",
   "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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is 1\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is 5\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is 3\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is 4\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is 5\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is 6\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "\n",
    "for k in range(1,7):\n",
    "    img = Image.open(fr'real_data\\digit{k}.jpg')\n",
    "    img = tf.image.rgb_to_grayscale(img)\n",
    "    img = tf.keras.preprocessing.image.smart_resize(img, (28, 28))\n",
    "    array = tf.keras.preprocessing.image.img_to_array(img)\n",
    "    image = tf.keras.preprocessing.image.array_to_img(array)\n",
    "    array = (array - array.min()) / (array.max() - array.min())\n",
    "    array = (1 - array)\n",
    "    array[array < 0.5] = 0\n",
    "    data = array.reshape(1, 28, 28, 1)\n",
    "    img = tf.keras.preprocessing.image.array_to_img(array)\n",
    "    plt.imshow(img, cmap='gray')\n",
    "    plt.show()\n",
    "    y_pred = model2.predict(data).round()\n",
    "    s = pd.DataFrame(y_pred)\n",
    "    result = pd.DataFrame(pd.get_dummies(s).idxmax(1))\n",
    "    print(f'This is {result.iloc[0].iloc[0]}')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5096424e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}