MNIST_Neural_Net_Kaggle.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MNIST kaggle.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "K08La8Zip2SH"
      },
      "source": [
        "# Import libraries\r\n",
        "import numpy as np\r\n",
        "import pandas as pd\r\n",
        "from sklearn.metrics import accuracy_score\r\n",
        "from sklearn import preprocessing\r\n",
        "from sklearn import metrics\r\n",
        "from sklearn.linear_model import LogisticRegression\r\n",
        "from sklearn.metrics import f1_score\r\n",
        "from tensorflow import keras\r\n",
        "from tensorflow.keras.utils import to_categorical # to One-Hot-Code values\r\n",
        "from tensorflow.keras.models import Sequential\r\n",
        "from tensorflow.keras.layers import Activation, Dropout, Dense, Flatten\r\n",
        "from tensorflow.keras.layers import Conv2D, MaxPooling2D # create convolutional layer\r\n",
        "from keras.models import Model\r\n",
        "import matplotlib.pyplot as plt # plotting graphs for visual analysis\r\n",
        "from sklearn.model_selection import train_test_split #splitting dataset\r\n",
        "%matplotlib inline"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ARdOx2ywqR7c"
      },
      "source": [
        "train_df = pd.read_csv('/content/drive/MyDrive/train.csv')\r\n",
        "test_df = pd.read_csv('/content/drive/MyDrive/test.csv')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pdMdaWOxqfvm",
        "outputId": "7e90eb41-b079-4378-876a-ca48d9e0113c"
      },
      "source": [
        "print(train_df.shape)\r\n",
        "print(test_df.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(42000, 785)\n",
            "(28000, 784)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NAwV7fSC_pit",
        "outputId": "9617582f-b2dd-4c5a-95c5-80bca0b89a20"
      },
      "source": [
        "train_df.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 42000 entries, 0 to 41999\n",
            "Columns: 785 entries, label to pixel783\n",
            "dtypes: int64(785)\n",
            "memory usage: 251.5 MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O1SBYXDfqhuP"
      },
      "source": [
        "#Separating labels and pixel values\r\n",
        "X_train = train_df.drop('label',axis=1)\r\n",
        "y_train = train_df['label']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q9hX_E67-BMg",
        "outputId": "9b21a887-8f73-4095-b97b-8ab7328d3f58"
      },
      "source": [
        "print(X_train.shape)\r\n",
        "print(y_train.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(42000, 784)\n",
            "(42000,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XDK1MH9ZrcTb"
      },
      "source": [
        "#Normalizing data\r\n",
        "X_train = X_train/255.\r\n",
        "test_df = test_df/255."
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s4-Bc11V_QVC"
      },
      "source": [
        "image_train, image_val, label_train, label_val = train_test_split(X_train, y_train,\r\n",
        "                                        test_size=0.1, random_state=44)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "frDxBi3arpA3",
        "outputId": "6d4cda87-5c34-411e-bf2d-6ce2e8972ea5"
      },
      "source": [
        "image_train.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(37800, 784)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aZ-hdtPy82iP",
        "outputId": "e8d46885-a23b-4fef-8faa-44a377ee22fe"
      },
      "source": [
        "label_train.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(37800,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mvQgFTxx6X9Y",
        "outputId": "4760d65b-7ecb-465d-8ae0-f9646b4a2377"
      },
      "source": [
        "# Build and fit the Logistic Regression model\r\n",
        "log_reg = LogisticRegression( C=0.1, solver='sag', random_state=4, verbose=1)\r\n",
        "log_reg.fit(image_train, label_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "convergence after 43 epochs took 55 seconds\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:   55.1s finished\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=0.1, class_weight=None, dual=False, fit_intercept=True,\n",
              "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
              "                   multi_class='auto', n_jobs=None, penalty='l2',\n",
              "                   random_state=4, solver='sag', tol=0.0001, verbose=1,\n",
              "                   warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nkYyk0NiB3Fj"
      },
      "source": [
        "y_pred = log_reg.predict(image_val)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "75JFBOIyB6qj",
        "outputId": "e0096d5a-a41f-44cb-d8b9-c7e27995a902"
      },
      "source": [
        "# Evaluate the Model Accuracy\r\n",
        "\r\n",
        "# Comparing y test to y predict\r\n",
        "print(\"Logistic Regression Model Accuracy (after tuning solver) on test data: \", metrics.accuracy_score(label_val, y_pred))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Logistic Regression Model Accuracy (after tuning solver) on test data:  0.9207142857142857\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9jFhrfD8Ev7d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 163
        },
        "outputId": "349852a2-0865-474b-a187-29d4a7e15ef2"
      },
      "source": [
        "test_result = log_reg.predict(test_df)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-12-a5c16ae41841>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mtest_result\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlog_reg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest_df\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m: name 'log_reg' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Rl9wE_588_xr"
      },
      "source": [
        "# For NN model\r\n",
        "# reshape images into 28*28*1 (-1 takes value as it is) for NN\r\n",
        "\r\n",
        "x_train = X_train.values.reshape(-1, 28, 28, 1)\r\n",
        "tesT_df=test_df.values.reshape(-1, 28, 28, 1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4N0EEJ9XC0jG"
      },
      "source": [
        "y_cat = to_categorical(y_train)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "abU3QzFk-nID"
      },
      "source": [
        "image_train, image_val, label_train, label_val = train_test_split(x_train, y_cat,\r\n",
        "                                        test_size=0.1, random_state=44)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "COs8kq6T9cjf",
        "outputId": "ec006b6d-7a05-4354-9295-7d9324873f7b"
      },
      "source": [
        "model=Sequential()\r\n",
        "\r\n",
        "model.add(Conv2D(64, (3,3), \r\n",
        "                 activation='relu',\r\n",
        "                 input_shape=(28, 28, 1)))\r\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\r\n",
        "#model.add(Dropout(0.25))\r\n",
        "\r\n",
        "model.add(Conv2D(64, (3,3),\r\n",
        "                activation='relu',\r\n",
        "                input_shape=(28, 28, 1)))\r\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\r\n",
        "#model.add(Dropout(0.25))\r\n",
        "\r\n",
        "model.add(Conv2D(64, (3,3),\r\n",
        "                activation='relu',\r\n",
        "                input_shape=(28, 28, 1)))\r\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\r\n",
        "\r\n",
        "model.add(Dropout(0.25))\r\n",
        "\r\n",
        "#2. Create 2 Dense layers, one is hidden and other is for prediction(output)\r\n",
        "#Flatten(convert 2D to 1D array) the results for Dense Neural Network\r\n",
        "model.add(Flatten())\r\n",
        "#model.add(Dropout(0.25))\r\n",
        "\r\n",
        "#Add a dense layer to recognize and define pixel patterns\r\n",
        "model.add(Dense(256, activation='relu'))\r\n",
        "\r\n",
        "#Add another dense layer to identify and find probability of class 0-9\r\n",
        "model.add(Dense(10, activation='softmax'))\r\n",
        "\r\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_3 (Conv2D)            (None, 26, 26, 64)        640       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_3 (MaxPooling2 (None, 13, 13, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_4 (Conv2D)            (None, 11, 11, 64)        36928     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_4 (MaxPooling2 (None, 5, 5, 64)          0         \n",
            "_________________________________________________________________\n",
            "conv2d_5 (Conv2D)            (None, 3, 3, 64)          36928     \n",
            "_________________________________________________________________\n",
            "max_pooling2d_5 (MaxPooling2 (None, 1, 1, 64)          0         \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 1, 1, 64)          0         \n",
            "_________________________________________________________________\n",
            "flatten_1 (Flatten)          (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 256)               16640     \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 10)                2570      \n",
            "=================================================================\n",
            "Total params: 93,706\n",
            "Trainable params: 93,706\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e97tNtV19d7H"
      },
      "source": [
        "model.compile(loss='categorical_crossentropy',\r\n",
        "                optimizer='Adam',    #rmsprop\r\n",
        "                metrics=['accuracy'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t9x-p7qL9lNK"
      },
      "source": [
        "epochs=5\r\n",
        "batch_size=64"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gyPL4pJD9pWL",
        "outputId": "e92873ce-727f-4f39-a663-54e828c13c17"
      },
      "source": [
        "history=model.fit(image_train,label_train,\r\n",
        "                  batch_size=batch_size, epochs=epochs,\r\n",
        "                  validation_data = (image_val, label_val),\r\n",
        "                  verbose=2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/5\n",
            "591/591 - 45s - loss: 0.3876 - accuracy: 0.8798 - val_loss: 0.1657 - val_accuracy: 0.9493\n",
            "Epoch 2/5\n",
            "591/591 - 44s - loss: 0.1367 - accuracy: 0.9591 - val_loss: 0.0944 - val_accuracy: 0.9719\n",
            "Epoch 3/5\n",
            "591/591 - 44s - loss: 0.1005 - accuracy: 0.9697 - val_loss: 0.0673 - val_accuracy: 0.9802\n",
            "Epoch 4/5\n",
            "591/591 - 44s - loss: 0.0804 - accuracy: 0.9752 - val_loss: 0.0633 - val_accuracy: 0.9812\n",
            "Epoch 5/5\n",
            "591/591 - 44s - loss: 0.0691 - accuracy: 0.9790 - val_loss: 0.0632 - val_accuracy: 0.9845\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 610
        },
        "id": "Yv92yovA-6jU",
        "outputId": "a1b14010-fae4-4184-9856-81cd3962c6e0"
      },
      "source": [
        "fig, ax = plt.subplots(2,1, figsize=(15, 10))\r\n",
        "ax[0].plot(history.history['loss'], color='b', label=\"Training loss\")\r\n",
        "ax[0].plot(history.history['val_loss'], color='r', label=\"validation loss\")\r\n",
        "ax[0].legend()\r\n",
        "\r\n",
        "ax[1].plot(history.history['accuracy'], color='b', label=\"Training accuracy\")\r\n",
        "ax[1].plot(history.history['val_accuracy'], color='r',label=\"Validation accuracy\")\r\n",
        "ax[1].legend()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7fcc3458dc90>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x720 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}