Devnagri_Script_Detection_CNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "language": "python",
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "version": "3.6.4",
      "file_extension": ".py",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "name": "python",
      "mimetype": "text/x-python"
    },
    "colab": {
      "name": "Devnagri Script Detection CNN.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Yash0411/Machine-Learning-initial-projects/blob/master/Devnagri_Script_Detection_CNN.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
        "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
        "trusted": true,
        "id": "xlWru4z4U_hj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\n",
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.optimizers import RMSprop\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "# Input data files are available in the read-only \"../input/\" directory\n",
        "# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RtWw_IXyVC7F",
        "colab_type": "code",
        "outputId": "14d2b055-6125-4e45-d776-2811a767d990",
        "colab": {
          "resources": {
            "http://localhost:8080/nbextensions/google.colab/files.js": {
              "data": "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",
              "ok": true,
              "headers": [
                [
                  "content-type",
                  "application/javascript"
                ]
              ],
              "status": 200,
              "status_text": "OK"
            }
          },
          "base_uri": "https://localhost:8080/",
          "height": 95
        }
      },
      "source": [
        "from google.colab import files\n",
        "files.upload() #upload kaggle.json\n",
        " \n",
        "!pip install -q kaggle\n",
        "!mkdir -p ~/.kaggle\n",
        "!cp kaggle.json ~/.kaggle/\n",
        "!ls ~/.kaggle\n",
        "!chmod 600 /root/.kaggle/kaggle.json"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "     <input type=\"file\" id=\"files-68d3c432-defb-4ed4-b5b1-8ae5bc958801\" name=\"files[]\" multiple disabled />\n",
              "     <output id=\"result-68d3c432-defb-4ed4-b5b1-8ae5bc958801\">\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 object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Saving kaggle.json to kaggle (1).json\n",
            "kaggle.json\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_VZdCdZZXuwR",
        "colab_type": "code",
        "outputId": "09dd9b87-b0d2-4f0e-c6f1-c6b06b3d3844",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "!kaggle datasets download -d somnath796/devnagri-handwritten-character"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "devnagri-handwritten-character.zip: Skipping, found more recently modified local copy (use --force to force download)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "18VhuMxvVZe6",
        "colab_type": "code",
        "outputId": "e39234fb-c567-45a5-8bf5-80ed508c7a72",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 46
        }
      },
      "source": [
        "!unzip -q /content/devnagri-handwritten-character.zip -d .\n",
        "!ls"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "replace ./DEVNAGARI_NEW/TEST/1/001_01.jpg? [y]es, [n]o, [A]ll, [N]one, [r]ename: "
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "QuJGCUgbU_iC",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "8642aab8-0ff6-499e-8dcf-c4b9ee834a2e"
      },
      "source": [
        "import os\n",
        "\n",
        "training_dir = os.listdir('/content/DEVNAGARI_NEW/TRAIN/')\n",
        "testing_dir = os.listdir('/content/DEVNAGARI_NEW/TEST/')\n",
        "\n",
        "print(len(training_dir))\n",
        "print(len(testing_dir))\n",
        "\n"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "48\n",
            "48\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "EtMG30pDU_i8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "d0ecd535-2087-4f4e-8798-ecde32fb5d66"
      },
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Conv2D\n",
        "from keras.layers import MaxPooling2D\n",
        "from keras.layers import Flatten\n",
        "from keras.layers import Dense\n",
        "\n",
        "classifier = Sequential()\n",
        "\n",
        "\n",
        "classifier.add(Conv2D(32, (3, 3), input_shape = (64, 64, 3), activation = 'relu'))\n",
        "classifier.add(MaxPooling2D(pool_size = (2, 2)))\n",
        "\n",
        "classifier.add(Conv2D(64, (3, 3), activation = 'relu'))\n",
        "classifier.add(MaxPooling2D(pool_size = (2, 2)))\n",
        "\n",
        "classifier.add(Conv2D(64, (3, 3), activation = 'relu'))\n",
        "classifier.add(MaxPooling2D(pool_size = (2, 2)))\n",
        "\n",
        "classifier.add(Flatten())\n",
        "\n",
        "classifier.add(Dense(units=64, activation='relu'))\n",
        "classifier.add(Dense(units=128, activation='relu'))\n",
        "classifier.add(Dense(units = 48, activation = 'softmax'))\n",
        "\n",
        "classifier.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics = ['accuracy'])"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "0k67JcjaU_jM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "outputId": "9cb27dc8-82a2-4c07-801e-1815a036e082"
      },
      "source": [
        "TRAINING_DIR = \"/content/DEVNAGARI_NEW/TRAIN\"\n",
        "train_datagen = ImageDataGenerator( rescale=1./255.)\n",
        "\n",
        "training_set = train_datagen.flow_from_directory(TRAINING_DIR,\n",
        "                                                 target_size = (64, 64),\n",
        "                                                 batch_size = 32,\n",
        "                                                 class_mode = 'categorical')\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "VALIDATION_DIR = \"/content/DEVNAGARI_NEW/TEST\"\n",
        "validation_datagen = ImageDataGenerator(rescale = 1.0 / 255)\n",
        "\n",
        "test_set = validation_datagen.flow_from_directory(VALIDATION_DIR,\n",
        "                                            target_size = (64, 64),\n",
        "                                            batch_size = 32,\n",
        "                                            class_mode = 'categorical')"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 6528 images belonging to 48 classes.\n",
            "Found 3312 images belonging to 48 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "VsZPqE3-U_jc",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "3fec2164-cf40-4671-a15b-5f7b00424dfd"
      },
      "source": [
        "history = classifier.fit_generator(training_set,\n",
        "                         epochs = 32,\n",
        "                         validation_data = test_set)\n",
        "\n"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/32\n",
            "204/204 [==============================] - 5s 25ms/step - loss: 3.1872 - accuracy: 0.1702 - val_loss: 2.5264 - val_accuracy: 0.4460\n",
            "Epoch 2/32\n",
            "204/204 [==============================] - 3s 17ms/step - loss: 1.2663 - accuracy: 0.6186 - val_loss: 0.7619 - val_accuracy: 0.6643\n",
            "Epoch 3/32\n",
            "204/204 [==============================] - 3s 17ms/step - loss: 0.7419 - accuracy: 0.7672 - val_loss: 1.2451 - val_accuracy: 0.7307\n",
            "Epoch 4/32\n",
            "204/204 [==============================] - 3s 17ms/step - loss: 0.5220 - accuracy: 0.8346 - val_loss: 0.7959 - val_accuracy: 0.7509\n",
            "Epoch 5/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.3843 - accuracy: 0.8753 - val_loss: 0.4314 - val_accuracy: 0.7533\n",
            "Epoch 6/32\n",
            "204/204 [==============================] - 3s 17ms/step - loss: 0.2809 - accuracy: 0.9082 - val_loss: 1.3176 - val_accuracy: 0.7409\n",
            "Epoch 7/32\n",
            "204/204 [==============================] - 3s 17ms/step - loss: 0.2199 - accuracy: 0.9248 - val_loss: 0.7845 - val_accuracy: 0.7557\n",
            "Epoch 8/32\n",
            "204/204 [==============================] - 3s 17ms/step - loss: 0.1825 - accuracy: 0.9367 - val_loss: 0.8679 - val_accuracy: 0.7732\n",
            "Epoch 9/32\n",
            "204/204 [==============================] - 3s 17ms/step - loss: 0.1349 - accuracy: 0.9519 - val_loss: 0.4317 - val_accuracy: 0.7920\n",
            "Epoch 10/32\n",
            "204/204 [==============================] - 3s 17ms/step - loss: 0.1069 - accuracy: 0.9632 - val_loss: 1.5792 - val_accuracy: 0.7355\n",
            "Epoch 11/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.1155 - accuracy: 0.9614 - val_loss: 0.9594 - val_accuracy: 0.7823\n",
            "Epoch 12/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0848 - accuracy: 0.9691 - val_loss: 0.3488 - val_accuracy: 0.7334\n",
            "Epoch 13/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0783 - accuracy: 0.9758 - val_loss: 0.5740 - val_accuracy: 0.7636\n",
            "Epoch 14/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0593 - accuracy: 0.9782 - val_loss: 0.8212 - val_accuracy: 0.7796\n",
            "Epoch 15/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0950 - accuracy: 0.9669 - val_loss: 0.6217 - val_accuracy: 0.7775\n",
            "Epoch 16/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0577 - accuracy: 0.9816 - val_loss: 0.7678 - val_accuracy: 0.7802\n",
            "Epoch 17/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0687 - accuracy: 0.9756 - val_loss: 0.3443 - val_accuracy: 0.7651\n",
            "Epoch 18/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0494 - accuracy: 0.9838 - val_loss: 1.4957 - val_accuracy: 0.7751\n",
            "Epoch 19/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0632 - accuracy: 0.9779 - val_loss: 0.5393 - val_accuracy: 0.7693\n",
            "Epoch 20/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0514 - accuracy: 0.9851 - val_loss: 1.2156 - val_accuracy: 0.8022\n",
            "Epoch 21/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0528 - accuracy: 0.9821 - val_loss: 1.1874 - val_accuracy: 0.7835\n",
            "Epoch 22/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0368 - accuracy: 0.9865 - val_loss: 0.5036 - val_accuracy: 0.7880\n",
            "Epoch 23/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0750 - accuracy: 0.9752 - val_loss: 2.0931 - val_accuracy: 0.7871\n",
            "Epoch 24/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0271 - accuracy: 0.9913 - val_loss: 0.7983 - val_accuracy: 0.8025\n",
            "Epoch 25/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0137 - accuracy: 0.9954 - val_loss: 2.4854 - val_accuracy: 0.8113\n",
            "Epoch 26/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0535 - accuracy: 0.9822 - val_loss: 0.5637 - val_accuracy: 0.7572\n",
            "Epoch 27/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0866 - accuracy: 0.9706 - val_loss: 1.4463 - val_accuracy: 0.7856\n",
            "Epoch 28/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0407 - accuracy: 0.9864 - val_loss: 1.1650 - val_accuracy: 0.7889\n",
            "Epoch 29/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0130 - accuracy: 0.9959 - val_loss: 0.6269 - val_accuracy: 0.7853\n",
            "Epoch 30/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0423 - accuracy: 0.9853 - val_loss: 0.9963 - val_accuracy: 0.7805\n",
            "Epoch 31/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0622 - accuracy: 0.9786 - val_loss: 1.3657 - val_accuracy: 0.7739\n",
            "Epoch 32/32\n",
            "204/204 [==============================] - 3s 16ms/step - loss: 0.0562 - accuracy: 0.9819 - val_loss: 0.3360 - val_accuracy: 0.8089\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "trusted": true,
        "id": "oPD5d7UPU_jm",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 563
        },
        "outputId": "d7b40ee0-ebba-446d-9408-a94cfcecfe85"
      },
      "source": [
        "# PLOT LOSS AND ACCURACY\n",
        "%matplotlib inline\n",
        "\n",
        "import matplotlib.image  as mpimg\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "#-----------------------------------------------------------\n",
        "# Retrieve a list of list results on training and test data\n",
        "# sets for each training epoch\n",
        "#-----------------------------------------------------------\n",
        "acc=history.history['accuracy']\n",
        "val_acc=history.history['val_accuracy']\n",
        "loss=history.history['loss']\n",
        "val_loss=history.history['val_loss']\n",
        "\n",
        "epochs=range(len(acc)) # Get number of epochs\n",
        "\n",
        "#------------------------------------------------\n",
        "# Plot training and validation accuracy per epoch\n",
        "#------------------------------------------------\n",
        "plt.plot(epochs, acc, 'r', \"Training Accuracy\")\n",
        "plt.plot(epochs, val_acc, 'b', \"Validation Accuracy\")\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.figure()\n",
        "\n",
        "#------------------------------------------------\n",
        "# Plot training and validation loss per epoch\n",
        "#------------------------------------------------\n",
        "plt.plot(epochs, loss, 'r', \"Training Loss\")\n",
        "plt.plot(epochs, val_loss, 'b', \"Validation Loss\")\n",
        "\n",
        "\n",
        "plt.title('Training and validation loss')\n"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Training and validation loss')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6Dcjgbb-rlX7",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}