handwritten_reco.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyPfVtGAN2dimBJSSZdvW9Za",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/anish1301/bharat_intern_project2/blob/main/handwritten_reco.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "R1MWB0_KC7dW",
        "outputId": "d9712c82-9c4c-4c01-e6d0-a93b445c1978"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/5\n",
            "1875/1875 [==============================] - 19s 9ms/step - loss: 0.2937 - accuracy: 0.9154\n",
            "Epoch 2/5\n",
            "1875/1875 [==============================] - 8s 4ms/step - loss: 0.1420 - accuracy: 0.9574\n",
            "Epoch 3/5\n",
            "1875/1875 [==============================] - 8s 4ms/step - loss: 0.1071 - accuracy: 0.9669\n",
            "Epoch 4/5\n",
            "1875/1875 [==============================] - 8s 4ms/step - loss: 0.0882 - accuracy: 0.9725\n",
            "Epoch 5/5\n",
            "1875/1875 [==============================] - 7s 4ms/step - loss: 0.0753 - accuracy: 0.9761\n",
            "313/313 [==============================] - 1s 2ms/step - loss: 0.0679 - accuracy: 0.9797\n",
            "Test accuracy: 0.9797000288963318\n",
            "313/313 [==============================] - 1s 2ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x1000 with 25 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Import necessary libraries\n",
        "import tensorflow as tf\n",
        "from tensorflow import keras\n",
        "from tensorflow.keras.datasets import mnist\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Load and preprocess the dataset\n",
        "(train_images, train_labels), (test_images, test_labels) = mnist.load_data()\n",
        "\n",
        "# Normalize pixel values to be between 0 and 1\n",
        "train_images, test_images = train_images / 255.0, test_images / 255.0\n",
        "\n",
        "# Build the neural network model\n",
        "model = keras.Sequential([\n",
        "    keras.layers.Flatten(input_shape=(28, 28)),\n",
        "    keras.layers.Dense(128, activation='relu'),\n",
        "    keras.layers.Dropout(0.2),\n",
        "    keras.layers.Dense(10, activation='softmax')\n",
        "])\n",
        "\n",
        "# Compile the model\n",
        "model.compile(optimizer='adam',\n",
        "              loss='sparse_categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "# Train the model\n",
        "model.fit(train_images, train_labels, epochs=5)\n",
        "\n",
        "# Evaluate the model on the test data\n",
        "test_loss, test_acc = model.evaluate(test_images, test_labels)\n",
        "print(f\"Test accuracy: {test_acc}\")\n",
        "\n",
        "# Make predictions\n",
        "predictions = model.predict(test_images)\n",
        "\n",
        "# Show some predictions and their actual labels\n",
        "plt.figure(figsize=(10, 10))\n",
        "for i in range(25):\n",
        "    plt.subplot(5, 5, i+1)\n",
        "    plt.xticks([])\n",
        "    plt.yticks([])\n",
        "    plt.imshow(test_images[i], cmap=plt.cm.binary)\n",
        "    predicted_label = tf.argmax(predictions[i])\n",
        "    actual_label = test_labels[i]\n",
        "    if predicted_label == actual_label:\n",
        "        color = 'green'\n",
        "    else:\n",
        "        color = 'red'\n",
        "    plt.xlabel(f\"Predicted: {predicted_label}\\nActual: {actual_label}\", color=color)\n",
        "plt.show()\n"
      ]
    }
  ]
}