MNIST_2021_2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "MNIST 2021-2.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/tjrudrnr2/A.I/blob/main/MNIST_2021_2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "psCOlSnRzMVw"
      },
      "source": [
        "# MNIST Practice"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "g6xPjSr4x5zV"
      },
      "source": [
        "## Imports"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iZaF1tEMzOtu"
      },
      "source": [
        "import numpy as np\n",
        "from tensorflow import keras\n",
        "from tensorflow.keras import layers"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "n413y3ICx7so"
      },
      "source": [
        "## Prepare Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uYayXrySzoIX",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ca4706e8-267e-49bc-9aff-17c337dd3875"
      },
      "source": [
        "# Model / data parameters\n",
        "num_classes = 10\n",
        "input_shape = (28, 28, 1)\n",
        "\n",
        "# the data, split between train and test sets\n",
        "(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()\n",
        "\n",
        "# Scale images to the [0, 1] range\n",
        "x_train = x_train.astype(\"float32\") / 255\n",
        "x_test = x_test.astype(\"float32\") / 255\n",
        "# Make sure images have shape (28, 28, 1)\n",
        "x_train = np.expand_dims(x_train, -1)\n",
        "x_test = np.expand_dims(x_test, -1)\n",
        "print(\"x_train shape:\", x_train.shape)\n",
        "print(x_train.shape[0], \"train samples\")\n",
        "print(x_test.shape[0], \"test samples\")\n",
        "\n",
        "# convert class vectors to binary class matrices\n",
        "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
        "y_test = keras.utils.to_categorical(y_test, num_classes)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 0s 0us/step\n",
            "11501568/11490434 [==============================] - 0s 0us/step\n",
            "x_train shape: (60000, 28, 28, 1)\n",
            "60000 train samples\n",
            "10000 test samples\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "v5eUA4lcx_OI"
      },
      "source": [
        "## Explore Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qmyNxc6mwd1w",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 317
        },
        "outputId": "3ed48139-9fdf-486d-fc6a-166387e47a8e"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "\n",
        "img_id = 100\n",
        "print('Image shape:', x_train[img_id].shape)\n",
        "print('Image Label Array:', y_train[img_id], \n",
        "      'Image label shape:', y_train[img_id].shape, \n",
        "      '\\nImage Label:', np.argmax(y_train[img_id]))\n",
        "plt.imshow(x_train[img_id].reshape(28, 28), interpolation='nearest')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Image shape: (28, 28, 1)\n",
            "Image Label Array: [0. 0. 0. 0. 0. 1. 0. 0. 0. 0.] Image label shape: (10,) \n",
            "Image Label: 5\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-Ikv03jjwDnJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 533
        },
        "outputId": "8b822526-5417-491b-ab6c-28f5c3c0cbb7"
      },
      "source": [
        "# preview the images first\n",
        "plt.figure(figsize=(12,10))\n",
        "x, y = 10, 4\n",
        "for i in range(40):  \n",
        "    plt.subplot(y, x, i+1)\n",
        "    plt.imshow(x_train[i].reshape((28,28)),interpolation='nearest')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x720 with 40 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FcpA-8wVyBoe"
      },
      "source": [
        "## Make Model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aeFwt4_xzoK7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d76e14bd-a97b-4445-b768-bf157555ea16"
      },
      "source": [
        "model = keras.Sequential(\n",
        "    [\n",
        "        keras.Input(shape=input_shape),\n",
        "        layers.Conv2D(32, kernel_size=(3, 3), activation=\"relu\"),\n",
        "        layers.MaxPooling2D(pool_size=(2, 2)),\n",
        "        layers.Conv2D(64, kernel_size=(3, 3), activation=\"relu\"),\n",
        "        layers.MaxPooling2D(pool_size=(2, 2)),\n",
        "        layers.Flatten(),\n",
        "        layers.Dropout(0.5),\n",
        "        layers.Dense(num_classes, activation=\"softmax\"),\n",
        "    ]\n",
        ")\n",
        "\n",
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 26, 26, 32)        320       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 13, 13, 32)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 11, 11, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 5, 5, 64)         0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 1600)              0         \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 1600)              0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 10)                16010     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 34,826\n",
            "Trainable params: 34,826\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TZQipTVMyEA2"
      },
      "source": [
        "## Set Training Parameter & Train"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rWxjDyP6zwZX",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f0d85c42-d316-4c33-84a9-1275c66e6570"
      },
      "source": [
        "batch_size = 128\n",
        "epochs = 15\n",
        "\n",
        "model.compile(loss=\"categorical_crossentropy\", \n",
        "              optimizer=\"adam\", \n",
        "              metrics=[\"accuracy\"])\n",
        "\n",
        "history = model.fit(x_train, y_train, \n",
        "                    batch_size=batch_size, \n",
        "                    epochs=epochs, \n",
        "                    validation_split=0.1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/15\n",
            "422/422 [==============================] - 42s 97ms/step - loss: 0.3806 - accuracy: 0.8859 - val_loss: 0.0836 - val_accuracy: 0.9772\n",
            "Epoch 2/15\n",
            "422/422 [==============================] - 40s 94ms/step - loss: 0.1085 - accuracy: 0.9670 - val_loss: 0.0564 - val_accuracy: 0.9845\n",
            "Epoch 3/15\n",
            "422/422 [==============================] - 38s 90ms/step - loss: 0.0808 - accuracy: 0.9756 - val_loss: 0.0450 - val_accuracy: 0.9878\n",
            "Epoch 4/15\n",
            "422/422 [==============================] - 39s 92ms/step - loss: 0.0639 - accuracy: 0.9801 - val_loss: 0.0409 - val_accuracy: 0.9897\n",
            "Epoch 5/15\n",
            "422/422 [==============================] - 39s 93ms/step - loss: 0.0585 - accuracy: 0.9823 - val_loss: 0.0373 - val_accuracy: 0.9893\n",
            "Epoch 6/15\n",
            "422/422 [==============================] - 39s 93ms/step - loss: 0.0505 - accuracy: 0.9843 - val_loss: 0.0350 - val_accuracy: 0.9900\n",
            "Epoch 7/15\n",
            "422/422 [==============================] - 39s 93ms/step - loss: 0.0476 - accuracy: 0.9847 - val_loss: 0.0367 - val_accuracy: 0.9903\n",
            "Epoch 8/15\n",
            "422/422 [==============================] - 40s 95ms/step - loss: 0.0446 - accuracy: 0.9861 - val_loss: 0.0329 - val_accuracy: 0.9917\n",
            "Epoch 9/15\n",
            "422/422 [==============================] - 39s 93ms/step - loss: 0.0421 - accuracy: 0.9867 - val_loss: 0.0338 - val_accuracy: 0.9902\n",
            "Epoch 10/15\n",
            "422/422 [==============================] - 40s 96ms/step - loss: 0.0379 - accuracy: 0.9878 - val_loss: 0.0299 - val_accuracy: 0.9917\n",
            "Epoch 11/15\n",
            "422/422 [==============================] - 39s 92ms/step - loss: 0.0353 - accuracy: 0.9886 - val_loss: 0.0272 - val_accuracy: 0.9928\n",
            "Epoch 12/15\n",
            "422/422 [==============================] - 38s 91ms/step - loss: 0.0337 - accuracy: 0.9894 - val_loss: 0.0292 - val_accuracy: 0.9927\n",
            "Epoch 13/15\n",
            "422/422 [==============================] - 39s 92ms/step - loss: 0.0338 - accuracy: 0.9892 - val_loss: 0.0302 - val_accuracy: 0.9913\n",
            "Epoch 14/15\n",
            "422/422 [==============================] - 39s 92ms/step - loss: 0.0322 - accuracy: 0.9896 - val_loss: 0.0321 - val_accuracy: 0.9918\n",
            "Epoch 15/15\n",
            "422/422 [==============================] - 39s 92ms/step - loss: 0.0300 - accuracy: 0.9905 - val_loss: 0.0281 - val_accuracy: 0.9920\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3pAkEyvC9k_b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 562
        },
        "outputId": "025b6c54-e145-4f8f-a35d-01742bbf0da4"
      },
      "source": [
        "print(history.history.keys())\n",
        "train_accuracy = history.history['accuracy']\n",
        "val_accuracy = history.history['val_accuracy']\n",
        "train_loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "plt.plot(list(range(epochs)), train_accuracy, 'bo', label='Training accuracy')\n",
        "plt.plot(list(range(epochs)), val_accuracy, 'b', label='Validation accuracy')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "plt.figure()\n",
        "plt.plot(list(range(epochs)), train_loss, 'bo', label='Training loss')\n",
        "plt.plot(list(range(epochs)), val_loss, 'b', label='Validation loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "leR-L8qsyI3x"
      },
      "source": [
        "## Evaluate performance on Test Set"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AjekCYGBzwb7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4b22e04e-428f-4bd5-efab-ea7b492af279"
      },
      "source": [
        "score = model.evaluate(x_test, y_test, verbose=0)\n",
        "print(\"Test loss:\", score[0])\n",
        "print(\"Test accuracy:\", score[1])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test loss: 0.024434229359030724\n",
            "Test accuracy: 0.9912999868392944\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PTFe0i6NyONR"
      },
      "source": [
        "## Sample prediction"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lASOCHpdzwhs",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 406
        },
        "outputId": "91616dc8-243d-4aee-e134-b64c8a7e121a"
      },
      "source": [
        "img_id = 0\n",
        "\n",
        "sample_test_image = x_test[img_id]\n",
        "sample_test_label = y_test[img_id]\n",
        "\n",
        "print('Image shape:', sample_test_image.shape)\n",
        "print('True Image Label Array:', sample_test_label, \n",
        "      '\\nTrue Image Label Shape:', sample_test_label.shape,\n",
        "      '\\nTrue Image Label:', np.argmax(sample_test_label))\n",
        "plt.imshow(sample_test_image.reshape(28, 28), interpolation='nearest')\n",
        "plt.show()\n",
        "\n",
        "prediction = model.predict(sample_test_image.reshape(1, 28, 28, 1))\n",
        "#print('Predicted Image Label Array:')\n",
        "#print(*['{0:0.2f}'.format(i) for i in prediction[0]], sep='\\n')\n",
        "print('Predicted Image Label Array:', ['{0:0.2f}'.format(i) for i in prediction[0]],\n",
        "      '\\nPredicted Image Label Shape:', prediction.shape, \n",
        "      '\\nPredicted Image Label:', np.argmax(prediction))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Image shape: (28, 28, 1)\n",
            "True Image Label Array: [0. 0. 0. 0. 0. 0. 0. 1. 0. 0.] \n",
            "True Image Label Shape: (10,) \n",
            "True Image Label: 7\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAANiklEQVR4nO3df4wc9XnH8c8n/kV8QGtDcF3j4ISQqE4aSHWBRNDKESUFImSiJBRLtVyJ5lALElRRW0QVBalVSlEIok0aySluHESgaQBhJTSNa6W1UKljg4yxgdaEmsau8QFOaxPAP/DTP24cHXD7vWNndmft5/2SVrs7z87Oo/F9PLMzO/t1RAjA8e9tbTcAoD8IO5AEYQeSIOxAEoQdSGJ6Pxc207PiBA31c5FAKq/qZzoYBzxRrVbYbV8s6XZJ0yT9bUTcXHr9CRrSeb6wziIBFGyIdR1rXe/G254m6auSLpG0WNIy24u7fT8AvVXnM/u5kp6OiGci4qCkeyQtbaYtAE2rE/YFkn4y7vnOatrr2B6xvcn2pkM6UGNxAOro+dH4iFgZEcMRMTxDs3q9OAAd1An7LkkLxz0/vZoGYADVCftGSWfZfpftmZKulLSmmbYANK3rU28Rcdj2tZL+SWOn3lZFxLbGOgPQqFrn2SPiQUkPNtQLgB7i67JAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQIO5AEYQeSIOxAEoQdSIKwA0kQdiAJwg4kQdiBJGoN2Wx7h6T9kl6TdDgihptoCkDzaoW98rGIeKGB9wHQQ+zGA0nUDXtI+oHtR2yPTPQC2yO2N9nedEgHai4OQLfq7sZfEBG7bJ8maa3tpyJi/fgXRMRKSSsl6WTPjZrLA9ClWlv2iNhV3Y9Kul/SuU00BaB5XYfd9pDtk44+lvRxSVubagxAs+rsxs+TdL/to+/zrYj4fiNdAWhc12GPiGcknd1gLwB6iFNvQBKEHUiCsANJEHYgCcIOJNHEhTApvPjZj3asvXP508V5nxqdV6wfPDCjWF9wd7k+e+dLHWtHNj9RnBd5sGUHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQ4zz5Ff/xH3+pY+9TQT8szn1lz4UvK5R2HX+5Yu/35j9Vc+LHrR6NndKwN3foLxXmnr3uk6XZax5YdSIKwA0kQdiAJwg4kQdiBJAg7kARhB5JwRP8GaTnZc+M8X9i35TXpZ58+r2PthQ+W/8+c82R5Hf/0V1ysz/zg/xbrt3zgvo61i97+SnHe7718YrH+idmdr5Wv65U4WKxvODBUrC854VDXy37P964u1t87srHr927ThlinfbF3wj8otuxAEoQdSIKwA0kQdiAJwg4kQdiBJAg7kATXs0/R0Hc2FGr13vvkerPrr39pScfan5+/qLzsfy3/5v0tS97TRUdTM/2VI8X60Jbdxfop6+8t1n91Zuff25+9o/xb/MejSbfstlfZHrW9ddy0ubbX2t5e3c/pbZsA6prKbvw3JF38hmk3SFoXEWdJWlc9BzDAJg17RKyXtPcNk5dKWl09Xi3p8ob7AtCwbj+zz4uIox+onpPUcTAz2yOSRiTpBM3ucnEA6qp9ND7GrqTpeKVHRKyMiOGIGJ6hWXUXB6BL3YZ9j+35klTdjzbXEoBe6DbsayStqB6vkPRAM+0A6JVJP7Pbvltjv1x+qu2dkr4g6WZJ37Z9laRnJV3RyyZRdvi5PR1rQ/d2rknSa5O899B3Xuyio2bs+b2PFuvvn1n+8/3S3vd1rC36u2eK8x4uVo9Nk4Y9IpZ1KB2bv0IBJMXXZYEkCDuQBGEHkiDsQBKEHUiCS1zRmulnLCzWv3LjV4r1GZ5WrP/D7b/ZsXbK7oeL8x6P2LIDSRB2IAnCDiRB2IEkCDuQBGEHkiDsQBKcZ0drnvrDBcX6h2eVh7LedrA8HPXcJ15+yz0dz9iyA0kQdiAJwg4kQdiBJAg7kARhB5Ig7EASnGdHTx34xIc71h799G2TzF0eQej3r7uuWH/7v/1okvfPhS07kARhB5Ig7EAShB1IgrADSRB2IAnCDiTBeXb01H9f0nl7cqLL59GX/ddFxfrs7z9WrEexms+kW3bbq2yP2t46btpNtnfZ3lzdLu1tmwDqmspu/DckXTzB9Nsi4pzq9mCzbQFo2qRhj4j1kvb2oRcAPVTnAN21trdUu/lzOr3I9ojtTbY3HdKBGosDUEe3Yf+apDMlnSNpt6RbO70wIlZGxHBEDM+Y5MIGAL3TVdgjYk9EvBYRRyR9XdK5zbYFoGldhd32/HFPPylpa6fXAhgMk55nt323pCWSTrW9U9IXJC2xfY7GTmXukHR1D3vEAHvbSScV68t//aGOtX1HXi3OO/rFdxfrsw5sLNbxepOGPSKWTTD5jh70AqCH+LoskARhB5Ig7EAShB1IgrADSXCJK2rZftP7i/Xvnvo3HWtLt3+qOO+sBzm11iS27EAShB1IgrADSRB2IAnCDiRB2IEkCDuQBOfZUfR/v/ORYn3Lb/9Vsf7jw4c61l76y9OL887S7mIdbw1bdiAJwg4kQdiBJAg7kARhB5Ig7EAShB1IgvPsyU1f8MvF+vWf//tifZbLf0JXPra8Y+0d/8j16v3Elh1IgrADSRB2IAnCDiRB2IEkCDuQBGEHkuA8+3HO08v/xGd/d2ex/pkTXyzW79p/WrE+7/OdtydHinOiaZNu2W0vtP1D20/Y3mb7umr6XNtrbW+v7uf0vl0A3ZrKbvxhSZ+LiMWSPiLpGtuLJd0gaV1EnCVpXfUcwICaNOwRsTsiHq0e75f0pKQFkpZKWl29bLWky3vVJID63tJndtuLJH1I0gZJ8yLi6I+EPSdpXod5RiSNSNIJmt1tnwBqmvLReNsnSrpX0vURsW98LSJCUkw0X0SsjIjhiBieoVm1mgXQvSmF3fYMjQX9roi4r5q8x/b8qj5f0mhvWgTQhEl3421b0h2SnoyIL48rrZG0QtLN1f0DPekQ9Zz9vmL5z067s9bbf/WLnynWf/Gxh2u9P5ozlc/s50taLulx25uraTdqLOTftn2VpGclXdGbFgE0YdKwR8RDktyhfGGz7QDoFb4uCyRB2IEkCDuQBGEHkiDsQBJc4nocmLb4vR1rI/fU+/rD4lXXFOuL7vz3Wu+P/mHLDiRB2IEkCDuQBGEHkiDsQBKEHUiCsANJcJ79OPDUH3T+Yd/LZu/rWJuK0//lYPkFMeEPFGEAsWUHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSQ4z34MePWyc4v1dZfdWqgy5BbGsGUHkiDsQBKEHUiCsANJEHYgCcIOJEHYgSSmMj77QknflDRPUkhaGRG3275J0mclPV+99MaIeLBXjWb2P+dPK9bfOb37c+l37T+tWJ+xr3w9O1ezHzum8qWaw5I+FxGP2j5J0iO211a12yLiS71rD0BTpjI++25Ju6vH+20/KWlBrxsD0Ky39Jnd9iJJH5K0oZp0re0ttlfZnvC3kWyP2N5ke9MhHajVLIDuTTnstk+UdK+k6yNin6SvSTpT0jka2/JP+AXtiFgZEcMRMTxDsxpoGUA3phR22zM0FvS7IuI+SYqIPRHxWkQckfR1SeWrNQC0atKw27akOyQ9GRFfHjd9/riXfVLS1ubbA9CUqRyNP1/SckmP295cTbtR0jLb52js7MsOSVf3pEPU8hcvLi7WH/6tRcV67H68wW7QpqkcjX9IkicocU4dOIbwDTogCcIOJEHYgSQIO5AEYQeSIOxAEo4+Drl7sufGeb6wb8sDstkQ67Qv9k50qpwtO5AFYQeSIOxAEoQdSIKwA0kQdiAJwg4k0dfz7Lafl/TsuEmnSnqhbw28NYPa26D2JdFbt5rs7YyIeMdEhb6G/U0LtzdFxHBrDRQMam+D2pdEb93qV2/sxgNJEHYgibbDvrLl5ZcMam+D2pdEb93qS2+tfmYH0D9tb9kB9AlhB5JoJey2L7b9H7aftn1DGz10YnuH7cdtb7a9qeVeVtketb113LS5ttfa3l7dTzjGXku93WR7V7XuNtu+tKXeFtr+oe0nbG+zfV01vdV1V+irL+ut75/ZbU+T9J+SLpK0U9JGScsi4om+NtKB7R2ShiOi9S9g2P4NSS9J+mZEfKCadoukvRFxc/Uf5ZyI+JMB6e0mSS+1PYx3NVrR/PHDjEu6XNLvqsV1V+jrCvVhvbWxZT9X0tMR8UxEHJR0j6SlLfQx8CJivaS9b5i8VNLq6vFqjf2x9F2H3gZCROyOiEerx/slHR1mvNV1V+irL9oI+wJJPxn3fKcGa7z3kPQD24/YHmm7mQnMi4jd1ePnJM1rs5kJTDqMdz+9YZjxgVl33Qx/XhcH6N7sgoj4NUmXSLqm2l0dSDH2GWyQzp1OaRjvfplgmPGfa3PddTv8eV1thH2XpIXjnp9eTRsIEbGruh+VdL8GbyjqPUdH0K3uR1vu5+cGaRjviYYZ1wCsuzaHP28j7BslnWX7XbZnSrpS0poW+ngT20PVgRPZHpL0cQ3eUNRrJK2oHq+Q9ECLvbzOoAzj3WmYcbW87lof/jwi+n6TdKnGjsj/WNKfttFDh77eLemx6rat7d4k3a2x3bpDGju2cZWkUyStk7Rd0j9LmjtAvd0p6XFJWzQWrPkt9XaBxnbRt0jaXN0ubXvdFfrqy3rj67JAEhygA5Ig7EAShB1IgrADSRB2IAnCDiRB2IEk/h9BCfQTVPflJQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Predicted Image Label Array: ['0.00', '0.00', '0.00', '0.00', '0.00', '0.00', '0.00', '1.00', '0.00', '0.00'] \n",
            "Predicted Image Label Shape: (1, 10) \n",
            "Predicted Image Label: 7\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EtlqSr0_zwvP"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hrkokwhmzw1m"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jThVEBR6zw32"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ETKXSvC8zw6Q"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "spIz7InEzoNI"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}