DEEP_lEARNING_WITH_KERAS.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "DEEP lEARNING WITH KERAS.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/veshitala/Basics-of-Unsupervised-learning/blob/master/DEEP_lEARNING_WITH_KERAS.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "CxAkzwScLKzQ",
        "colab_type": "code",
        "outputId": "c305f678-7d4b-4891-b7a1-0fdfa61fd1f3",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "#   Train a ConvNet on the MNIST fashion data.  This data consists of 10 classes of fashion images such\n",
        "#   as shorts, dresses, shoes, purses, etc.  These images replace the handwritten digits in the classic MNIST dataset.\n",
        "#   This change makes it harder to get a high score and more closely reflects real world usage of image \n",
        "#   classification.  And at the same time, is still small enough for the average PC to train in a short time.\n",
        "# See https://github.com/zalandoresearch/fashion-mnist for information and code on Fashion MNIST\n",
        "\n",
        "# This code is based on MNIST example found at Keras.io\n",
        "\n",
        "from __future__ import print_function\n",
        "import keras\n",
        "from keras.datasets import fashion_mnist  # new with Keras 2.1.2.  Yah!!\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout, Flatten\n",
        "from keras.layers import Conv2D, MaxPooling2D\n",
        "from keras import backend as K\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "id": "4gXwUMhsNkph",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import seaborn as sns"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "gjOfJk6vLOBI",
        "colab_type": "code",
        "outputId": "fd3dedad-8490-4ebd-8446-a90967f11541",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 158
        }
      },
      "cell_type": "code",
      "source": [
        "#   Supress warning and informational messages\n",
        "import os\n",
        "os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3' \n",
        "\n",
        "# Number of classes - do not change unless the data changes\n",
        "num_classes = 10\n",
        "\n",
        "# sizes of batch and # of epochs of data\n",
        "batch_size = 128\n",
        "epochs = 24\n",
        "\n",
        "# input image dimensions\n",
        "img_rows, img_cols = 28, 28\n",
        "\n",
        "# the data, shuffled and split between train and test sets\n",
        "(x_train, y_train), (x_test, y_test) = fashion_mnist.load_data() "
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-labels-idx1-ubyte.gz\n",
            "32768/29515 [=================================] - 0s 3us/step\n",
            "Downloading data from http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/train-images-idx3-ubyte.gz\n",
            "26427392/26421880 [==============================] - 8s 0us/step\n",
            "Downloading data from http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-labels-idx1-ubyte.gz\n",
            "8192/5148 [===============================================] - 0s 0us/step\n",
            "Downloading data from http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/t10k-images-idx3-ubyte.gz\n",
            "4423680/4422102 [==============================] - 4s 1us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "6eU14gwSLkVT",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "#   Deal with format issues between different backends.  Some put the # of channels in the image before the width and height of image.\n",
        "if K.image_data_format() == 'channels_first':\n",
        "    x_train = x_train.reshape(x_train.shape[0], 1, img_rows, img_cols)\n",
        "    x_test = X_test.reshape(x_test.shape[0], 1, img_rows, img_cols)\n",
        "    input_shape = (1, img_rows, img_cols)\n",
        "else:\n",
        "    x_train = x_train.reshape(x_train.shape[0], img_rows, img_cols, 1)\n",
        "    x_test = x_test.reshape(x_test.shape[0], img_rows, img_cols, 1)\n",
        "    input_shape = (img_rows, img_cols, 1)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "lOjLWR2TLtMj",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "#   Type convert and scale the test and training data\n",
        "x_train = x_train.astype('float32')\n",
        "x_test = x_test.astype('float32')\n",
        "x_train /= 255\n",
        "x_test /= 255"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ywclracpL3sj",
        "colab_type": "code",
        "outputId": "8ae99323-b499-4e84-ee75-bc52f4e11e62",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 70
        }
      },
      "cell_type": "code",
      "source": [
        "print('x_train shape:', x_train.shape)\n",
        "print('x_test shape:', x_test.shape)\n",
        "\n",
        "print(y_train[:4])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "x_train shape: (60000, 28, 28, 1)\n",
            "x_test shape: (10000, 28, 28, 1)\n",
            "[9 0 0 3]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "XffahzauN9DM",
        "colab_type": "code",
        "outputId": "58c03815-2c6c-49ba-d3de-eec4a5c65de2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "y_train"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9, 0, 0, ..., 3, 0, 5], dtype=uint8)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "metadata": {
        "id": "fD2QYAKjN_zR",
        "colab_type": "code",
        "outputId": "084ddab5-a868-4a0d-814c-4eed64ae3f71",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "y_test"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9, 2, 1, ..., 8, 1, 5], dtype=uint8)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "metadata": {
        "id": "9L4J4gpMMBlq",
        "colab_type": "code",
        "outputId": "86e2c7dc-d64b-443a-f85b-1ccd5670ed96",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 702
        }
      },
      "cell_type": "code",
      "source": [
        "w=10\n",
        "h=10\n",
        "fig=plt.figure(figsize=(12, 12))\n",
        "columns = 4\n",
        "rows = 5\n",
        "for i in range(1, columns*rows +1):\n",
        "    img = np.random.randint(100)\n",
        "    fig.add_subplot(rows, columns, i)\n",
        "    plt.imshow(x_train[img][: ,: ,0])\n",
        "plt.show()\n",
        "\n",
        "#plt.imshow(x_train[1][: ,: ,0])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAp0AAAKtCAYAAABlr9WTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niIg8x00nEREREXnu/wEkjdB3mXNg0QAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fc31e0a1cc0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "FsFi4qZENtU0",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# convert class vectors to binary class matrices.  One-hot encoding\n",
        "#  3 => 0 0 0 1 0 0 0 0 0 0 and 1 => 0 1 0 0 0 0 0 0 0 0 \n",
        "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
        "y_test = keras.utils.to_categorical(y_test, num_classes)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "8zxAsAi6N7vb",
        "colab_type": "code",
        "outputId": "d03c3c96-e50f-4e66-c4f4-017dcc9e263c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 140
        }
      },
      "cell_type": "code",
      "source": [
        "display(y_train)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "array([[0., 0., 0., ..., 0., 0., 1.],\n",
              "       [1., 0., 0., ..., 0., 0., 0.],\n",
              "       [1., 0., 0., ..., 0., 0., 0.],\n",
              "       ...,\n",
              "       [0., 0., 0., ..., 0., 0., 0.],\n",
              "       [1., 0., 0., ..., 0., 0., 0.],\n",
              "       [0., 0., 0., ..., 0., 0., 0.]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "AYaY5T2yOJdd",
        "colab_type": "code",
        "outputId": "c363c7bd-ffcc-4d48-be24-c382745e48e8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 140
        }
      },
      "cell_type": "code",
      "source": [
        "display(y_test)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "array([[0., 0., 0., ..., 0., 0., 1.],\n",
              "       [0., 0., 1., ..., 0., 0., 0.],\n",
              "       [0., 1., 0., ..., 0., 0., 0.],\n",
              "       ...,\n",
              "       [0., 0., 0., ..., 0., 1., 0.],\n",
              "       [0., 1., 0., ..., 0., 0., 0.],\n",
              "       [0., 0., 0., ..., 0., 0., 0.]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "M3PEF14QOoMd",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model = Sequential()\n",
        "\n",
        "model.add(Conv2D(32, kernel_size=(3, 3),activation='relu', input_shape=input_shape))\n",
        "\n",
        "\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Conv2D(32, (3, 3), activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(128, activation='relu'))\n",
        "model.add(Dropout(0.5))\n",
        "model.add(Dense(num_classes, activation='softmax'))\n",
        "\n",
        "                "
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "KiblMz7NSHPI",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "#   define compile to minimize categorical loss, use ada delta optimized, and optimize to maximizing accuracy\n",
        "model.compile(loss=keras.losses.categorical_crossentropy,\n",
        "              optimizer=keras.optimizers.Adadelta(),\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "VCsMSWqoSUcu",
        "colab_type": "code",
        "outputId": "490109b4-5eb5-4735-b05c-613ea887f66f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 879
        }
      },
      "cell_type": "code",
      "source": [
        "hist = model.fit(x_train, y_train,\n",
        "          batch_size=batch_size,\n",
        "          epochs=epochs,\n",
        "          verbose=1,\n",
        "          validation_data=(x_test, y_test))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 60000 samples, validate on 10000 samples\n",
            "Epoch 1/24\n",
            "60000/60000 [==============================] - 8s 136us/step - loss: 0.6934 - acc: 0.7472 - val_loss: 0.4461 - val_acc: 0.8382\n",
            "Epoch 2/24\n",
            "60000/60000 [==============================] - 6s 105us/step - loss: 0.4330 - acc: 0.8443 - val_loss: 0.4076 - val_acc: 0.8452\n",
            "Epoch 3/24\n",
            "60000/60000 [==============================] - 6s 104us/step - loss: 0.3752 - acc: 0.8656 - val_loss: 0.3332 - val_acc: 0.8786\n",
            "Epoch 4/24\n",
            "60000/60000 [==============================] - 6s 104us/step - loss: 0.3431 - acc: 0.8758 - val_loss: 0.3244 - val_acc: 0.8844\n",
            "Epoch 5/24\n",
            "60000/60000 [==============================] - 6s 105us/step - loss: 0.3233 - acc: 0.8834 - val_loss: 0.3125 - val_acc: 0.8884\n",
            "Epoch 6/24\n",
            "60000/60000 [==============================] - 6s 105us/step - loss: 0.3027 - acc: 0.8902 - val_loss: 0.3141 - val_acc: 0.8874\n",
            "Epoch 7/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.2890 - acc: 0.8956 - val_loss: 0.2947 - val_acc: 0.8935\n",
            "Epoch 8/24\n",
            "60000/60000 [==============================] - 6s 105us/step - loss: 0.2787 - acc: 0.8995 - val_loss: 0.2822 - val_acc: 0.8988\n",
            "Epoch 9/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.2658 - acc: 0.9028 - val_loss: 0.2821 - val_acc: 0.9002\n",
            "Epoch 10/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.2582 - acc: 0.9071 - val_loss: 0.2708 - val_acc: 0.9033\n",
            "Epoch 11/24\n",
            "60000/60000 [==============================] - 6s 107us/step - loss: 0.2498 - acc: 0.9094 - val_loss: 0.2701 - val_acc: 0.9000\n",
            "Epoch 12/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.2404 - acc: 0.9131 - val_loss: 0.2726 - val_acc: 0.9019\n",
            "Epoch 13/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.2334 - acc: 0.9157 - val_loss: 0.2568 - val_acc: 0.9061\n",
            "Epoch 14/24\n",
            "60000/60000 [==============================] - 6s 107us/step - loss: 0.2282 - acc: 0.9172 - val_loss: 0.2590 - val_acc: 0.9082\n",
            "Epoch 15/24\n",
            "60000/60000 [==============================] - 6s 107us/step - loss: 0.2192 - acc: 0.9197 - val_loss: 0.2488 - val_acc: 0.9080\n",
            "Epoch 16/24\n",
            "60000/60000 [==============================] - 6s 107us/step - loss: 0.2153 - acc: 0.9224 - val_loss: 0.2497 - val_acc: 0.9099\n",
            "Epoch 17/24\n",
            "60000/60000 [==============================] - 6s 107us/step - loss: 0.2108 - acc: 0.9227 - val_loss: 0.2549 - val_acc: 0.9120\n",
            "Epoch 18/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.2061 - acc: 0.9245 - val_loss: 0.2716 - val_acc: 0.9031\n",
            "Epoch 19/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.2012 - acc: 0.9258 - val_loss: 0.2538 - val_acc: 0.9122\n",
            "Epoch 20/24\n",
            "60000/60000 [==============================] - 6s 104us/step - loss: 0.1983 - acc: 0.9267 - val_loss: 0.2432 - val_acc: 0.9141\n",
            "Epoch 21/24\n",
            "60000/60000 [==============================] - 6s 105us/step - loss: 0.1939 - acc: 0.9294 - val_loss: 0.2509 - val_acc: 0.9108\n",
            "Epoch 22/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.1883 - acc: 0.9303 - val_loss: 0.2552 - val_acc: 0.9099\n",
            "Epoch 23/24\n",
            "60000/60000 [==============================] - 6s 105us/step - loss: 0.1850 - acc: 0.9308 - val_loss: 0.2689 - val_acc: 0.9128\n",
            "Epoch 24/24\n",
            "60000/60000 [==============================] - 6s 106us/step - loss: 0.1802 - acc: 0.9335 - val_loss: 0.2440 - val_acc: 0.9155\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "-pIFfcuoRiDm",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "val_accuracy = hist.history['val_acc']\n",
        "accuracy = hist.history['acc']\n",
        "val_loss = hist.history['loss']\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "81X_6l4xRt8g",
        "colab_type": "code",
        "outputId": "bb11d030-e89c-4089-9222-4236a287b869",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 347
        }
      },
      "cell_type": "code",
      "source": [
        "plt.plot(val_accuracy , label  = 'validation accuracy')\n",
        "\n",
        "plt.plot(accuracy , label = 'model accuracy')\n",
        "\n",
        "plt.plot(val_loss , label = 'validation loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeEAAAFKCAYAAAAqkecjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n2UgZfRC5BwAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fc28842b208>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "-1iBZrVjSnza",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "score = model.evaluate(x_test, y_test, verbose=0)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "JQXYluuhRoKE",
        "colab_type": "code",
        "outputId": "54a4f527-5d13-41df-fa85-8b882ed057ea",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        }
      },
      "cell_type": "code",
      "source": [
        "print('accuracy: ',score[1])\n",
        "print('loss: ',score[0])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "accuracy:  0.9155\n",
            "loss:  0.24396675983071328\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "v7pRyjsaRkTn",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "test = x_test[0]\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "jsGIkhSYU4M9",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "test = test.reshape(1,28 ,28,1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "bULFf5jKVFX4",
        "colab_type": "code",
        "outputId": "b6dab8b1-113b-4ccf-b108-fb9b7379d020",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "test.shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1, 28, 28, 1)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "metadata": {
        "id": "kqqewG5wVSrs",
        "colab_type": "code",
        "outputId": "3abd9123-ab91-4f4d-cf4e-3947dd9b5331",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "model.predict(test).argmax()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "9"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "metadata": {
        "id": "31xcbaH8VfB3",
        "colab_type": "code",
        "outputId": "4545ebdb-417f-4923-fba0-d1206d36f25c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "y_test[0]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 1.], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "metadata": {
        "id": "mexE1WxIVZnh",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}