mnist_number_prediction.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "lab78_Eremin_Alexander.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "weP6bdHFPz5t"
      },
      "source": [
        "import  os\n",
        "import  numpy as np\n",
        "import  pandas  as pd\n",
        "%matplotlib inline\n",
        "import  matplotlib.pyplot as plt\n",
        "import gzip\n",
        "\n",
        "\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout, Activation, Flatten\n",
        "from keras.layers import Conv2D, MaxPooling2D\n",
        "from keras.utils import np_utils\n",
        "from keras.callbacks import EarlyStopping\n",
        "\n",
        "from sklearn.metrics import plot_confusion_matrix\n",
        "from sklearn.metrics import mean_absolute_error\n",
        "from sklearn.metrics import mean_squared_error\n",
        "from sklearn.metrics import r2_score\n",
        "from sklearn.metrics import plot_confusion_matrix\n",
        "import math"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Tg7qJPhI6gz6",
        "outputId": "41a1059e-4234-40c7-89cf-e9610a30ecdb"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZeBiR1XRP65R"
      },
      "source": [
        "# 0. Загрузка дата сета\n",
        "\n",
        "\n",
        "---\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5nB_qrUCGd5T"
      },
      "source": [
        "def load_mnist(path, kind='train'):\n",
        "    import os\n",
        "    import gzip\n",
        "    import numpy as np\n",
        "\n",
        "    labels_path = os.path.join(path,\n",
        "                               f'{kind}-labels-idx1-ubyte.gz')\n",
        "    images_path = os.path.join(path,\n",
        "                               f'{kind}-images-idx3-ubyte.gz')\n",
        "\n",
        "    with gzip.open(labels_path, 'rb') as lbpath:\n",
        "        labels = np.frombuffer(lbpath.read(), dtype=np.uint8,\n",
        "                               offset=8)\n",
        "\n",
        "    with gzip.open(images_path, 'rb') as imgpath:\n",
        "        images = np.frombuffer(imgpath.read(), dtype=np.uint8,\n",
        "                               offset=16).reshape(len(labels), 28, 28)\n",
        "    \n",
        "    return images, labels"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-ou2vm-xFPXV"
      },
      "source": [
        "test_img_path = \"/content/drive/My Drive/ML/lab7-8/t10k-images-idx3-ubyte.gz\"\n",
        "test_label_path = \"/content/drive/My Drive/ML/lab7-8/t10k-labels-idx1-ubyte.gz\"\n",
        "train_img_path =  \"/content/drive/My Drive/ML/lab7-8/train-images-idx3-ubyte.gz\"\n",
        "train_label_path = \"/content/drive/My Drive/ML/lab7-8/train-labels-idx1-ubyte.gz\"\n",
        "\n",
        "path = '/content/drive/My Drive/ML/lab7-8/'\n",
        "X_train, Y_train = load_mnist(path, kind='train')\n",
        "X_test, Y_test = load_mnist(path, kind='t10k')\n",
        "# X_test = readLabels(test_img_path)\n",
        "# Y_test = readLabels(test_label_path)\n",
        "# X_train = readLabels(train_img_path)\n",
        "# Y_train = readLabels(train_label_path)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RwmZQVHXFTy5",
        "outputId": "0f28c3d8-d220-4b32-9f8e-b244a76cf2f5"
      },
      "source": [
        "X_train.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(60000, 28, 28)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1aMQmC1nHc5x",
        "outputId": "80e4f15b-9592-41d0-d95d-5040010055b9"
      },
      "source": [
        "Y_train"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([5, 0, 4, ..., 5, 6, 8], dtype=uint8)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 223
        },
        "id": "W1e03OR8HC4i",
        "outputId": "f311e231-c476-40a2-ae0c-a6e38819a9af"
      },
      "source": [
        "\n",
        "labels = [0,1,2,3,4,5,6,7,8,9]\n",
        "\n",
        "for i, label in enumerate(labels):\n",
        "    plt.subplot(2, 5, i+1)\n",
        "    plt.imshow(X_train[np.min(np.where(Y_train == i))], \n",
        "               cmap='gray', \n",
        "               interpolation='bilinear')\n",
        "    plt.title(label)\n",
        "    plt.axis('off')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 10 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PBryvPO5JSbM"
      },
      "source": [
        "# 2. Построение нейронной сети"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PVn2fbYlHn2t",
        "outputId": "3ff1510f-8b2d-4f9a-97c7-4d0b4e561fc3"
      },
      "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",
        "X_train = X_train.astype('float32')\n",
        "X_test = X_test.astype('float32')\n",
        "X_train /= 255\n",
        "X_test /= 255\n",
        "print(\"Training matrix shape\", X_train.shape)\n",
        "print(\"Testing matrix shape\", X_test.shape)\n",
        "input_shape = (28, 28, 1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training matrix shape (60000, 28, 28, 1)\n",
            "Testing matrix shape (10000, 28, 28, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2p8kcnFjHqE2"
      },
      "source": [
        "n_classes = 10\n",
        "\n",
        "Y_train = np_utils.to_categorical(Y_train, n_classes)\n",
        "Y_test = np_utils.to_categorical(Y_test, n_classes)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vsR35l6-Hur_"
      },
      "source": [
        "model = Sequential()\n",
        "model.add(Conv2D(256, kernel_size=(3, 3),\n",
        "                 activation='relu',\n",
        "                 input_shape=input_shape))\n",
        "model.add(Conv2D(256, (3, 3), activation='relu'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(512, activation='relu'))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(256, activation='relu'))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(n_classes, activation='softmax'))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5DI5RH1QJr69"
      },
      "source": [
        "# 3. Обучение и построение метрик ошибки"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pPcLbHjWJ0bA"
      },
      "source": [
        "model.compile(loss='categorical_crossentropy', \n",
        "              optimizer='adam',\n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UP83r0V9JtdL"
      },
      "source": [
        "def root_mean_squared_error(y_true, y_pred):\n",
        "        return K.sqrt(K.mean(K.square(y_pred - y_true))) "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NEIQ_pBYIaky"
      },
      "source": [
        "model.compile(optimizer = \"rmsprop\", loss = root_mean_squared_error, \n",
        "              metrics =[\"accuracy\"])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EsO7kjyDJ-41"
      },
      "source": [
        "early_stopping_monitor = EarlyStopping(patience=3)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-f_WgTCkKAmp",
        "outputId": "31e99a40-f54c-4059-f0fb-9e48115ea3df"
      },
      "source": [
        "model_train = model.fit(X_train, Y_train,\n",
        "                        batch_size=128, \n",
        "                        epochs=3,\n",
        "                        callbacks=[early_stopping_monitor],\n",
        "                        validation_data=(X_test, Y_test))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/3\n",
            "469/469 [==============================] - 50s 92ms/step - loss: 0.2999 - accuracy: 0.9008 - val_loss: 0.0344 - val_accuracy: 0.9884\n",
            "Epoch 2/3\n",
            "469/469 [==============================] - 42s 90ms/step - loss: 0.0430 - accuracy: 0.9866 - val_loss: 0.0298 - val_accuracy: 0.9906\n",
            "Epoch 3/3\n",
            "469/469 [==============================] - 42s 90ms/step - loss: 0.0261 - accuracy: 0.9916 - val_loss: 0.0421 - val_accuracy: 0.9875\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "H0hw0vgEMxvJ",
        "outputId": "e49c98e0-0bd3-4cf3-b92d-644028315ca8"
      },
      "source": [
        "model_train.history"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': [0.9578166604042053, 0.9864333271980286, 0.9917166829109192],\n",
              " 'loss': [0.13365347683429718, 0.042712002992630005, 0.02593539096415043],\n",
              " 'val_accuracy': [0.9883999824523926, 0.9905999898910522, 0.987500011920929],\n",
              " 'val_loss': [0.034369904547929764, 0.02982674166560173, 0.04208217188715935]}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 607
        },
        "id": "-V2UIH9jLZtE",
        "outputId": "a108196c-115a-46ab-b654-1a4da5a57c22"
      },
      "source": [
        "plt.figure(figsize=(20, 10))\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(model_train.history['accuracy'], label='train')\n",
        "plt.plot(model_train.history['val_accuracy'], label='test')\n",
        "plt.title('Accuracy')\n",
        "plt.legend(loc='center right')\n",
        "plt.ylim(0.9, )\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(model_train.history['loss'], label='train')\n",
        "plt.plot(model_train.history['val_loss'], label='test')\n",
        "plt.title('Loss')\n",
        "plt.legend(loc='center right');"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "B3WPtqBxKlsP",
        "outputId": "02342a92-19f6-4534-c48b-f868dafb20be"
      },
      "source": [
        "score = model.evaluate(X_test, Y_test, verbose=0)\n",
        "print('Test score:', score[0])\n",
        "print('Test accuracy:', score[1])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Test score: 0.042082175612449646\n",
            "Test accuracy: 0.987500011920929\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KWOv2tsdKjfZ",
        "outputId": "77a9c09e-76a5-466e-abb0-3ff9929e0c23"
      },
      "source": [
        "Y_pred = model.predict_classes(X_test, verbose=0)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/sequential.py:450: UserWarning: `model.predict_classes()` is deprecated and will be removed after 2021-01-01. Please use instead:* `np.argmax(model.predict(x), axis=-1)`,   if your model does multi-class classification   (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype(\"int32\")`,   if your model does binary classification   (e.g. if it uses a `sigmoid` last-layer activation).\n",
            "  warnings.warn('`model.predict_classes()` is deprecated and '\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KsEt-dACNyn5"
      },
      "source": [
        "testing_labels = []\n",
        "for lbl in Y_test:\n",
        "  for i, val in enumerate(lbl):\n",
        "    if val == 1:\n",
        "      testing_labels.append(i)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 712
        },
        "id": "iBMoDJYMKJJ9",
        "outputId": "82ede6d4-6fca-4b6e-abe5-74b971aa3634"
      },
      "source": [
        "def check_error(Y_pred, Y_test):\n",
        "  sum = 0.0\n",
        "  max_deviation = 0\n",
        "  min_deviation = 10**10\n",
        "\n",
        "  for i, elem in enumerate(Y_test):\n",
        "    sum += math.fabs(Y_pred[i]-elem)\n",
        "    if math.fabs(Y_pred[i]-elem) > max_deviation:\n",
        "      max_deviation = math.fabs(Y_pred[i]-elem)\n",
        "    if math.fabs(Y_pred[i] - elem) < min_deviation:\n",
        "      min_deviation = math.fabs(Y_pred[i]-elem)\n",
        "\n",
        "  print(\"Max_deviation: \", max_deviation)\n",
        "  print(\"Min_deviation: \", min_deviation)\n",
        "  print(\"Arithmetic_mean: \", sum/len(Y_pred))\n",
        "  print(\"MAE: \", mean_absolute_error(Y_test, Y_pred))\n",
        "  print(\"RMSE: \", mean_squared_error(Y_test, Y_pred))\n",
        "  print(\"R^2: \", r2_score(Y_test, Y_pred))\n",
        "  plt.figure(figsize=(10, 10))\n",
        "  plt.scatter(Y_test, Y_pred)\n",
        "  plt.plot([0, max(Y_test)], [0, max(Y_pred)])\n",
        "  plt.xlabel('Actual', fontsize=14)\n",
        "  plt.ylabel('Prediction', fontsize=14)\n",
        "\n",
        "check_error(Y_pred, testing_labels)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Max_deviation:  9.0\n",
            "Min_deviation:  0.0\n",
            "Arithmetic_mean:  0.0461\n",
            "MAE:  0.0461\n",
            "RMSE:  0.2247\n",
            "R^2:  0.9732027744838379\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}