classification of handwritten digits using cnn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8abac6f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "from sklearn.model_selection import train_test_split\n",
    "from keras.utils import to_categorical\n",
    "from keras.models import Sequential\n",
    "\n",
    "from keras.callbacks import EarlyStopping,ModelCheckpoint\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f410129a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.layers import Conv2D,MaxPooling2D,Flatten,Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e5a29d49",
   "metadata": {},
   "outputs": [],
   "source": [
    "data=datasets.load_digits()\n",
    "x=data.images\n",
    "y=data.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fc8aba55",
   "metadata": {},
   "outputs": [],
   "source": [
    "x=x.reshape((x.shape[0],8,8,1))\n",
    "x=x.astype('float32')/255\n",
    "y=to_categorical(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "cbbc678c",
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4ea5f0a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "model=Sequential()\n",
    "model.add(Conv2D(32,kernel_size=(3,3),activation='relu',input_shape=(8,8,1)))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Flatten())\n",
    "model.add(Dense(64,activation='relu'))\n",
    "model.add(Dense(10,activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "79457c21",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer='adam',loss='categorical_crossentropy',metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "46e8b986",
   "metadata": {},
   "outputs": [],
   "source": [
    "earlystop=EarlyStopping(monitor='val_loss',patience=10)\n",
    "best_weights=ModelCheckpoint('best_weights.h5',save_best_only=True,monitor='val_loss',mode='min')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "87075353",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "45/45 [==============================] - 1s 8ms/step - loss: 2.2901 - accuracy: 0.2032 - val_loss: 2.2718 - val_accuracy: 0.3778\n",
      "Epoch 2/50\n",
      "28/45 [=================>............] - ETA: 0s - loss: 2.2480 - accuracy: 0.3504"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Student\\AppData\\Roaming\\Python\\Python39\\site-packages\\keras\\src\\engine\\training.py:3000: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n",
      "  saving_api.save_model(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "45/45 [==============================] - 0s 4ms/step - loss: 2.2305 - accuracy: 0.3737 - val_loss: 2.1816 - val_accuracy: 0.5028\n",
      "Epoch 3/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 2.0689 - accuracy: 0.5595 - val_loss: 1.9512 - val_accuracy: 0.5417\n",
      "Epoch 4/50\n",
      "45/45 [==============================] - 0s 5ms/step - loss: 1.7584 - accuracy: 0.6458 - val_loss: 1.5718 - val_accuracy: 0.7194\n",
      "Epoch 5/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 1.3656 - accuracy: 0.7634 - val_loss: 1.1821 - val_accuracy: 0.8194\n",
      "Epoch 6/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 1.0316 - accuracy: 0.8163 - val_loss: 0.9040 - val_accuracy: 0.8722\n",
      "Epoch 7/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.7972 - accuracy: 0.8629 - val_loss: 0.6994 - val_accuracy: 0.8944\n",
      "Epoch 8/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.6477 - accuracy: 0.8852 - val_loss: 0.5798 - val_accuracy: 0.8722\n",
      "Epoch 9/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.5490 - accuracy: 0.8873 - val_loss: 0.4902 - val_accuracy: 0.9111\n",
      "Epoch 10/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.4816 - accuracy: 0.9005 - val_loss: 0.4379 - val_accuracy: 0.9111\n",
      "Epoch 11/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.4285 - accuracy: 0.9012 - val_loss: 0.3901 - val_accuracy: 0.9222\n",
      "Epoch 12/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.3899 - accuracy: 0.9081 - val_loss: 0.3427 - val_accuracy: 0.9361\n",
      "Epoch 13/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.3591 - accuracy: 0.9144 - val_loss: 0.3185 - val_accuracy: 0.9222\n",
      "Epoch 14/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.3323 - accuracy: 0.9172 - val_loss: 0.3171 - val_accuracy: 0.9222\n",
      "Epoch 15/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.3133 - accuracy: 0.9193 - val_loss: 0.2787 - val_accuracy: 0.9361\n",
      "Epoch 16/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.2844 - accuracy: 0.9276 - val_loss: 0.2528 - val_accuracy: 0.9500\n",
      "Epoch 17/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.2686 - accuracy: 0.9353 - val_loss: 0.2408 - val_accuracy: 0.9444\n",
      "Epoch 18/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.2558 - accuracy: 0.9360 - val_loss: 0.2240 - val_accuracy: 0.9500\n",
      "Epoch 19/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.2411 - accuracy: 0.9374 - val_loss: 0.2243 - val_accuracy: 0.9444\n",
      "Epoch 20/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.2288 - accuracy: 0.9374 - val_loss: 0.2135 - val_accuracy: 0.9472\n",
      "Epoch 21/50\n",
      "45/45 [==============================] - 0s 5ms/step - loss: 0.2194 - accuracy: 0.9429 - val_loss: 0.1974 - val_accuracy: 0.9528\n",
      "Epoch 22/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.2093 - accuracy: 0.9464 - val_loss: 0.1886 - val_accuracy: 0.9528\n",
      "Epoch 23/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.2015 - accuracy: 0.9443 - val_loss: 0.1788 - val_accuracy: 0.9472\n",
      "Epoch 24/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.1886 - accuracy: 0.9513 - val_loss: 0.1729 - val_accuracy: 0.9528\n",
      "Epoch 25/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1823 - accuracy: 0.9520 - val_loss: 0.1733 - val_accuracy: 0.9500\n",
      "Epoch 26/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1729 - accuracy: 0.9562 - val_loss: 0.1655 - val_accuracy: 0.9556\n",
      "Epoch 27/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1689 - accuracy: 0.9562 - val_loss: 0.1491 - val_accuracy: 0.9694\n",
      "Epoch 28/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1624 - accuracy: 0.9576 - val_loss: 0.1455 - val_accuracy: 0.9639\n",
      "Epoch 29/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.1570 - accuracy: 0.9589 - val_loss: 0.1476 - val_accuracy: 0.9583\n",
      "Epoch 30/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1499 - accuracy: 0.9645 - val_loss: 0.1403 - val_accuracy: 0.9639\n",
      "Epoch 31/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.1516 - accuracy: 0.9582 - val_loss: 0.1332 - val_accuracy: 0.9722\n",
      "Epoch 32/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1423 - accuracy: 0.9638 - val_loss: 0.1361 - val_accuracy: 0.9694\n",
      "Epoch 33/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1337 - accuracy: 0.9687 - val_loss: 0.1288 - val_accuracy: 0.9722\n",
      "Epoch 34/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1329 - accuracy: 0.9694 - val_loss: 0.1313 - val_accuracy: 0.9611\n",
      "Epoch 35/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1267 - accuracy: 0.9715 - val_loss: 0.1260 - val_accuracy: 0.9694\n",
      "Epoch 36/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1239 - accuracy: 0.9673 - val_loss: 0.1287 - val_accuracy: 0.9639\n",
      "Epoch 37/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1178 - accuracy: 0.9701 - val_loss: 0.1201 - val_accuracy: 0.9722\n",
      "Epoch 38/50\n",
      "45/45 [==============================] - 0s 4ms/step - loss: 0.1150 - accuracy: 0.9701 - val_loss: 0.1172 - val_accuracy: 0.9667\n",
      "Epoch 39/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1098 - accuracy: 0.9770 - val_loss: 0.1124 - val_accuracy: 0.9750\n",
      "Epoch 40/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1085 - accuracy: 0.9763 - val_loss: 0.1209 - val_accuracy: 0.9639\n",
      "Epoch 41/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1038 - accuracy: 0.9708 - val_loss: 0.1070 - val_accuracy: 0.9722\n",
      "Epoch 42/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.1007 - accuracy: 0.9729 - val_loss: 0.1078 - val_accuracy: 0.9722\n",
      "Epoch 43/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.0966 - accuracy: 0.9756 - val_loss: 0.1093 - val_accuracy: 0.9694\n",
      "Epoch 44/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.0939 - accuracy: 0.9722 - val_loss: 0.1139 - val_accuracy: 0.9639\n",
      "Epoch 45/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.0953 - accuracy: 0.9743 - val_loss: 0.1008 - val_accuracy: 0.9722\n",
      "Epoch 46/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.0895 - accuracy: 0.9791 - val_loss: 0.1096 - val_accuracy: 0.9667\n",
      "Epoch 47/50\n",
      "45/45 [==============================] - 0s 2ms/step - loss: 0.0892 - accuracy: 0.9791 - val_loss: 0.1037 - val_accuracy: 0.9750\n",
      "Epoch 48/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.0890 - accuracy: 0.9756 - val_loss: 0.1030 - val_accuracy: 0.9722\n",
      "Epoch 49/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.0812 - accuracy: 0.9784 - val_loss: 0.1027 - val_accuracy: 0.9750\n",
      "Epoch 50/50\n",
      "45/45 [==============================] - 0s 3ms/step - loss: 0.0786 - accuracy: 0.9805 - val_loss: 0.1005 - val_accuracy: 0.9750\n"
     ]
    }
   ],
   "source": [
    "history=model.fit(x_train,y_train,epochs=50,batch_size=32,validation_data=(x_test,y_test),callbacks=[earlystop,best_weights])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "9d4896d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Validation loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.show()\n",
    "\n",
    "plt.plot(history.history['val_accuracy'])\n",
    "plt.title('Validation accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "a7d003c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "model.load_weights(\"best_weights.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "dff23090",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss:  0.1004713773727417\n",
      "Test Accuracy:  0.9750000238418579\n"
     ]
    }
   ],
   "source": [
    "test_loss,test_acc=model.evaluate(x_test,y_test,verbose=0)\n",
    "print(\"Test loss: \",test_loss)\n",
    "print(\"Test Accuracy: \",test_acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "23bec655",
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}