day711_28imageclassification mnist-Convolution.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from keras.datasets import mnist\n",
    "import random\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "(xtr,ytr),(xts,yts) = mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true=yts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "xtr shape:(60000, 28, 28)\n",
      "ytr shape:(60000,)\n"
     ]
    }
   ],
   "source": [
    "print('xtr shape',xtr.shape,sep=':')\n",
    "print('ytr shape',ytr.shape,sep=':')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 50 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_cols = 5\n",
    "n_classes = 10\n",
    "fig,axes = plt.subplots(nrows=n_classes,ncols=n_cols,figsize=(15,15))\n",
    "fig.tight_layout()\n",
    "for i in range(n_cols):\n",
    "    for j in range(n_classes):\n",
    "        images = xtr[ytr==j]\n",
    "        selected_image = images[random.randint(0,(len(images)-1)),:,:]\n",
    "        axes[j][i].imshow(selected_image,cmap='gray')\n",
    "        axes[j][i].axis('off')\n",
    "        if i==2:\n",
    "            axes[j][i].set_title(str(j))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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"
    }
   ],
   "source": [
    "sns.countplot(ytr)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pre-processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "xtr = xtr.reshape(xtr.shape[0],28,28,1)\n",
    "xtr = xtr/255\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "xts = xts.reshape(xts.shape[0],28,28,1)\n",
    "xts = xts/255"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# OneHot Encoding ytr array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.utils.np_utils import to_categorical\n",
    "ytr=to_categorical(ytr,10) #10 represents number of classes/categories \n",
    "yts= to_categorical(yts,10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 10)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ytr.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Creating neural net model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense , Flatten\n",
    "from keras.layers.convolutional import Conv2D ,MaxPooling2D\n",
    "from keras.optimizers import Adam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_4\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_6 (Conv2D)            (None, 24, 24, 40)        1040      \n",
      "_________________________________________________________________\n",
      "conv2d_7 (Conv2D)            (None, 20, 20, 32)        32032     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2 (None, 10, 10, 32)        0         \n",
      "_________________________________________________________________\n",
      "flatten_4 (Flatten)          (None, 3200)              0         \n",
      "_________________________________________________________________\n",
      "dense_7 (Dense)              (None, 256)               819456    \n",
      "_________________________________________________________________\n",
      "dense_8 (Dense)              (None, 10)                2570      \n",
      "=================================================================\n",
      "Total params: 855,098\n",
      "Trainable params: 855,098\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "def create_model():\n",
    "    model = Sequential()\n",
    "    model.add(Conv2D(40,kernel_size=(5,5),input_shape = (28,28,1),activation = 'relu'))\n",
    "    model.add(Conv2D(32,(5,5),activation = 'relu'))\n",
    "    model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "    model.add(Flatten())\n",
    "    model.add(Dense(256,activation='relu'))\n",
    "    model.add(Dense(10,activation='softmax'))\n",
    "    model.compile(Adam(lr=0.001),loss='categorical_crossentropy',metrics=['accuracy'])\n",
    "    return model\n",
    "model = create_model()\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/10\n",
      "60000/60000 [==============================] - 192s 3ms/step - loss: 0.1120 - accuracy: 0.9654 - val_loss: 0.0413 - val_accuracy: 0.9872\n",
      "Epoch 2/10\n",
      "60000/60000 [==============================] - 194s 3ms/step - loss: 0.0375 - accuracy: 0.9888 - val_loss: 0.0325 - val_accuracy: 0.9892\n",
      "Epoch 3/10\n",
      "60000/60000 [==============================] - 196s 3ms/step - loss: 0.0247 - accuracy: 0.9923 - val_loss: 0.0505 - val_accuracy: 0.9854\n",
      "Epoch 4/10\n",
      "60000/60000 [==============================] - 183s 3ms/step - loss: 0.0178 - accuracy: 0.9943 - val_loss: 0.0383 - val_accuracy: 0.9898\n",
      "Epoch 5/10\n",
      "60000/60000 [==============================] - 187s 3ms/step - loss: 0.0134 - accuracy: 0.9952 - val_loss: 0.0421 - val_accuracy: 0.9895\n",
      "Epoch 6/10\n",
      "60000/60000 [==============================] - 191s 3ms/step - loss: 0.0095 - accuracy: 0.9967 - val_loss: 0.0454 - val_accuracy: 0.9885\n",
      "Epoch 7/10\n",
      "60000/60000 [==============================] - 168s 3ms/step - loss: 0.0091 - accuracy: 0.9970 - val_loss: 0.0374 - val_accuracy: 0.9913\n",
      "Epoch 8/10\n",
      "60000/60000 [==============================] - 170s 3ms/step - loss: 0.0074 - accuracy: 0.9976 - val_loss: 0.0285 - val_accuracy: 0.9928\n",
      "Epoch 9/10\n",
      "60000/60000 [==============================] - 186s 3ms/step - loss: 0.0057 - accuracy: 0.9980 - val_loss: 0.0307 - val_accuracy: 0.9928\n",
      "Epoch 10/10\n",
      "60000/60000 [==============================] - 182s 3ms/step - loss: 0.0055 - accuracy: 0.9982 - val_loss: 0.0656 - val_accuracy: 0.9866\n"
     ]
    }
   ],
   "source": [
    "h = model.fit(xtr,ytr,epochs=10,verbose=1,validation_data=(xts,yts),batch_size=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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"
    }
   ],
   "source": [
    "plt.plot(h.history['loss'],label='loss')  #val_loss decreasing = overfitted , val_loss increasing = underfitted\n",
    "plt.plot(h.history['val_loss'],label = 'val_loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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"
    }
   ],
   "source": [
    "plt.plot(h.history['accuracy'],label='acc')  #val_loss decreasing = overfitted , val_loss increasing = underfitted\n",
    "plt.plot(h.history['val_accuracy'],label = 'val_acc')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predicting images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "img=cv2.imread(r\"C:\\Users\\amanv\\Documents\\ai&mlclass\\Datasets\\digit.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(389, 385, 3)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "img.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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"
    }
   ],
   "source": [
    "img=cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)\n",
    "img=cv2.resize(img,(28,28))\n",
    "plt.imshow(img,cmap='gray')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "img = img.reshape(1,28,28,1)\n",
    "img = img/255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6], dtype=int64)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.predict_classes(img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "y_pred = model.predict_classes(xts)\n",
    "print(confusion_matrix(y_true,y_pred))"
   ]
  },
  {
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}