digit_recognition.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "digit_recognition.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyMgj7fBtwnJhBL/vDsRW8dp",
      "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/StuPidMRE/Deeplearn/blob/main/digit_recognition.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FDRD6NUyubd_",
        "outputId": "91eeba91-a036-4292-8b2b-59c9be06c077"
      },
      "source": [
        "import tensorflow as tf\n",
        "(x_train,y_train),(x_test,y_test)=tf.keras.datasets.mnist.load_data()"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 0s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JhDnR5TMtzT9"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "image_index=7777\n",
        "print(y_train[image_index])\n",
        "plt.imshow(x_train[image_index],cmap='Greys')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JWbLYrvhu84j",
        "outputId": "5b75f766-a2fd-413c-d455-a313a37f0362"
      },
      "source": [
        "x_train.shape"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(60000, 28, 28)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qZN1cztgvAJR"
      },
      "source": [
        "x_train=x_train.reshape(x_train.shape[0],28,28,1)\n",
        "x_test=x_test.reshape(x_test.shape[0],28,28,1)\n",
        "input_shape=(28,28,1)\n",
        "x_train=x_train.astype('float32')\n",
        "x_test=x_test.astype('float32')\n",
        "x_train/=255\n",
        "x_test/=255\n"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iLseAxLhvmBN"
      },
      "source": [
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Dense,Conv2D,Dropout,Flatten,MaxPooling2D\n",
        "\n",
        "model=Sequential()\n",
        "model.add(Conv2D(28,kernel_size=(3,3),input_shape=input_shape))\n",
        "model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(128,activation=tf.nn.relu))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(10,activation=tf.nn.softmax))"
      ],
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ilr5cg3nwvu8",
        "outputId": "aa748fa5-ecc9-42ac-a2dd-bd9bb1aee03a"
      },
      "source": [
        "model.compile(optimizer='adam',\n",
        "              loss='sparse_categorical_crossentropy',\n",
        "              metrics=['accuracy'])\n",
        "model.fit(x=x_train,y=y_train,epochs=10)"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "1875/1875 [==============================] - 26s 13ms/step - loss: 0.3521 - accuracy: 0.8953\n",
            "Epoch 2/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0833 - accuracy: 0.9742\n",
            "Epoch 3/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0589 - accuracy: 0.9818\n",
            "Epoch 4/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0426 - accuracy: 0.9867\n",
            "Epoch 5/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0326 - accuracy: 0.9893\n",
            "Epoch 6/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0258 - accuracy: 0.9918\n",
            "Epoch 7/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0240 - accuracy: 0.9922\n",
            "Epoch 8/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0194 - accuracy: 0.9933\n",
            "Epoch 9/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0196 - accuracy: 0.9929\n",
            "Epoch 10/10\n",
            "1875/1875 [==============================] - 25s 13ms/step - loss: 0.0137 - accuracy: 0.9956\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tensorflow.python.keras.callbacks.History at 0x7f4756c86990>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5DlEh7L9xbrE",
        "outputId": "4dc347a2-e0c6-4591-d226-c3a66b70f70c"
      },
      "source": [
        "model.evaluate(x_test,y_test)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 2s 5ms/step - loss: 0.0600 - accuracy: 0.9849\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.06004621461033821, 0.9848999977111816]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "zqX2TMQ1xjro",
        "outputId": "0d34c4e6-9e26-4b89-dee4-1253b3e07aef"
      },
      "source": [
        "image_index=4444\n",
        "plt.imshow(x_test[image_index].reshape(28,28),cmap='Greys')\n",
        "pred=model.predict(x_test[image_index].reshape(1,28,28,1))\n",
        "print(pred.argmax)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<built-in method argmax of numpy.ndarray object at 0x7f4754409f30>\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAANfElEQVR4nO3db6xU9Z3H8c9HthViq8Jyc0MoLt0GJaSxtBnJJiWNpllEEoP1gYEHDaumlweagCFRYqMlMfgv25I+MI23SgqmC2nSGnlA3LqkCUGT6mhYRbCVVUwhCEPQlMYoQr/74B7MLd45c5k58we/71dyMzPnO+eebw587pmZ35zzc0QIwBffJf1uAEBvEHYgCcIOJEHYgSQIO5DEP/VyYzNnzoy5c+f2cpNAKocOHdKJEyc8Ua2jsNteKunnkqZIeioiHi17/ty5c1Wv1zvZJIAStVqtaa3tl/G2p0h6QtJNkhZIWml7Qbu/D0B3dfKefZGkgxHxTkSclrRd0vJq2gJQtU7CPlvSX8Y9Plws+we2R2zXbdcbjUYHmwPQia5/Gh8RoxFRi4ja0NBQtzcHoIlOwn5E0pxxj79WLAMwgDoJ+yuS5tn+uu0vS1ohaUc1bQGoWttDbxFxxvbdkv5bY0NvmyPizco6A1CpjsbZI2KnpJ0V9QKgi/i6LJAEYQeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kQdiCJjqZstn1I0ilJZyWdiYhaFU0BqF5HYS/cEBEnKvg9ALqIl/FAEp2GPST93vartkcmeoLtEdt12/VGo9Hh5gC0q9OwL46I70i6SdJdtr93/hMiYjQiahFRGxoa6nBzANrVUdgj4khxe1zSs5IWVdEUgOq1HXbbl9n+6rn7kpZI2ldVYwCq1cmn8cOSnrV97vf8V0Q8X0lXACrXdtgj4h1J36qwFwBdxNAbkARhB5Ig7EAShB1IgrADSVRxIgwG2NmzZ0vrt99+e2n9mWeeKa0XQ69tufzyy0vrDzzwQGl93bp1bW87I47sQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AE4+wD4IMPPiitP/bYY22v//zz5WcdHz58uLTeahz90ksvLa0/8sgjTWt33HFH6brXXnttaX3FihWl9dmzZ5fWs+HIDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJMM4+AObNm1dabzUO302rV68urT/00EOl9ZkzZ7a97eHh4dJ6q3Pt169f3/a2v4g4sgNJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoyz98DJkyc7qndybfZOPfHEE6X1Sy7heHGxaPkvZXuz7eO2941bNsP2C7bfLm6nd7dNAJ2azJ/lX0laet6y9ZJ2RcQ8SbuKxwAGWMuwR8RuSee/zlwuaUtxf4ukWyruC0DF2n3DNRwRR4v770tq+iVm2yO267brjUajzc0B6FTHn65EREiKkvpoRNQiojY0NNTp5gC0qd2wH7M9S5KK2+PVtQSgG9oN+w5Jq4r7qyQ9V007ALql5Ti77W2Srpc00/ZhST+R9Kik39i+U9J7km7rZpMXu7Vr1/a7haZazc/ezXH0M2fOlNZbncfPZ0AXpmXYI2Jlk9L3K+4FQBfx9ScgCcIOJEHYgSQIO5AEYQeS4BTXHjhw4EBpferUqaX1Wq1WWt+zZ88F93TOxo0b2163Uy+++GJp/eDBg6X13bt3V9nOFx5HdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IgnH2Hmh1mui9995bWr/vvvtK69dcc03T2pEjR0rXffDBB0vr06d378LBo6OjpfVWl9DmMtYXhr0FJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kwzj4APvroo9L6tGnTSuv79u1rWmt1GeunnnqqtD424U9z/ZxOemRkpG/bvhhxZAeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJNxqHLVKtVot6vV6z7Y3KG644YbS+rvvvltab3Xd+bJx+Fb/vvv37y+ttzqfffv27aX1hx9+uGmt1ZTMrXz66ael9Yznu9dqNdXr9Qm//NByb9jebPu47X3jlm2wfcT23uJnWZUNA6jeZP70/UrS0gmWb4qIhcXPzmrbAlC1lmGPiN2STvagFwBd1Mmbmrttv168zG/6xs72iO267Xqj0ehgcwA60W7YfyHpG5IWSjoq6afNnhgRoxFRi4ja0NBQm5sD0Km2wh4RxyLibET8XdIvJS2qti0AVWsr7LZnjXv4A0nNz7EEMBBans9ue5uk6yXNtH1Y0k8kXW97oaSQdEjS6i72eNF78sknS+vz588vra9eXb57y66/3mru93vuuae0/vLLL5fWT506VVrvpozj6J1oGfaIWDnB4qe70AuALuJPI5AEYQeSIOxAEoQdSIKwA0lwKekeuPrqq0vrrYa/Nm3aVFrfubP5eUg33nhj6bqthtZOnz5dWm/1rchly5qfELlt27bSdW+99dbSOi4MR3YgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIJx9gHw+OOPl9bXrFlTWi87hfbDDz8sXbfVlM2LFy8urV955ZWl9bfeeqtpbevWraXrLl060XVO0S6O7EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBOPsA2DKlCml9auuuqq0vnHjxirbqdRLL73UtNZqOuklS5ZU3U5qHNmBJAg7kARhB5Ig7EAShB1IgrADSRB2IAnG2dFVJ06c6HcLKLQ8stueY/sPtvfbftP2mmL5DNsv2H67uJ3e/XYBtGsyL+PPSFoXEQsk/Zuku2wvkLRe0q6ImCdpV/EYwIBqGfaIOBoRrxX3T0k6IGm2pOWSthRP2yLplm41CaBzF/QBne25kr4t6Y+ShiPiaFF6X9Jwk3VGbNdt1xuNRgetAujEpMNu+yuSfitpbUT8dXwtxs5omPCshogYjYhaRNRaTQIIoHsmFXbbX9JY0H8dEb8rFh+zPauoz5J0vDstAqhCy6E325b0tKQDEfGzcaUdklZJerS4fa4rHeILa9q0aaX1qVOn9qiTHCYzzv5dST+U9IbtvcWy+zUW8t/YvlPSe5Ju606LAKrQMuwRsUeSm5S/X207ALqFr8sCSRB2IAnCDiRB2IEkCDuQBKe4oiMff/xxaX3Dhg1NazfffHPpuldccUU7LaEJjuxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kATj7OiqscshTGzBggU97AQc2YEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcbZ0ZFPPvmk3y1gkjiyA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EASk5mffY6krZKGJYWk0Yj4ue0Nkn4kqVE89f6I2NmtRjGY9u/f3/a61113XYWdoJXJfKnmjKR1EfGa7a9KetX2C0VtU0T8Z/faA1CVyczPflTS0eL+KdsHJM3udmMAqnVB79ltz5X0bUl/LBbdbft125ttT2+yzojtuu16o9GY6CkAemDSYbf9FUm/lbQ2Iv4q6ReSviFpocaO/D+daL2IGI2IWkTUhoaGKmgZQDsmFXbbX9JY0H8dEb+TpIg4FhFnI+Lvkn4paVH32gTQqZZh99jlQZ+WdCAifjZu+axxT/uBpH3VtwegKpP5NP67kn4o6Q3be4tl90taaXuhxobjDkla3ZUOMdCmT5/wo5rPzJgxo2lt8eLFVbeDEpP5NH6PpIku/s2YOnAR4Rt0QBKEHUiCsANJEHYgCcIOJEHYgSS4lDQ6Mn/+/NI650MMDo7sQBKEHUiCsANJEHYgCcIOJEHYgSQIO5CEI6J3G7Mbkt4bt2impBM9a+DCDGpvg9qXRG/tqrK3f4mICa//1tOwf27jdj0ian1roMSg9jaofUn01q5e9cbLeCAJwg4k0e+wj/Z5+2UGtbdB7Uuit3b1pLe+vmcH0Dv9PrID6BHCDiTRl7DbXmr7T7YP2l7fjx6asX3I9hu299qu97mXzbaP2943btkM2y/Yfru4Lb9we29722D7SLHv9tpe1qfe5tj+g+39tt+0vaZY3td9V9JXT/Zbz9+z254i6c+S/l3SYUmvSFoZEe1P9F0h24ck1SKi71/AsP09SX+TtDUivlkse1zSyYh4tPhDOT0i7huQ3jZI+lu/p/EuZiuaNX6acUm3SPoP9XHflfR1m3qw3/pxZF8k6WBEvBMRpyVtl7S8D30MvIjYLenkeYuXS9pS3N+isf8sPdekt4EQEUcj4rXi/ilJ56YZ7+u+K+mrJ/oR9tmS/jLu8WEN1nzvIen3tl+1PdLvZiYwHBFHi/vvSxruZzMTaDmNdy+dN834wOy7dqY/7xQf0H3e4oj4jqSbJN1VvFwdSDH2HmyQxk4nNY13r0wwzfhn+rnv2p3+vFP9CPsRSXPGPf5asWwgRMSR4va4pGc1eFNRHzs3g25xe7zP/XxmkKbxnmiacQ3Avuvn9Of9CPsrkubZ/rrtL0taIWlHH/r4HNuXFR+cyPZlkpZo8Kai3iFpVXF/laTn+tjLPxiUabybTTOuPu+7vk9/HhE9/5G0TGOfyP+fpB/3o4cmff2rpP8tft7sd2+StmnsZd2nGvts405J/yxpl6S3Jf2PpBkD1Nszkt6Q9LrGgjWrT70t1thL9Ncl7S1+lvV735X01ZP9xtdlgST4gA5IgrADSRB2IAnCDiRB2IEkCDuQBGEHkvh/3cEPVjF3ogoAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "er2n2mwfx_dO"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}