2/cnn_0_99_.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "_rUTmC8HpkOU"
   },
   "outputs": [],
   "source": [
    "# https://colab.research.google.com/drive/1CpV4osCtZdmtnDStA4qrYjIAlWnTjtVK?usp=sharing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "HcT3ANdtXmjk"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "\n",
    "np.random.seed(2)\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import itertools\n",
    "\n",
    "from keras.utils.np_utils import to_categorical # convert to one-hot-encoding\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten, Conv2D, MaxPool2D\n",
    "from keras.optimizers import RMSprop\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.callbacks import ReduceLROnPlateau"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "GSboKOp-VO7n",
    "outputId": "73276950-9543-4fd4-b6f5-3d88169eabcc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mounted at /content/drive\n"
     ]
    }
   ],
   "source": [
    "from google.colab import drive\n",
    "\n",
    "drive.mount('/content/drive')\n",
    "\n",
    "!cp /content/drive/MyDrive/Abdulla_OCR/Character/anotation_28x28/Test.csv .\n",
    "!cp /content/drive/MyDrive/Abdulla_OCR/Character/anotation_28x28/Train.csv .\n",
    "\n",
    "!rm -rf sample_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "enMBFPZLVRMM"
   },
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import seaborn as sns\n",
    "\n",
    "train = pd.read_csv(\"Train.csv\")\n",
    "test = pd.read_csv(\"Test.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "1pmGo6loWRIe",
    "outputId": "35287df4-d028-4305-b728-d58b68994e61"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     6000\n",
       "26    5999\n",
       "20    5999\n",
       "21    5999\n",
       "22    5999\n",
       "23    5999\n",
       "24    5999\n",
       "25    5999\n",
       "27    5999\n",
       "18    5999\n",
       "28    5999\n",
       "29    5999\n",
       "30    5999\n",
       "31    5999\n",
       "32    5999\n",
       "33    5999\n",
       "19    5999\n",
       "17    5999\n",
       "1     5999\n",
       "8     5999\n",
       "2     5999\n",
       "3     5999\n",
       "4     5999\n",
       "5     5999\n",
       "6     5999\n",
       "7     5999\n",
       "9     5999\n",
       "16    5999\n",
       "10    5999\n",
       "11    5999\n",
       "12    5999\n",
       "13    5999\n",
       "14    5999\n",
       "15    5999\n",
       "34    5999\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Y_train = train[\"label\"]\n",
    "\n",
    "# Drop 'label' column\n",
    "X_train = train.drop(labels = [\"label\"],axis = 1) \n",
    "\n",
    "# free some space\n",
    "del train \n",
    "\n",
    "g = sns.countplot(Y_train)\n",
    "\n",
    "Y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6Xj_yM6YWOUM",
    "outputId": "7b3dc3d4-1311-43cd-f0ef-5a7680182f35"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count       784\n",
       "unique        1\n",
       "top       False\n",
       "freq        784\n",
       "dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check the data\n",
    "X_train.isnull().any().describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "sliD4w8nWM0P",
    "outputId": "8a518393-f269-4023-f41f-3632256084c9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count       784\n",
       "unique        1\n",
       "top       False\n",
       "freq        784\n",
       "dtype: object"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.isnull().any().describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "id": "XUcDWzR8WLM9"
   },
   "outputs": [],
   "source": [
    "# Normalize the data\n",
    "X_train = X_train / 255.0\n",
    "test = test / 255.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "CDME31IbWJhk"
   },
   "outputs": [],
   "source": [
    "# Reshape image in 3 dimensions (height = 28px, width = 28px , canal = 1)\n",
    "X_train = X_train.values.reshape(-1,28,28,1)\n",
    "test = test.values.reshape(-1,28,28,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "LQiqgLHoWIoG"
   },
   "outputs": [],
   "source": [
    "from keras.utils import to_categorical\n",
    "\n",
    "# Encode labels to one hot vectors (ex : 2 -> [0,0,1,0,0,0,0,0,0,0])\n",
    "Y_train = to_categorical(Y_train, num_classes = 35)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "VRDDgdpQWD8f"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Split the train and the validation set for the fitting\n",
    "X_train, X_val, Y_train, Y_val = train_test_split(X_train, Y_train, test_size = 0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "nGY-GLhYVbYP"
   },
   "outputs": [],
   "source": [
    "from keras.utils import to_categorical\n",
    "\n",
    "y_train = to_categorical(Y_train, num_classes = 35, dtype = 'float32')\n",
    "y_test = to_categorical(Y_val, num_classes = 35, dtype = 'float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4M6qt1zIVfLL",
    "outputId": "8d0597d2-b4a4-49a6-8331-ee5e8c7d061e"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(188969, 35, 35)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "caVVu9RgVh5-"
   },
   "outputs": [],
   "source": [
    "# Converting the arrays to float so that we can get decimal points after division\n",
    "x_train = X_train.astype('float32')\n",
    "x_test = X_val.astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "WVRBSoPZV0XP"
   },
   "outputs": [],
   "source": [
    "# Normalizing the RGB codes by dividing it to the maximum RGB value.\n",
    "x_train /= 255\n",
    "x_test /= 255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "L_NjSAOXV18y",
    "outputId": "8012bb7c-9e69-4248-abd7-f459ac66452a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: (188969, 28, 28, 1)\n",
      "y_train shape: (188969, 35, 35)\n",
      "Number of images in x_train 188969\n",
      "Number of images in x_test 20997\n"
     ]
    }
   ],
   "source": [
    "#Checking the shapes of x_train and y_train \n",
    "print('x_train shape:', x_train.shape)\n",
    "print('y_train shape:', y_train.shape)\n",
    "print('Number of images in x_train', x_train.shape[0])\n",
    "print('Number of images in x_test', x_test.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "lUGM8CYEV32s"
   },
   "outputs": [],
   "source": [
    "#Initializing the input shape\n",
    "input_shape = (28, 28, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HU0hRCWXUWb-",
    "outputId": "a2cb3526-6c3b-43e5-d88d-3f877c40388c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/30\n",
      "5906/5906 [==============================] - 44s 5ms/step - loss: 0.4242 - accuracy: 0.8636\n",
      "Epoch 2/30\n",
      "5906/5906 [==============================] - 27s 5ms/step - loss: 0.1967 - accuracy: 0.9376\n",
      "Epoch 3/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1547 - accuracy: 0.9517\n",
      "Epoch 4/30\n",
      "5906/5906 [==============================] - 26s 4ms/step - loss: 0.1343 - accuracy: 0.9590\n",
      "Epoch 5/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1201 - accuracy: 0.9640\n",
      "Epoch 6/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1108 - accuracy: 0.9672\n",
      "Epoch 7/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1112 - accuracy: 0.9693\n",
      "Epoch 8/30\n",
      "5906/5906 [==============================] - 26s 4ms/step - loss: 0.1086 - accuracy: 0.9705\n",
      "Epoch 9/30\n",
      "5906/5906 [==============================] - 26s 4ms/step - loss: 0.1068 - accuracy: 0.9722\n",
      "Epoch 10/30\n",
      "5906/5906 [==============================] - 26s 4ms/step - loss: 0.1084 - accuracy: 0.9735\n",
      "Epoch 11/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1074 - accuracy: 0.9749\n",
      "Epoch 12/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1146 - accuracy: 0.9747\n",
      "Epoch 13/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1130 - accuracy: 0.9761\n",
      "Epoch 14/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1109 - accuracy: 0.9775\n",
      "Epoch 15/30\n",
      "5906/5906 [==============================] - 26s 4ms/step - loss: 0.1213 - accuracy: 0.9770\n",
      "Epoch 16/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1140 - accuracy: 0.9788\n",
      "Epoch 17/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1194 - accuracy: 0.9786\n",
      "Epoch 18/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1280 - accuracy: 0.9788\n",
      "Epoch 19/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1239 - accuracy: 0.9799\n",
      "Epoch 20/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1333 - accuracy: 0.9801\n",
      "Epoch 21/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1326 - accuracy: 0.9808\n",
      "Epoch 22/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1365 - accuracy: 0.9815\n",
      "Epoch 23/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1371 - accuracy: 0.9818\n",
      "Epoch 24/30\n",
      "5906/5906 [==============================] - 26s 4ms/step - loss: 0.1404 - accuracy: 0.9823\n",
      "Epoch 25/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1418 - accuracy: 0.9831\n",
      "Epoch 26/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1536 - accuracy: 0.9829\n",
      "Epoch 27/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1553 - accuracy: 0.9833\n",
      "Epoch 28/30\n",
      "5906/5906 [==============================] - 24s 4ms/step - loss: 0.1502 - accuracy: 0.9840\n",
      "Epoch 29/30\n",
      "5906/5906 [==============================] - 24s 4ms/step - loss: 0.1556 - accuracy: 0.9844\n",
      "Epoch 30/30\n",
      "5906/5906 [==============================] - 25s 4ms/step - loss: 0.1586 - accuracy: 0.9849\n",
      "657/657 [==============================] - 3s 4ms/step - loss: 0.8927 - accuracy: 0.9602\n",
      "5906/5906 [==============================] - 19s 3ms/step - loss: 0.0464 - accuracy: 0.9942\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.046373460441827774, 0.9941948056221008]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "\n",
    "# load the data\n",
    "X_train, X_val, Y_train, Y_val\n",
    "\n",
    "# define the model architecture\n",
    "model = keras.Sequential([\n",
    "    keras.layers.Conv2D(32, (5, 5), padding=\"same\", input_shape=[28, 28, 1]),\n",
    "    keras.layers.MaxPool2D((2,2)),\n",
    "    keras.layers.Conv2D(64, (5, 5), padding=\"same\"),\n",
    "    keras.layers.MaxPool2D((2,2)),\n",
    "    keras.layers.Flatten(),\n",
    "    keras.layers.Dense(1024, activation='relu'),\n",
    "    keras.layers.Dropout(0.2),\n",
    "    keras.layers.Dense(35, activation='softmax')\n",
    "])\n",
    "model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# train the model\n",
    "model.fit(x=X_train,y=Y_train, epochs=30)\n",
    "model.evaluate(X_val, Y_val)\n",
    "model.evaluate(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "lDmeKhy1UzMQ"
   },
   "outputs": [],
   "source": [
    "model.save(\"model.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "3yqQQEC7whya",
    "outputId": "7da6545e-bd8c-4bb5-f0d2-63201111d023"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 28, 28, 32)        832       \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 14, 14, 32)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 14, 14, 64)        51264     \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 7, 7, 64)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 3136)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 1024)              3212288   \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 1024)              0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 35)                35875     \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 3,300,259\n",
      "Trainable params: 3,300,259\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Om38rRpj4U34",
    "outputId": "1c95c301-f52e-4478-e2c8-d46ab08c35a6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5906/5906 [==============================] - 20s 3ms/step - loss: 0.0464 - accuracy: 0.9942\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.046373460441827774, 0.9941948056221008]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.evaluate(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "R8Ncx-PcZE56"
   },
   "outputs": [],
   "source": [
    "!cp -r model.h5 /content/drive/MyDrive/Model/"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "provenance": []
  },
  "gpuClass": "standard",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}