mnist.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "emN7PokBgOL3"
   },
   "source": [
    "# **Accurate Image Classification with MNIST Dataset**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### CNN Notes: https://drive.google.com/file/d/1Qp6rV9yLGyj_S8_MlAu4XLJybuLA3cII/view?usp=sharing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2SzhuYI6gh8p"
   },
   "source": [
    "## Importing Required Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "Vwrr0GlSpiF-"
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline \n",
    "# Importing the required Keras modules containing model and layers\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Conv2D, Dropout, Flatten, MaxPooling2D"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xJujKMtxf95z"
   },
   "source": [
    "## Downloading the MNIST Dataset\n",
    "#### Tensorflow and Keras allow us to import and download the MNIST dataset directly from their API\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4FT5Ufe_ffWc",
    "outputId": "0b3cb516-041e-4bcd-fb9a-08c339cf5a68"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
      "11493376/11490434 [==============================] - 38s 3us/step\n"
     ]
    }
   ],
   "source": [
    "(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Wee0yKC8gtFn"
   },
   "source": [
    "### Greyscale visualization of an image fom the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 298
    },
    "id": "EdG7_qWqplAu",
    "outputId": "9b1f109c-1d03-4cce-a59e-05a49c4b583e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2302a1aa2c8>"
      ]
     },
     "execution_count": 3,
     "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": [
    "img_index = 6547 \n",
    "print(y_train[img_index]) \n",
    "plt.imshow(x_train[img_index], cmap='Greys')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "UxIY_pZKg3yd"
   },
   "source": [
    "## Reshaping and Normalizing the Images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "0xBVVDS0qGV4",
    "outputId": "d9a0a9b4-958e-4082-ef19-ea7a90294a71"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 28, 28)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "usrsc7NTqQDp",
    "outputId": "c5ecfad3-a440-48b6-d9b9-50ab2534897a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: (60000, 28, 28, 1)\n",
      "Number of images in x_train 60000\n",
      "Number of images in x_test 10000\n"
     ]
    }
   ],
   "source": [
    "x_train = x_train.reshape(x_train.shape[0], 28, 28, 1)\n",
    "x_test = x_test.reshape(x_test.shape[0], 28, 28, 1)\n",
    "input_shape = (28, 28, 1)\n",
    "\n",
    "x_train = x_train.astype('float32')\n",
    "x_test = x_test.astype('float32')\n",
    "\n",
    "x_train /= 255\n",
    "x_test /= 255\n",
    "print('x_train shape:', x_train.shape)\n",
    "print('Number of images in x_train', x_train.shape[0])\n",
    "print('Number of images in x_test', x_test.shape[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rZyLZN_6g6kV"
   },
   "source": [
    "## Creating your own CNN model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "B4-bOjSJhOrX"
   },
   "source": [
    "### Adding layers into the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "jMqqsDadqTuQ"
   },
   "outputs": [],
   "source": [
    "# Creating a Sequential Model and adding the layers\n",
    "model = Sequential()\n",
    "model.add(Conv2D(28, kernel_size=(3,3), input_shape=input_shape))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(Flatten()) \n",
    "model.add(Dense(128, activation=tf.nn.relu))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(10,activation=tf.nn.softmax))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Viz87YmEhJqH"
   },
   "source": [
    "### Fitting, Compilation and Evaluation\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "5Ors35YEqXFw",
    "outputId": "fbe6b873-a509-4782-f3aa-706a5710f69b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/10\n",
      "60000/60000 [==============================] - 138s 2ms/sample - loss: 0.2095 - accuracy: 0.9376 - val_loss: 0.0881 - val_accuracy: 0.9721\n",
      "Epoch 2/10\n",
      "60000/60000 [==============================] - 128s 2ms/sample - loss: 0.0802 - accuracy: 0.9752 - val_loss: 0.0595 - val_accuracy: 0.9814\n",
      "Epoch 3/10\n",
      "60000/60000 [==============================] - 128s 2ms/sample - loss: 0.0576 - accuracy: 0.9816 - val_loss: 0.0527 - val_accuracy: 0.9822\n",
      "Epoch 4/10\n",
      "60000/60000 [==============================] - 130s 2ms/sample - loss: 0.0425 - accuracy: 0.9864 - val_loss: 0.0508 - val_accuracy: 0.9849\n",
      "Epoch 5/10\n",
      "60000/60000 [==============================] - 146s 2ms/sample - loss: 0.0354 - accuracy: 0.9887 - val_loss: 0.0478 - val_accuracy: 0.9854\n",
      "Epoch 6/10\n",
      "60000/60000 [==============================] - 130s 2ms/sample - loss: 0.0281 - accuracy: 0.9908 - val_loss: 0.0561 - val_accuracy: 0.9824\n",
      "Epoch 7/10\n",
      "60000/60000 [==============================] - 123s 2ms/sample - loss: 0.0239 - accuracy: 0.9922 - val_loss: 0.0589 - val_accuracy: 0.9852\n",
      "Epoch 8/10\n",
      "60000/60000 [==============================] - 128s 2ms/sample - loss: 0.0235 - accuracy: 0.9920 - val_loss: 0.0535 - val_accuracy: 0.9849\n",
      "Epoch 9/10\n",
      "60000/60000 [==============================] - 128s 2ms/sample - loss: 0.0180 - accuracy: 0.9935 - val_loss: 0.0593 - val_accuracy: 0.9853\n",
      "Epoch 10/10\n",
      "60000/60000 [==============================] - 127s 2ms/sample - loss: 0.0170 - accuracy: 0.9942 - val_loss: 0.0659 - val_accuracy: 0.9840\n"
     ]
    }
   ],
   "source": [
    "model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
    "history=model.fit(x=x_train,y=y_train, epochs=10, validation_data= ((x_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "t5X2vLjWqaS5",
    "outputId": "335062ea-b47f-4d98-a248-ff48cd3498b6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000/10000 [==============================] - 6s 591us/sample - loss: 0.0659 - accuracy: 0.9840\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.06590935274491885, 0.984]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(x_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cg-Y9mZfhYvE"
   },
   "source": [
    "## Graphical visualization of the results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "id": "-wkAXIipqkXE",
    "outputId": "670797c8-0462-4c62-c4f0-d780451ac395"
   },
   "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": [
    "loss_train = history.history['loss']\n",
    "loss_val = history.history['val_loss']\n",
    "epochs = range(1,11)\n",
    "plt.plot(epochs, loss_train, 'g', label='Training loss')\n",
    "plt.plot(epochs, loss_val, 'b', label='validation loss')\n",
    "plt.title('Training and Validation loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "id": "d8uhKc-pqpf-",
    "outputId": "b403a5cc-25a2-4abf-c1fa-b41c86362834"
   },
   "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": [
    "accuracy_train = history.history['accuracy']\n",
    "accuracy_val = history.history['val_accuracy']\n",
    "epochs = range(1,11)\n",
    "plt.plot(epochs, accuracy_train, 'g', label='Training accuracy')\n",
    "plt.plot(epochs, accuracy_val, 'b', label='validation accuracy')\n",
    "plt.title('Training and Validation accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2nqrCEY5hl67"
   },
   "source": [
    "## Making individual predictions as a method of testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 70
    },
    "id": "pi7jmP3ZqzFi",
    "outputId": "285919de-c4e5-4569-bcba-348982d5588c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9\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"
    }
   ],
   "source": [
    "image_index = 3798\n",
    "plt.imshow(x_test[image_index].reshape(28, 28),cmap='Greys')\n",
    "pred = model.predict(x_test[image_index].reshape(1, 28, 28, 1))\n",
    "print(pred.argmax())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "MdCBaF6hhstR"
   },
   "source": [
    "# **THANK YOU!!**"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "mnist.ipynb",
   "provenance": []
  },
  "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": 1
}