Week 4 ─ Multi-class Classifier.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Exercise 8 - Answer.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "wYtuKeK0dImp",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import csv\n",
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from google.colab import files"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EmMyh9_mkDHF",
        "colab_type": "text"
      },
      "source": [
        "The data for this exercise is available at: https://www.kaggle.com/datamunge/sign-language-mnist/home\n",
        "\n",
        "Sign up and download to find 2 CSV files: sign_mnist_test.csv and sign_mnist_train.csv -- You will upload both of them using this button before you can continue.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IcLOZlnnc_N7",
        "colab_type": "code",
        "outputId": "f902689f-06ab-422b-a15d-ceee438a19cb",
        "colab": {
          "height": 106
        }
      },
      "source": [
        "uploaded=files.upload()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-9979a0a3-4162-45c4-98fc-4e133ab5ad52\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-9979a0a3-4162-45c4-98fc-4e133ab5ad52\">\n",
              "      Upload widget is only available when the cell has been executed in the\n",
              "      current browser session. Please rerun this cell to enable.\n",
              "      </output>\n",
              "      <script src=\"/nbextensions/google.colab/files.js\"></script> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML at 0x7f2bf1cffa90>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving sign_mnist_test.csv to sign_mnist_test.csv\n",
            "Saving sign_mnist_train.csv to sign_mnist_train.csv\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4kxw-_rmcnVu",
        "colab_type": "code",
        "outputId": "3d62714a-acc3-4aef-98ec-10fee8e7ab50",
        "colab": {
          "height": 86
        }
      },
      "source": [
        "def get_data(filename):\n",
        "    with open(filename) as training_file:\n",
        "        csv_reader = csv.reader(training_file, delimiter=',')\n",
        "        first_line = True\n",
        "        temp_images = []\n",
        "        temp_labels = []\n",
        "        for row in csv_reader:\n",
        "            if first_line:\n",
        "                # print(\"Ignoring first line\")\n",
        "                first_line = False\n",
        "            else:\n",
        "                temp_labels.append(row[0])\n",
        "                image_data = row[1:785]\n",
        "                image_data_as_array = np.array_split(image_data, 28)\n",
        "                temp_images.append(image_data_as_array)\n",
        "        images = np.array(temp_images).astype('float')\n",
        "        labels = np.array(temp_labels).astype('float')\n",
        "    return images, labels\n",
        "\n",
        "\n",
        "training_images, training_labels = get_data('sign_mnist_train.csv')\n",
        "testing_images, testing_labels = get_data('sign_mnist_test.csv')\n",
        "\n",
        "print(training_images.shape)\n",
        "print(training_labels.shape)\n",
        "print(testing_images.shape)\n",
        "print(testing_labels.shape)\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(27455, 28, 28)\n",
            "(27455,)\n",
            "(7172, 28, 28)\n",
            "(7172,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "awoqRpyZdQkD",
        "colab_type": "code",
        "outputId": "a223a6f6-29f2-4f5f-8ccf-80f36d3368d1",
        "colab": {
          "height": 52
        }
      },
      "source": [
        "training_images = np.expand_dims(training_images, axis=3)\n",
        "testing_images = np.expand_dims(testing_images, axis=3)\n",
        "\n",
        "train_datagen = ImageDataGenerator(\n",
        "    rescale=1. / 255,\n",
        "    rotation_range=40,\n",
        "    width_shift_range=0.2,\n",
        "    height_shift_range=0.2,\n",
        "    shear_range=0.2,\n",
        "    zoom_range=0.2,\n",
        "    horizontal_flip=True,\n",
        "    fill_mode='nearest')\n",
        "\n",
        "validation_datagen = ImageDataGenerator(\n",
        "    rescale=1. / 255)\n",
        "\n",
        "print(training_images.shape)\n",
        "print(testing_images.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(27455, 28, 28, 1)\n",
            "(7172, 28, 28, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Rmb7S32cgRqS",
        "colab_type": "code",
        "outputId": "d1cbfc47-3a0a-4605-f1be-fa968032283d",
        "colab": {
          "height": 570
        }
      },
      "source": [
        "model = tf.keras.models.Sequential([\n",
        "    tf.keras.layers.Conv2D(64, (3, 3), activation='relu', input_shape=(28, 28, 1)),\n",
        "    tf.keras.layers.MaxPooling2D(2, 2),\n",
        "    tf.keras.layers.Conv2D(64, (3, 3), activation='relu'),\n",
        "    tf.keras.layers.MaxPooling2D(2, 2),\n",
        "    tf.keras.layers.Flatten(),\n",
        "    tf.keras.layers.Dense(128, activation=tf.nn.relu),\n",
        "    tf.keras.layers.Dense(26, activation=tf.nn.softmax)])\n",
        "\n",
        "model.compile(optimizer = tf.train.AdamOptimizer(),\n",
        "              loss = 'sparse_categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "history = model.fit_generator(train_datagen.flow(training_images, training_labels, batch_size=32),\n",
        "                              steps_per_epoch=len(training_images) / 32,\n",
        "                              epochs=15,\n",
        "                              validation_data=validation_datagen.flow(testing_images, testing_labels, batch_size=32),\n",
        "                              validation_steps=len(testing_images) / 32)\n",
        "\n",
        "model.evaluate(testing_images, testing_labels)\n"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/15\n",
            "857/857 [==============================] - 21s 24ms/step - loss: 2.8540 - acc: 0.1562 - val_loss: 1.9644 - val_acc: 0.3100\n",
            "Epoch 2/15\n",
            "857/857 [==============================] - 46s 54ms/step - loss: 2.1131 - acc: 0.2188 - val_loss: 1.5558 - val_acc: 0.4855\n",
            "Epoch 3/15\n",
            "857/857 [==============================] - 40s 47ms/step - loss: 1.7423 - acc: 0.3438 - val_loss: 1.1643 - val_acc: 0.6115\n",
            "Epoch 4/15\n",
            "857/857 [==============================] - 30s 35ms/step - loss: 1.4933 - acc: 0.4688 - val_loss: 1.1572 - val_acc: 0.6064\n",
            "Epoch 5/15\n",
            "857/857 [==============================] - 22s 26ms/step - loss: 1.3521 - acc: 0.7500 - val_loss: 0.8973 - val_acc: 0.7167\n",
            "Epoch 6/15\n",
            "857/857 [==============================] - 21s 24ms/step - loss: 1.2332 - acc: 0.5625 - val_loss: 0.8082 - val_acc: 0.7200\n",
            "Epoch 7/15\n",
            "857/857 [==============================] - 44s 51ms/step - loss: 1.1631 - acc: 0.5938 - val_loss: 0.8671 - val_acc: 0.7352\n",
            "Epoch 8/15\n",
            "857/857 [==============================] - 58s 67ms/step - loss: 1.0857 - acc: 0.7188 - val_loss: 0.7608 - val_acc: 0.7949\n",
            "Epoch 9/15\n",
            "857/857 [==============================] - 23s 26ms/step - loss: 1.0197 - acc: 0.6562 - val_loss: 0.6978 - val_acc: 0.7674\n",
            "Epoch 10/15\n",
            "857/857 [==============================] - 28s 33ms/step - loss: 0.9711 - acc: 0.6875 - val_loss: 0.7027 - val_acc: 0.7984\n",
            "Epoch 11/15\n",
            "857/857 [==============================] - 53s 61ms/step - loss: 0.9076 - acc: 0.5625 - val_loss: 0.5784 - val_acc: 0.8238\n",
            "Epoch 12/15\n",
            "857/857 [==============================] - 64s 75ms/step - loss: 0.8764 - acc: 0.5938 - val_loss: 0.6079 - val_acc: 0.8133\n",
            "Epoch 13/15\n",
            "857/857 [==============================] - 28s 33ms/step - loss: 0.8410 - acc: 0.7500 - val_loss: 0.4547 - val_acc: 0.8182\n",
            "Epoch 14/15\n",
            "857/857 [==============================] - 22s 26ms/step - loss: 0.8045 - acc: 0.5312 - val_loss: 0.2415 - val_acc: 0.8496\n",
            "Epoch 15/15\n",
            "857/857 [==============================] - 47s 55ms/step - loss: 0.7836 - acc: 0.7188 - val_loss: 0.3857 - val_acc: 0.8425\n",
            "7172/7172 [==============================] - 4s 596us/step - loss: 6.9243 - acc: 0.5661\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[6.92426086682151, 0.56609035]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_Q3Zpr46dsij",
        "colab_type": "code",
        "outputId": "fd4001fb-9a21-4192-dbdb-9a2f2174e769",
        "colab": {
          "height": 561
        }
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "acc = history.history['acc']\n",
        "val_acc = history.history['val_acc']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(len(acc))\n",
        "\n",
        "plt.plot(epochs, acc, 'r', label='Training accuracy')\n",
        "plt.plot(epochs, val_acc, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend()\n",
        "plt.figure()\n",
        "\n",
        "plt.plot(epochs, loss, 'r', label='Training Loss')\n",
        "plt.plot(epochs, val_loss, 'b', label='Validation Loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEQCAYAAABY5nToAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f2be0af5310>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEQCAYAAABY5nToAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEMufiEiDWP5ERBr0X+qpJyKRhKK8AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f2bd4a223d0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}