MNIST.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "vFnB69XE-FVT"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.metrics import classification_report"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "(xtrain , ytrain) , (xtest , ytest) = tf.keras.datasets.mnist.load_data()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "rHh3rl6p-vXW",
        "outputId": "50b71bd1-b1ad-4294-f830-d92a9f184f41"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "\u001b[1m11490434/11490434\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "xtrain[1].shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "SYJhWRhG_ni7",
        "outputId": "b73df3b9-1e95-4b8a-d13f-691c8ac96e6e"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(28, 28)"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(xtrain[1])\n",
        "plt.title(f\"digit = {ytrain[1]}\" )"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 469
        },
        "collapsed": true,
        "id": "ExqKtE8Rs1xS",
        "outputId": "b44eaa22-72a4-41d2-bc93-1560d851e7c6"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'digit = 0')"
            ]
          },
          "metadata": {},
          "execution_count": 7
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "xtrain,xtest = xtrain/255 , xtest/255"
      ],
      "metadata": {
        "id": "nFVR_g_ns6oS"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model = tf.keras.Sequential([\n",
        "    tf.keras.Input(shape=(28,28)),\n",
        "    tf.keras.layers.Flatten(),\n",
        "    tf.keras.layers.Dense(100,activation = 'relu'),\n",
        "    tf.keras.layers.Dense(50,activation = 'relu'),\n",
        "    tf.keras.layers.Dense(10,activation = 'softmax')\n",
        "])"
      ],
      "metadata": {
        "id": "QVSO8rL9tMj2"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(optimizer = 'adam', loss = 'sparse_categorical_crossentropy', metrics = ['accuracy'])"
      ],
      "metadata": {
        "id": "O-CK4AKMtTBv"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.fit(xtrain,ytrain,epochs = 10,verbose = 1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "qo76CSCftVlj",
        "outputId": "48317eb8-2345-4ac3-8f6f-ab412b8789b7"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.9594 - loss: 0.1375\n",
            "Epoch 2/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3ms/step - accuracy: 0.9611 - loss: 0.1307\n",
            "Epoch 3/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.9632 - loss: 0.1241\n",
            "Epoch 4/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - accuracy: 0.9645 - loss: 0.1198\n",
            "Epoch 5/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - accuracy: 0.9652 - loss: 0.1165\n",
            "Epoch 6/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - accuracy: 0.9672 - loss: 0.1101\n",
            "Epoch 7/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3ms/step - accuracy: 0.9690 - loss: 0.1060\n",
            "Epoch 8/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.9691 - loss: 0.1011\n",
            "Epoch 9/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.9737 - loss: 0.0915\n",
            "Epoch 10/10\n",
            "\u001b[1m1875/1875\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3ms/step - accuracy: 0.9719 - loss: 0.0920\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.src.callbacks.history.History at 0x7e5687b36770>"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_loss, test_accuracy = model.evaluate(xtest, ytest, verbose=1)\n",
        "print(f\"test_loss = {test_loss}\")\n",
        "print(f\"test_accuracy = {test_accuracy}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "h0nuyVaZtXoC",
        "outputId": "c79fc5e5-c34e-45a8-bc74-8b1cf9e8b51c"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.9610 - loss: 0.1282\n",
            "test_loss = 0.11382211744785309\n",
            "test_accuracy = 0.9650999903678894\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#CLASSIFICATION REPORT ON TEST DATA\n",
        "y_pred = model.predict(xtest)\n",
        "y_pred_classes = y_pred.argmax(axis=1)\n",
        "print(classification_report(ytest, y_pred_classes))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "IHuMhEdBtowm",
        "outputId": "63d73af5-a764-40fb-fb4b-2ada831cc90a"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.97      0.99      0.98       980\n",
            "           1       0.98      0.99      0.99      1135\n",
            "           2       0.96      0.96      0.96      1032\n",
            "           3       0.95      0.97      0.96      1010\n",
            "           4       0.96      0.97      0.96       982\n",
            "           5       0.97      0.94      0.96       892\n",
            "           6       0.98      0.96      0.97       958\n",
            "           7       0.97      0.96      0.96      1028\n",
            "           8       0.95      0.96      0.95       974\n",
            "           9       0.96      0.95      0.96      1009\n",
            "\n",
            "    accuracy                           0.97     10000\n",
            "   macro avg       0.97      0.96      0.96     10000\n",
            "weighted avg       0.97      0.97      0.97     10000\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#INCORRECT PREDICTIONS ON TEST DATA\n",
        "misclassified_indices = [i for i, (true, pred) in enumerate(zip(ytest, y_pred_classes)) if true != pred]\n",
        "count = len(misclassified_indices)\n",
        "print(f\"Total misclassified_indices are {count}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "7eboSivYtuBQ",
        "outputId": "1f7c102a-becc-4593-f887-2453ac3cd35b"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Total misclassified_indices are 349\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "for index in misclassified_indices[:5]:\n",
        "    plt.imshow(xtest[index].reshape(28, 28), cmap='gray')\n",
        "    plt.title(f\"True: {ytest[index]}, Pred: {y_pred_classes[index]}\")\n",
        "    plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "collapsed": true,
        "id": "r6zNZKw_t76V",
        "outputId": "419fc73c-2f84-4e26-e55d-6e445ca0e128"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#predict the test data\n",
        "plt.imshow(xtest[110])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 447
        },
        "collapsed": true,
        "id": "UJYfW7QDzpPi",
        "outputId": "5a281035-0c5e-49a5-bf81-1bb2c712db37"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7e567fd981c0>"
            ]
          },
          "metadata": {},
          "execution_count": 54
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predict = np.argmax(model.predict(xtest)[110])\n",
        "print(f\"Number present in the image is {predict}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "S5OaGgOyzOpN",
        "outputId": "0c79baa4-86b0-46a1-998e-473ec724aba4"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step\n",
            "Number present in the image is 8\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#TESTING THE MODEL ON NEW DATA\n",
        "test_image = tf.keras.preprocessing.image.load_img(\"new_image_path\")\n",
        "test_image = tf.image.rgb_to_grayscale(test_image)\n",
        "test_image = tf.keras.preprocessing.image.img_to_array(test_image)"
      ],
      "metadata": {
        "id": "dPW0uhEDuIC4"
      },
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_image.shape"
      ],
      "metadata": {
        "id": "lmCCr8lYubBC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_image = tf.image.resize(test_image,[28,28])"
      ],
      "metadata": {
        "id": "Ir7CrbMOudsJ"
      },
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_image = test_image.numpy().squeeze()\n",
        "test_image.astype(\"float64\")\n",
        "test_image = test_image/255"
      ],
      "metadata": {
        "id": "mFrHm5QsuhWN"
      },
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(test_image)"
      ],
      "metadata": {
        "id": "pjG231OBvyzm"
      },
      "execution_count": null,
      "outputs": []
    },
    { "cell_type": "code",
      "source": [
        "v = np.argmax(model.predict(test_image.reshape(1,28,28)))\n",
        "print(f\"Number present in the image is {v}\")"
      ],
      "metadata": {
        "id": "8hHs4z5HuldC"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}