The_simplest_fashion_mnist_+_Confusion_Matrix.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "The-simplest-fashion-mnist + Confusion Matrix.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": [],
      "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/lc0/deeplearning-playground/blob/master/The_simplest_fashion_mnist_+_Confusion_Matrix.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "metadata": {
        "id": "_96YiOnl8niT",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "\n",
        "# tf.enable_eager_execution()\n",
        "\n",
        "fashion_mnist = keras.datasets.fashion_mnist\n",
        "(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()\n",
        "\n",
        "train_images_norm = train_images/256\n",
        "test_images_norm = test_images/256"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "bedaOE1a8veg",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model = keras.Sequential([\n",
        "    keras.layers.Flatten(input_shape=(28, 28)),\n",
        "    keras.layers.Dense(128, activation=tf.nn.relu),\n",
        "    keras.layers.Dense(10, activation=tf.nn.softmax)\n",
        "])\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "4dx_arvPJ9xA",
        "colab_type": "code",
        "outputId": "e362500f-535f-45c3-866d-d0c48f516f54",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        }
      },
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "flatten (Flatten)            (None, 784)               0         \n",
            "_________________________________________________________________\n",
            "dense (Dense)                (None, 128)               100480    \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 10)                1290      \n",
            "=================================================================\n",
            "Total params: 101,770\n",
            "Trainable params: 101,770\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "5TaydQ0i8zoW",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model.compile(optimizer=tf.train.AdamOptimizer(),\n",
        "              loss='sparse_categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "pzuSpy6e82ae",
        "colab_type": "code",
        "outputId": "d17d9a21-0d58-4fe5-a7a0-853007568033",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        }
      },
      "cell_type": "code",
      "source": [
        "model.fit(train_images_norm, train_labels, epochs=5)\n"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/5\n",
            "60000/60000 [==============================] - 7s 119us/step - loss: 0.5024 - acc: 0.8229\n",
            "Epoch 2/5\n",
            "60000/60000 [==============================] - 6s 102us/step - loss: 0.3790 - acc: 0.8634\n",
            "Epoch 3/5\n",
            "60000/60000 [==============================] - 6s 105us/step - loss: 0.3390 - acc: 0.8759\n",
            "Epoch 4/5\n",
            "60000/60000 [==============================] - 6s 102us/step - loss: 0.3140 - acc: 0.8842\n",
            "Epoch 5/5\n",
            "60000/60000 [==============================] - 6s 104us/step - loss: 0.2985 - acc: 0.8905\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f9396ff00b8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "metadata": {
        "id": "TXDglY4n8612",
        "colab_type": "code",
        "outputId": "adb966ce-dd33-4830-fcef-103c7795affc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "loss, accuracy = model.evaluate(test_images_norm, test_labels)\n",
        "print('Test loss:', loss)\n",
        "print('Test accuracy:', accuracy)"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "10000/10000 [==============================] - 1s 75us/step\n",
            "Test loss: 0.341156424677372\n",
            "Test accuracy: 0.8785\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "KqFKr4WVAXhF",
        "colab_type": "code",
        "outputId": "ff569009-be6d-499d-8294-268f642bcc07",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 299
        }
      },
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "\n",
        "%matplotlib inline\n",
        "\n",
        "image_id = np.random.randint(len(test_images))\n",
        "print(\"Image ID {} - {}\".format(image_id, test_labels[image_id]))\n",
        "plt.imshow(test_images[image_id], cmap='Greys')"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Image ID 7678 - 6\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f9390ceacc0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPoAAAD4CAYAAADFJPs2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f9396d9ff60>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "0DJbfIse9LoT",
        "colab_type": "code",
        "outputId": "b26c619a-3f3f-4ae8-d07a-9dffdf7becdf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        }
      },
      "cell_type": "code",
      "source": [
        "image_data = (test_images[image_id]/256)\n",
        "model.predict(image_data.reshape(1, 28, 28))"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1.9141521e-04, 2.3384351e-05, 5.1224966e-02, 4.5451528e-04,\n",
              "        5.0013477e-01, 8.7937156e-07, 4.4793499e-01, 8.1495465e-08,\n",
              "        3.4769622e-05, 1.7106058e-07]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "metadata": {
        "id": "JeVT2hWtAxeP",
        "colab_type": "code",
        "outputId": "dc8fca00-ce6f-4c2b-b6f1-9612c9e1edfd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "np.argmax(model.predict(image_data.reshape(1, 28, 28)))"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "metadata": {
        "id": "Nn3MZOEkBjcc",
        "colab_type": "code",
        "outputId": "be49d95d-c61d-4679-f961-c300b23a4a1d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "cell_type": "code",
      "source": [
        "model.save('fashion-mnist.hd5')"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:TensorFlow optimizers do not make it possible to access optimizer attributes or optimizer state after instantiation. As a result, we cannot save the optimizer as part of the model save file.You will have to compile your model again after loading it. Prefer using a Keras optimizer instead (see keras.io/optimizers).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "-HHCgOkNF9h1",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from google.colab import files\n",
        "\n",
        "files.download('fashion-mnist.hd5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "wcoGCalSD16C",
        "colab_type": "code",
        "outputId": "64256b51-557d-4eda-b68b-75feb7e9cba6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        }
      },
      "cell_type": "code",
      "source": [
        "!ls -laF"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "total 432\n",
            "drwxr-xr-x 1 root root   4096 Nov 18 22:28 ./\n",
            "drwxr-xr-x 1 root root   4096 Nov 18 22:26 ../\n",
            "drwxr-xr-x 4 root root   4096 Nov 15 19:23 .config/\n",
            "-rw-r--r-- 1 root root 422848 Nov 18 23:05 fashion-mnist.hd5\n",
            "drwxr-xr-x 2 root root   4096 Nov 15 19:33 sample_data/\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "X0DdqicuD5Z2",
        "colab_type": "code",
        "outputId": "30965729-180d-46ff-9901-774700d3ffbb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.models import load_model\n",
        "\n",
        "newmodel = load_model('fashion-mnist.hd5')"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:No training configuration found in save file: the model was *not* compiled. Compile it manually.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "YcOfXZKcD_ue",
        "colab_type": "code",
        "outputId": "93776a99-d247-4696-ac8e-9534fc9ecd11",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 367
        }
      },
      "cell_type": "code",
      "source": [
        "newmodel.compile(optimizer=tf.train.AdamOptimizer(),\n",
        "              loss='sparse_categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "\n",
        "image_id = np.random.randint(len(test_images))\n",
        "print(\"Image ID {} - target class {}\".format(image_id, test_labels[image_id]))\n",
        "plt.imshow(test_images[image_id], cmap='Greys')\n",
        "\n",
        "image_data = (test_images[image_id]/256)\n",
        "print(model.predict(image_data.reshape(1, 28, 28)))\n",
        "\n",
        "print(\"\\n\\nPredicted - {}\".format(np.argmax(model.predict(image_data.reshape(1, 28, 28)))))"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Image ID 4900 - target class 8\n",
            "[[1.2879003e-07 1.2247987e-09 6.5735357e-08 2.9130150e-11 1.3501520e-07\n",
            "  1.3028882e-10 2.3023703e-07 7.0167577e-10 9.9999940e-01 1.0344341e-08]]\n",
            "\n",
            "\n",
            "Predicted - 8\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPoAAAD4CAYAAADFJPs2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f937dd847f0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "YENiZQ0j70fx",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Confusion matrix"
      ]
    },
    {
      "metadata": {
        "id": "k8VFi8pW72Zt",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "?tf.confusion_matrix"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "pIBv7Bt076SN",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "58cc1a33-1f70-4e54-b4d7-fecb98f6ddd3"
      },
      "cell_type": "code",
      "source": [
        "np.argmax(model.predict(test_images_norm), axis=1)"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9, 2, 1, ..., 8, 1, 5])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "metadata": {
        "id": "ofd5hyNa9R9A",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import confusion_matrix"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "JZstVywq-BW4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 187
        },
        "outputId": "03423aeb-0d1d-4099-ec18-286d2eac7c55"
      },
      "cell_type": "code",
      "source": [
        "confusion_matrix(test_labels, np.argmax(model.predict(test_images_norm), axis=1))"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[816,   4,   9,  47,   7,   1, 103,   0,  13,   0],\n",
              "       [  1, 967,   0,  27,   2,   0,   3,   0,   0,   0],\n",
              "       [ 17,   2, 757,  24, 145,   0,  55,   0,   0,   0],\n",
              "       [ 17,   5,   7, 933,  17,   0,  17,   0,   4,   0],\n",
              "       [  1,   1,  73,  63, 828,   0,  34,   0,   0,   0],\n",
              "       [  0,   0,   0,   1,   0, 967,   0,  17,   1,  14],\n",
              "       [107,   0,  79,  52, 106,   0, 644,   0,  12,   0],\n",
              "       [  0,   0,   0,   0,   0,  30,   0, 931,   0,  39],\n",
              "       [  5,   0,   2,   6,   4,   2,   0,   3, 978,   0],\n",
              "       [  0,   0,   0,   0,   0,   7,   1,  28,   0, 964]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "metadata": {
        "id": "UpFi5No3-jQW",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "cm = confusion_matrix(test_labels, np.argmax(model.predict(test_images_norm), axis=1))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "OgTgBfx-_FGi",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import seaborn as sns; sns.set()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "JeirhjJBABnA",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "fashion_mnist_labels = [\"T-shirt/top\",  # index 0\n",
        "                        \"Trouser\",      # index 1\n",
        "                        \"Pullover\",     # index 2 \n",
        "                        \"Dress\",        # index 3 \n",
        "                        \"Coat\",         # index 4\n",
        "                        \"Sandal\",       # index 5\n",
        "                        \"Shirt\",        # index 6 \n",
        "                        \"Sneaker\",      # index 7 \n",
        "                        \"Bag\",          # index 8 \n",
        "                        \"Ankle boot\"]   # index 9"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "DrPQOvXQAhR8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 401
        },
        "outputId": "fbc7d98a-9552-4fa3-f34d-dd4963f993f4"
      },
      "cell_type": "code",
      "source": [
        "confusion_matrix = sns.heatmap(cm, annot=True, fmt=\"d\", yticklabels=fashion_mnist_labels, xticklabels=fashion_mnist_labels)\n",
        "confusion_matrix.set_xticklabels(confusion_matrix.get_xticklabels(), rotation = 90)\n",
        "\n",
        "confusion_matrix"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f937cc6c3c8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeYAAAFvCAYAAACfLGjfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nojALgiAIgoyIwiwIgiAIMvJ/BZjDbZscrUkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7f937cc4aac8>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "A1ahOzo1CR6c",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}