01_keras_cifar10_cnn.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "01-keras-cifar10-cnn.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.7.6"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/mehrotrasan16/Keras-Deep-Learning/blob/master/01_keras_cifar10_cnn.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "m7KBpffWzlxH"
      },
      "source": [
        "### Import TensorFlow"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "iAve6DCL4JH4",
        "colab": {}
      },
      "source": [
        "import tensorflow as tf\n",
        "from tensorflow.keras import datasets, layers, models\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "from tensorflow import keras"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "jRFxccghyMVo"
      },
      "source": [
        "### Download and prepare the CIFAR10 dataset\n",
        "\n",
        "\n",
        "The CIFAR10 dataset contains 60,000 color images in 10 classes, with 6,000 images in each class. The dataset is divided into 50,000 training images and 10,000 testing images. The classes are mutually exclusive and there is no overlap between them."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "JWoEqyMuXFF4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "acfd5fa5-fac5-4b54-e2ec-f2631897ff53"
      },
      "source": [
        "(train_images, train_labels), (test_images, test_labels) = datasets.cifar10.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"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
            "170500096/170498071 [==============================] - 11s 0us/step\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "7wArwCTJJlUa"
      },
      "source": [
        "### Verify the data\n",
        "\n",
        "To verify that the dataset looks correct, let's plot the first 25 images from the training set and display the class name below each image.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "K3PAELE2eSU9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 589
        },
        "outputId": "4e963ff7-60c5-4085-9998-1b91280b1037"
      },
      "source": [
        "class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',\n",
        "               'dog', 'frog', 'horse', 'ship', 'truck']\n",
        "\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.grid(False)\n",
        "    plt.imshow(train_images[i], cmap=plt.cm.binary)\n",
        "    # The CIFAR labels happen to be arrays, \n",
        "    # which is why you need the extra index\n",
        "    plt.xlabel(class_names[train_labels[i][0]])\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 25 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "exAXW6C8vfe8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "da3034c4-af72-4867-bff8-7d970dfbd349"
      },
      "source": [
        "len(train_images),len(test_images)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(50000, 10000)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K9rMp2-qwZQO",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "0fc64cde-6797-4fe3-b7c8-98872059b35f"
      },
      "source": [
        "len(train_images)/20"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2500.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GJdXtnolYbIJ",
        "colab_type": "text"
      },
      "source": [
        "## 68.839"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "Oewp-wYg31t9"
      },
      "source": [
        "### Create the model (your code goes here)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "L9YmGQBQPrdn",
        "colab": {}
      },
      "source": [
        "#INSERT YOUR MODEL CODE HERE\n",
        "#YOU CAN ALSO CHANGE THE CODE BELOW TO COMPILE AND RUN THE MODEL\n",
        "#setup model\n",
        "model = keras.Sequential([\n",
        "    keras.layers.Conv2D(kernel_size=3,filters=12,use_bias=False,padding='same',input_shape=(32,32,3)),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "\n",
        "    keras.layers.Conv2D(kernel_size=6,filters=24,use_bias=False,padding='same',strides=2),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    \n",
        "    keras.layers.Conv2D(kernel_size=6,filters=32,use_bias=False,padding='same',strides=2),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    \n",
        "    keras.layers.Flatten(),\n",
        "    \n",
        "    keras.layers.Dense(128),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.4),\n",
        "\n",
        "    keras.layers.Dense(200),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.4),\n",
        "    \n",
        "    keras.layers.Dense(10, activation = tf.nn.softmax)    \n",
        "])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "P3odqfHP4M67"
      },
      "source": [
        "### Compile and train the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "MdDzI75PUXrG",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 697
        },
        "outputId": "e2076a53-ca4e-4923-a3f8-bf64ce6379a6"
      },
      "source": [
        "model.compile(optimizer='adam',\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "history = model.fit(train_images, train_labels, epochs=20, \n",
        "                    validation_data=(test_images, test_labels))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 2.0443 - accuracy: 0.4152 - val_loss: 1.9468 - val_accuracy: 0.5145\n",
            "Epoch 2/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.9386 - accuracy: 0.5201 - val_loss: 1.9811 - val_accuracy: 0.4748\n",
            "Epoch 3/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8897 - accuracy: 0.5710 - val_loss: 1.8486 - val_accuracy: 0.6087\n",
            "Epoch 4/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8569 - accuracy: 0.6028 - val_loss: 1.9459 - val_accuracy: 0.5113\n",
            "Epoch 5/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8281 - accuracy: 0.6318 - val_loss: 1.8034 - val_accuracy: 0.6544\n",
            "Epoch 6/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8111 - accuracy: 0.6493 - val_loss: 1.7801 - val_accuracy: 0.6786\n",
            "Epoch 7/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7977 - accuracy: 0.6628 - val_loss: 1.8511 - val_accuracy: 0.6073\n",
            "Epoch 8/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7872 - accuracy: 0.6724 - val_loss: 1.8497 - val_accuracy: 0.6097\n",
            "Epoch 9/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7740 - accuracy: 0.6862 - val_loss: 1.7967 - val_accuracy: 0.6632\n",
            "Epoch 10/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7668 - accuracy: 0.6930 - val_loss: 1.8486 - val_accuracy: 0.6102\n",
            "Epoch 11/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7580 - accuracy: 0.7017 - val_loss: 1.7832 - val_accuracy: 0.6737\n",
            "Epoch 12/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7533 - accuracy: 0.7066 - val_loss: 1.8964 - val_accuracy: 0.5619\n",
            "Epoch 13/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7473 - accuracy: 0.7128 - val_loss: 1.7701 - val_accuracy: 0.6884\n",
            "Epoch 14/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7355 - accuracy: 0.7248 - val_loss: 1.7487 - val_accuracy: 0.7112\n",
            "Epoch 15/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7297 - accuracy: 0.7305 - val_loss: 1.7814 - val_accuracy: 0.6774\n",
            "Epoch 16/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7260 - accuracy: 0.7339 - val_loss: 1.7984 - val_accuracy: 0.6599\n",
            "Epoch 17/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7198 - accuracy: 0.7409 - val_loss: 1.8177 - val_accuracy: 0.6409\n",
            "Epoch 18/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7169 - accuracy: 0.7436 - val_loss: 1.9990 - val_accuracy: 0.4606\n",
            "Epoch 19/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7121 - accuracy: 0.7487 - val_loss: 1.7380 - val_accuracy: 0.7215\n",
            "Epoch 20/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7074 - accuracy: 0.7535 - val_loss: 1.7515 - val_accuracy: 0.7063\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "jKgyC5K_4O0d"
      },
      "source": [
        "### Evaluate the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "gtyDF0MKUcM7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "outputId": "e5d93897-12b0-474d-e3f4-65643e9de58a"
      },
      "source": [
        "plt.plot(history.history['accuracy'], label='accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.ylim([0.5, 1])\n",
        "plt.legend(loc='lower right')\n",
        "\n",
        "test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.7515 - accuracy: 0.7063\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "0LvwaKhtUdOo",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "dbb5855c-1954-4ecb-a272-0c9d1758d44c"
      },
      "source": [
        "print(test_acc)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.7063000202178955\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pVQpo2vGEDdO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YsVmYuioYhoN",
        "colab_type": "text"
      },
      "source": [
        "## 69.29"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "ZxTTlLt8E7X8",
        "colab": {}
      },
      "source": [
        "#INSERT YOUR MODEL CODE HERE\n",
        "#YOU CAN ALSO CHANGE THE CODE BELOW TO COMPILE AND RUN THE MODEL\n",
        "#setup model\n",
        "model = keras.Sequential([\n",
        "    keras.layers.Conv2D(kernel_size=3,filters=12,use_bias=False,padding='same',input_shape=(32,32,3)),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "\n",
        "    keras.layers.Conv2D(kernel_size=6,filters=24,use_bias=False,padding='same',strides=2),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    \n",
        "    keras.layers.Conv2D(kernel_size=12,filters=32,use_bias=False,padding='same',strides=2),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    \n",
        "    # keras.layers.Conv2D(kernel_size=24,filters=32,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    \n",
        "    keras.layers.Flatten(),\n",
        "    \n",
        "    # keras.layers.Dense(128),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Dropout(0.5),\n",
        "\n",
        "    keras.layers.Dense(200),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "\n",
        "    keras.layers.Dense(500),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "    \n",
        "    keras.layers.Dense(10, activation = tf.nn.softmax)    \n",
        "])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "cSLv13hIE7YD"
      },
      "source": [
        "### Compile and train the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "jQvWqFPRE7YD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 697
        },
        "outputId": "0231c26b-acbf-48ba-8dcb-57e1a8346ef2"
      },
      "source": [
        "model.compile(optimizer='adam',\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "history = model.fit(train_images, train_labels, epochs=20, \n",
        "                    validation_data=(test_images, test_labels))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 2.0628 - accuracy: 0.3918 - val_loss: 2.0334 - val_accuracy: 0.4203\n",
            "Epoch 2/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.9665 - accuracy: 0.4899 - val_loss: 1.9670 - val_accuracy: 0.4907\n",
            "Epoch 3/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.9158 - accuracy: 0.5423 - val_loss: 1.9355 - val_accuracy: 0.5228\n",
            "Epoch 4/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8834 - accuracy: 0.5742 - val_loss: 1.9481 - val_accuracy: 0.5094\n",
            "Epoch 5/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8623 - accuracy: 0.5957 - val_loss: 1.9254 - val_accuracy: 0.5337\n",
            "Epoch 6/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8389 - accuracy: 0.6200 - val_loss: 1.8388 - val_accuracy: 0.6183\n",
            "Epoch 7/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8246 - accuracy: 0.6347 - val_loss: 1.8450 - val_accuracy: 0.6125\n",
            "Epoch 8/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8147 - accuracy: 0.6438 - val_loss: 1.8156 - val_accuracy: 0.6430\n",
            "Epoch 9/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8023 - accuracy: 0.6569 - val_loss: 1.8713 - val_accuracy: 0.5839\n",
            "Epoch 10/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7942 - accuracy: 0.6648 - val_loss: 1.8023 - val_accuracy: 0.6572\n",
            "Epoch 11/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7837 - accuracy: 0.6755 - val_loss: 1.8551 - val_accuracy: 0.6026\n",
            "Epoch 12/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7764 - accuracy: 0.6824 - val_loss: 1.7989 - val_accuracy: 0.6601\n",
            "Epoch 13/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7678 - accuracy: 0.6923 - val_loss: 1.8000 - val_accuracy: 0.6604\n",
            "Epoch 14/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7630 - accuracy: 0.6963 - val_loss: 1.8787 - val_accuracy: 0.5789\n",
            "Epoch 15/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7550 - accuracy: 0.7053 - val_loss: 1.7834 - val_accuracy: 0.6763\n",
            "Epoch 16/20\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.7503 - accuracy: 0.7096 - val_loss: 1.7953 - val_accuracy: 0.6640\n",
            "Epoch 17/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7434 - accuracy: 0.7166 - val_loss: 1.7766 - val_accuracy: 0.6823\n",
            "Epoch 18/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7383 - accuracy: 0.7212 - val_loss: 1.8394 - val_accuracy: 0.6181\n",
            "Epoch 19/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7353 - accuracy: 0.7240 - val_loss: 1.7714 - val_accuracy: 0.6879\n",
            "Epoch 20/20\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7268 - accuracy: 0.7332 - val_loss: 1.8010 - val_accuracy: 0.6587\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "hsTO6l9ME7YG"
      },
      "source": [
        "### Evaluate the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "kvgP1EISE7YG",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "outputId": "74592dec-b7d8-417a-c7bf-44bc6295e8a6"
      },
      "source": [
        "plt.plot(history.history['accuracy'], label='accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.ylim([0.5, 1])\n",
        "plt.legend(loc='lower right')\n",
        "\n",
        "test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8010 - accuracy: 0.6587\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "iJHzJQPtE7YJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "f01a0617-01dd-4bc7-f7f1-f297f0ef20e7"
      },
      "source": [
        "print(test_acc)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.6586999893188477\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "MibsAm_PE7YL",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iIfFWwMTYm1s",
        "colab_type": "text"
      },
      "source": [
        "# 63.20"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "iGUkzF7_Mgkk",
        "colab": {}
      },
      "source": [
        "#INSERT YOUR MODEL CODE HERE\n",
        "#YOU CAN ALSO CHANGE THE CODE BELOW TO COMPILE AND RUN THE MODEL\n",
        "#setup model\n",
        "model = keras.Sequential([\n",
        "    keras.layers.Conv2D(kernel_size=3,filters=12,use_bias=False,padding='same',input_shape=(32,32,3)),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    # keras.layers.Dropout(0.5),\n",
        "    keras.layers.MaxPool2D(),\n",
        "\n",
        "    keras.layers.Conv2D(kernel_size=6,filters=24,use_bias=False,padding='same',strides=2),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    # keras.layers.Dropout(0.5),\n",
        "    keras.layers.MaxPool2D(),\n",
        "    \n",
        "    keras.layers.Conv2D(kernel_size=12,filters=32,use_bias=False,padding='same',strides=2),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.MaxPool2D(),\n",
        "    #keras.layers.Dropout(0.5),\n",
        "    \n",
        "    # keras.layers.Conv2D(kernel_size=24,filters=32,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    \n",
        "    keras.layers.Flatten(),\n",
        "    \n",
        "    # keras.layers.Dense(128),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Dropout(0.5),\n",
        "\n",
        "    # keras.layers.Dense(200),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Dropout(0.5),\n",
        "\n",
        "    keras.layers.Dense(700),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "    \n",
        "    keras.layers.Dense(10, activation = tf.nn.softmax)    \n",
        "])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "PQyFJg5HMgk6"
      },
      "source": [
        "### Compile and train the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "45ddfUDhMgk8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 697
        },
        "outputId": "d8957350-89ec-41a4-c488-5ac98349d2d8"
      },
      "source": [
        "model.compile(optimizer='sgd',\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "history = model.fit(train_images, train_labels, epochs=20, \n",
        "                    validation_data=(test_images, test_labels))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 2.0781 - accuracy: 0.3854 - val_loss: 1.9930 - val_accuracy: 0.4711\n",
            "Epoch 2/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9665 - accuracy: 0.4975 - val_loss: 1.9653 - val_accuracy: 0.4958\n",
            "Epoch 3/20\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.9158 - accuracy: 0.5509 - val_loss: 1.9690 - val_accuracy: 0.4886\n",
            "Epoch 4/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8825 - accuracy: 0.5823 - val_loss: 1.9958 - val_accuracy: 0.4619\n",
            "Epoch 5/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8588 - accuracy: 0.6057 - val_loss: 1.8863 - val_accuracy: 0.5748\n",
            "Epoch 6/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8423 - accuracy: 0.6222 - val_loss: 1.9115 - val_accuracy: 0.5518\n",
            "Epoch 7/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8263 - accuracy: 0.6373 - val_loss: 1.8555 - val_accuracy: 0.6061\n",
            "Epoch 8/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8163 - accuracy: 0.6471 - val_loss: 1.8662 - val_accuracy: 0.5954\n",
            "Epoch 9/20\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8077 - accuracy: 0.6556 - val_loss: 1.8551 - val_accuracy: 0.6049\n",
            "Epoch 10/20\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.7998 - accuracy: 0.6642 - val_loss: 1.9353 - val_accuracy: 0.5216\n",
            "Epoch 11/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7919 - accuracy: 0.6716 - val_loss: 1.9056 - val_accuracy: 0.5520\n",
            "Epoch 12/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7853 - accuracy: 0.6777 - val_loss: 1.8688 - val_accuracy: 0.5917\n",
            "Epoch 13/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7816 - accuracy: 0.6821 - val_loss: 1.8547 - val_accuracy: 0.6067\n",
            "Epoch 14/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7743 - accuracy: 0.6905 - val_loss: 1.8400 - val_accuracy: 0.6183\n",
            "Epoch 15/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7704 - accuracy: 0.6942 - val_loss: 1.8820 - val_accuracy: 0.5775\n",
            "Epoch 16/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7636 - accuracy: 0.7001 - val_loss: 1.7974 - val_accuracy: 0.6650\n",
            "Epoch 17/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7603 - accuracy: 0.7042 - val_loss: 1.8230 - val_accuracy: 0.6385\n",
            "Epoch 18/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7555 - accuracy: 0.7091 - val_loss: 1.9350 - val_accuracy: 0.5218\n",
            "Epoch 19/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7530 - accuracy: 0.7114 - val_loss: 1.8315 - val_accuracy: 0.6294\n",
            "Epoch 20/20\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7479 - accuracy: 0.7161 - val_loss: 1.8110 - val_accuracy: 0.6469\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "eiP-wnUWMglD"
      },
      "source": [
        "### Evaluate the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "XpIzozVoMglE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "outputId": "e0ac1c0d-9d1e-4a4a-b97f-fe6cb001dfc4"
      },
      "source": [
        "plt.plot(history.history['accuracy'], label='accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.ylim([0.5, 1])\n",
        "plt.legend(loc='lower right')\n",
        "\n",
        "test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8110 - accuracy: 0.6469\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "LFoHcSBqMglI",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "9257cd88-8e60-4a2c-ddfe-cb0c62409e44"
      },
      "source": [
        "print(test_acc)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.6468999981880188\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "WX8nGyJuMglM",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jzispAQXYsCv",
        "colab_type": "text"
      },
      "source": [
        "# 10"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "NLBUPvEuVkSa",
        "colab": {}
      },
      "source": [
        "#INSERT YOUR MODEL CODE HERE\n",
        "#YOU CAN ALSO CHANGE THE CODE BELOW TO COMPILE AND RUN THE MODEL\n",
        "#setup model\n",
        "model = keras.Sequential([\n",
        "    # keras.layers.Conv2D(kernel_size=3,filters=32,use_bias=False,padding='same',input_shape=(32,32,3)),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Conv2D(kernel_size=3,filters=32,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.MaxPool2D(pool_size=(2, 2),padding='same'),\n",
        "    # keras.layers.Dropout(0.2),\n",
        "    # keras.layers.Conv2D(kernel_size=3,filters=64,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Conv2D(kernel_size=3,filters=64,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.MaxPool2D(pool_size=(2, 2),padding='same'),\n",
        "    # keras.layers.Dropout(0.2),\n",
        "    # keras.layers.Conv2D(kernel_size=3,filters=128,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Conv2D(kernel_size=3,filters=128,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.MaxPool2D(pool_size=(2, 2),padding='same'),\n",
        "    # keras.layers.Dropout(0.2),\n",
        "    # keras.layers.Conv2D(kernel_size=3,filters=256,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Conv2D(kernel_size=3,filters=256,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.MaxPool2D(pool_size=(2, 2),padding='same'),\n",
        "    # keras.layers.Dropout(0.2),\n",
        "\n",
        "    # keras.layers.Flatten(),\n",
        "    \n",
        "    keras.layers.Dense(512),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "\n",
        "    keras.layers.Dense(256),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "\n",
        "    keras.layers.Dense(700),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "    \n",
        "    keras.layers.Dense(10, activation = tf.nn.softmax)    \n",
        "])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "XmBdQFWHVkS8"
      },
      "source": [
        "### Compile and train the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "rCcLXCjCVkS-",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 918
        },
        "outputId": "f721a0bf-c36f-4019-8685-5b9b60b8e28c"
      },
      "source": [
        "model.compile(optimizer='sgd',\n",
        "              loss='categorical_crossentropy', #tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "history = model.fit(train_images, train_labels, epochs=20, \n",
        "                    validation_data=(test_images, test_labels))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-61-11ceae13cf23>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m history = model.fit(train_images, train_labels, epochs=20, \n\u001b[0;32m----> 6\u001b[0;31m                     validation_data=(test_images, test_labels))\n\u001b[0m",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36m_method_wrapper\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    106\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_method_wrapper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    107\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_in_multi_worker_mode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint: disable=protected-access\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 108\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    109\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    110\u001b[0m     \u001b[0;31m# Running inside `run_distribute_coordinator` already.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1096\u001b[0m                 batch_size=batch_size):\n\u001b[1;32m   1097\u001b[0m               \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_train_batch_begin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1098\u001b[0;31m               \u001b[0mtmp_logs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1099\u001b[0m               \u001b[0;32mif\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshould_sync\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1100\u001b[0m                 \u001b[0mcontext\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masync_wait\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    778\u001b[0m       \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    779\u001b[0m         \u001b[0mcompiler\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"nonXla\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 780\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    781\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    782\u001b[0m       \u001b[0mnew_tracing_count\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_tracing_count\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m_call\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    821\u001b[0m       \u001b[0;31m# This is the first call of __call__, so we have to initialize.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    822\u001b[0m       \u001b[0minitializers\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 823\u001b[0;31m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_initialize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madd_initializers_to\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minitializers\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    824\u001b[0m     \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    825\u001b[0m       \u001b[0;31m# At this point we know that the initialization is complete (or less\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m_initialize\u001b[0;34m(self, args, kwds, add_initializers_to)\u001b[0m\n\u001b[1;32m    695\u001b[0m     self._concrete_stateful_fn = (\n\u001b[1;32m    696\u001b[0m         self._stateful_fn._get_concrete_function_internal_garbage_collected(  # pylint: disable=protected-access\n\u001b[0;32m--> 697\u001b[0;31m             *args, **kwds))\n\u001b[0m\u001b[1;32m    698\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    699\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0minvalid_creator_scope\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0munused_args\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0munused_kwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_get_concrete_function_internal_garbage_collected\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2853\u001b[0m       \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2854\u001b[0m     \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2855\u001b[0;31m       \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_define_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2856\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2857\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_maybe_define_function\u001b[0;34m(self, args, kwargs)\u001b[0m\n\u001b[1;32m   3211\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3212\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmissed\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcall_context_key\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3213\u001b[0;31m       \u001b[0mgraph_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_create_graph_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3214\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprimary\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcache_key\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3215\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_create_graph_function\u001b[0;34m(self, args, kwargs, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m   3073\u001b[0m             \u001b[0marg_names\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0marg_names\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3074\u001b[0m             \u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3075\u001b[0;31m             capture_by_value=self._capture_by_value),\n\u001b[0m\u001b[1;32m   3076\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_attributes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3077\u001b[0m         \u001b[0mfunction_spec\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunction_spec\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py\u001b[0m in \u001b[0;36mfunc_graph_from_py_func\u001b[0;34m(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m    984\u001b[0m         \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moriginal_func\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_decorator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munwrap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpython_func\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    985\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 986\u001b[0;31m       \u001b[0mfunc_outputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpython_func\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mfunc_args\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfunc_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    987\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    988\u001b[0m       \u001b[0;31m# invariant: `func_outputs` contains only Tensors, CompositeTensors,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36mwrapped_fn\u001b[0;34m(*args, **kwds)\u001b[0m\n\u001b[1;32m    598\u001b[0m         \u001b[0;31m# __wrapped__ allows AutoGraph to swap in a converted function. We give\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    599\u001b[0m         \u001b[0;31m# the function a weak reference to itself to avoid a reference cycle.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 600\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mweak_wrapped_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__wrapped__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    601\u001b[0m     \u001b[0mweak_wrapped_fn\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mweakref\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mref\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwrapped_fn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    602\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    971\u001b[0m           \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint:disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    972\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"ag_error_metadata\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 973\u001b[0;31m               \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    974\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    975\u001b[0m               \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: in user code:\n\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:806 train_function  *\n        return step_function(self, iterator)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:796 step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:1211 run\n        return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2585 call_for_each_replica\n        return self._call_for_each_replica(fn, args, kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2945 _call_for_each_replica\n        return fn(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:789 run_step  **\n        outputs = model.train_step(data)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:749 train_step\n        y, y_pred, sample_weight, regularization_losses=self.losses)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/compile_utils.py:204 __call__\n        loss_value = loss_obj(y_t, y_p, sample_weight=sw)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/losses.py:149 __call__\n        losses = ag_call(y_true, y_pred)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/losses.py:253 call  **\n        return ag_fn(y_true, y_pred, **self._fn_kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/util/dispatch.py:201 wrapper\n        return target(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/losses.py:1535 categorical_crossentropy\n        return K.categorical_crossentropy(y_true, y_pred, from_logits=from_logits)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/util/dispatch.py:201 wrapper\n        return target(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/backend.py:4687 categorical_crossentropy\n        target.shape.assert_is_compatible_with(output.shape)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/tensor_shape.py:1134 assert_is_compatible_with\n        raise ValueError(\"Shapes %s and %s are incompatible\" % (self, other))\n\n    ValueError: Shapes (None, 1) and (None, 32, 32, 10) are incompatible\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "YySn84wnVkTF"
      },
      "source": [
        "### Evaluate the model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3oJCsWGtXtRd",
        "colab_type": "text"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "raWRV2lz8xLv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras import Sequential\n",
        "from keras.optimizers import SGD,Adam\n",
        "from keras.layers import Conv2D,MaxPooling2D,Dropout,Flatten,Dense"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f--qiqfU8ofE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = keras.Sequential()\n",
        "model.add(Conv2D(32, (3, 3), activation='relu', kernel_initializer='he_uniform', padding='same', input_shape=(32, 32, 3)))\n",
        "model.add(Conv2D(32, (3, 3), activation='relu', kernel_initializer='he_uniform', padding='same'))\n",
        "model.add(MaxPooling2D((2, 2)))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Conv2D(64, (3, 3), activation='relu', kernel_initializer='he_uniform', padding='same'))\n",
        "model.add(Conv2D(64, (3, 3), activation='relu', kernel_initializer='he_uniform', padding='same'))\n",
        "model.add(MaxPooling2D((2, 2)))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Conv2D(128, (3, 3), activation='relu', kernel_initializer='he_uniform', padding='same'))\n",
        "model.add(Conv2D(128, (3, 3), activation='relu', kernel_initializer='he_uniform', padding='same'))\n",
        "model.add(MaxPooling2D((2, 2)))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Flatten())\n",
        "model.add(Dense(128, activation='relu', kernel_initializer='he_uniform'))\n",
        "model.add(Dropout(0.2))\n",
        "model.add(Dense(10, activation='softmax'))\n",
        "# compile model\n",
        "opt = SGD(lr=0.001, momentum=0.9)\n",
        "model.compile(optimizer=opt, loss='categorical_crossentropy', metrics=['accuracy'])\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "A83fEwePVkTF",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "875fae42-88fd-48c8-ef55-d2effdf27c28"
      },
      "source": [
        "plt.plot(history.history['accuracy'], label='accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.ylim([0.5, 1])\n",
        "plt.legend(loc='lower right')\n",
        "\n",
        "# fit model\n",
        "history = model.fit(train_images, train_labels, epochs=100, batch_size=64, validation_data=(test_images, test_labels), verbose=0)\n",
        "\n",
        "test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "ValueError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-82-6a01ed508126>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0;31m# fit model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m \u001b[0mhistory\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_images\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrain_labels\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mepochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m100\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m64\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalidation_data\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest_images\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest_labels\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     10\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     11\u001b[0m \u001b[0mtest_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtest_acc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mevaluate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtest_images\u001b[0m\u001b[0;34m,\u001b[0m  \u001b[0mtest_labels\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36m_method_wrapper\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    106\u001b[0m   \u001b[0;32mdef\u001b[0m \u001b[0m_method_wrapper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    107\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_in_multi_worker_mode\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint: disable=protected-access\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 108\u001b[0;31m       \u001b[0;32mreturn\u001b[0m \u001b[0mmethod\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    109\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    110\u001b[0m     \u001b[0;31m# Running inside `run_distribute_coordinator` already.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)\u001b[0m\n\u001b[1;32m   1096\u001b[0m                 batch_size=batch_size):\n\u001b[1;32m   1097\u001b[0m               \u001b[0mcallbacks\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mon_train_batch_begin\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1098\u001b[0;31m               \u001b[0mtmp_logs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0miterator\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1099\u001b[0m               \u001b[0;32mif\u001b[0m \u001b[0mdata_handler\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshould_sync\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1100\u001b[0m                 \u001b[0mcontext\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0masync_wait\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    778\u001b[0m       \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    779\u001b[0m         \u001b[0mcompiler\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"nonXla\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 780\u001b[0;31m         \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    781\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    782\u001b[0m       \u001b[0mnew_tracing_count\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_tracing_count\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m_call\u001b[0;34m(self, *args, **kwds)\u001b[0m\n\u001b[1;32m    821\u001b[0m       \u001b[0;31m# This is the first call of __call__, so we have to initialize.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    822\u001b[0m       \u001b[0minitializers\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 823\u001b[0;31m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_initialize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwds\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0madd_initializers_to\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0minitializers\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    824\u001b[0m     \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    825\u001b[0m       \u001b[0;31m# At this point we know that the initialization is complete (or less\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36m_initialize\u001b[0;34m(self, args, kwds, add_initializers_to)\u001b[0m\n\u001b[1;32m    695\u001b[0m     self._concrete_stateful_fn = (\n\u001b[1;32m    696\u001b[0m         self._stateful_fn._get_concrete_function_internal_garbage_collected(  # pylint: disable=protected-access\n\u001b[0;32m--> 697\u001b[0;31m             *args, **kwds))\n\u001b[0m\u001b[1;32m    698\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    699\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0minvalid_creator_scope\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0munused_args\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0munused_kwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_get_concrete_function_internal_garbage_collected\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2853\u001b[0m       \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2854\u001b[0m     \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_lock\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2855\u001b[0;31m       \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_maybe_define_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2856\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2857\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_maybe_define_function\u001b[0;34m(self, args, kwargs)\u001b[0m\n\u001b[1;32m   3211\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3212\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmissed\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcall_context_key\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3213\u001b[0;31m       \u001b[0mgraph_function\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_create_graph_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3214\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_cache\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprimary\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mcache_key\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3215\u001b[0m       \u001b[0;32mreturn\u001b[0m \u001b[0mgraph_function\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/function.py\u001b[0m in \u001b[0;36m_create_graph_function\u001b[0;34m(self, args, kwargs, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m   3073\u001b[0m             \u001b[0marg_names\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0marg_names\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3074\u001b[0m             \u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0moverride_flat_arg_shapes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3075\u001b[0;31m             capture_by_value=self._capture_by_value),\n\u001b[0m\u001b[1;32m   3076\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_function_attributes\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3077\u001b[0m         \u001b[0mfunction_spec\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfunction_spec\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py\u001b[0m in \u001b[0;36mfunc_graph_from_py_func\u001b[0;34m(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes)\u001b[0m\n\u001b[1;32m    984\u001b[0m         \u001b[0m_\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moriginal_func\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_decorator\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munwrap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpython_func\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    985\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 986\u001b[0;31m       \u001b[0mfunc_outputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpython_func\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mfunc_args\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mfunc_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    987\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    988\u001b[0m       \u001b[0;31m# invariant: `func_outputs` contains only Tensors, CompositeTensors,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/def_function.py\u001b[0m in \u001b[0;36mwrapped_fn\u001b[0;34m(*args, **kwds)\u001b[0m\n\u001b[1;32m    598\u001b[0m         \u001b[0;31m# __wrapped__ allows AutoGraph to swap in a converted function. We give\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    599\u001b[0m         \u001b[0;31m# the function a weak reference to itself to avoid a reference cycle.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 600\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mweak_wrapped_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__wrapped__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    601\u001b[0m     \u001b[0mweak_wrapped_fn\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mweakref\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mref\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwrapped_fn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    602\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/func_graph.py\u001b[0m in \u001b[0;36mwrapper\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    971\u001b[0m           \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint:disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    972\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"ag_error_metadata\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 973\u001b[0;31m               \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    974\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    975\u001b[0m               \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mValueError\u001b[0m: in user code:\n\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:806 train_function  *\n        return step_function(self, iterator)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:796 step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:1211 run\n        return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2585 call_for_each_replica\n        return self._call_for_each_replica(fn, args, kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/distribute/distribute_lib.py:2945 _call_for_each_replica\n        return fn(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:789 run_step  **\n        outputs = model.train_step(data)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training.py:749 train_step\n        y, y_pred, sample_weight, regularization_losses=self.losses)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/compile_utils.py:204 __call__\n        loss_value = loss_obj(y_t, y_p, sample_weight=sw)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/losses.py:149 __call__\n        losses = ag_call(y_true, y_pred)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/losses.py:253 call  **\n        return ag_fn(y_true, y_pred, **self._fn_kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/util/dispatch.py:201 wrapper\n        return target(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/losses.py:1535 categorical_crossentropy\n        return K.categorical_crossentropy(y_true, y_pred, from_logits=from_logits)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/util/dispatch.py:201 wrapper\n        return target(*args, **kwargs)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/backend.py:4687 categorical_crossentropy\n        target.shape.assert_is_compatible_with(output.shape)\n    /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/tensor_shape.py:1134 assert_is_compatible_with\n        raise ValueError(\"Shapes %s and %s are incompatible\" % (self, other))\n\n    ValueError: Shapes (None, 1) and (None, 10) are incompatible\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "o_Er_kiYVkTJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "72f9e81d-323e-477a-ad93-fe51b4e2b8c1"
      },
      "source": [
        "print(test_acc)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.5834000110626221\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "C84AkXHiVkTN",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WaTnoSbRZQua",
        "colab_type": "text"
      },
      "source": [
        "# 10"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "ctctyOA8YJSU",
        "colab": {}
      },
      "source": [
        "#INSERT YOUR MODEL CODE HERE\n",
        "#YOU CAN ALSO CHANGE THE CODE BELOW TO COMPILE AND RUN THE MODEL\n",
        "#setup model\n",
        "model = keras.Sequential([\n",
        "    # keras.layers.Conv2D(kernel_size=6,filters=64,use_bias=False,padding='same',input_shape=(32,32,3)),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    #keras.layers.Dropout(0.5),\n",
        "    #keras.layers.MaxPool2D(),\n",
        "\n",
        "    keras.layers.Conv2D(kernel_size=6,filters=128,use_bias=False,padding='same',strides=2),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    #keras.layers.Dropout(0.5),\n",
        "    #keras.layers.MaxPool2D(),\n",
        "    \n",
        "    keras.layers.Conv2D(kernel_size=6,filters=256,use_bias=False,padding='same',strides=2),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    #keras.layers.MaxPool2D(),\n",
        "    #keras.layers.Dropout(0.5),\n",
        "    \n",
        "    # keras.layers.Conv2D(kernel_size=24,filters=32,use_bias=False,padding='same',strides=2),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    \n",
        "    keras.layers.Flatten(),\n",
        "    \n",
        "    # keras.layers.Dense(128),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Dropout(0.5),\n",
        "\n",
        "    # keras.layers.Dense(200),\n",
        "    # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    # keras.layers.Activation('relu'),\n",
        "    # keras.layers.Dropout(0.5),\n",
        "\n",
        "    keras.layers.Dense(700),\n",
        "    keras.layers.BatchNormalization(center=True,scale=False),\n",
        "    keras.layers.Activation('relu'),\n",
        "    keras.layers.Dropout(0.5),\n",
        "    \n",
        "    keras.layers.Dense(10, activation = tf.nn.softmax)    \n",
        "])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "iTPJUudEYJSl"
      },
      "source": [
        "### Compile and train the model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "Elw1Zp8nYJSo",
        "colab": {}
      },
      "source": [
        "model.compile(optimizer=keras.optimizers.Adam(),\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "history = model.fit(train_images, train_labels, epochs=15, \n",
        "                    validation_data=(test_images, test_labels))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "oLy7VvAbYJSy"
      },
      "source": [
        "### Evaluate the model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "zm42PYhuYJS0"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "DYjP2S4rYJS1",
        "colab": {}
      },
      "source": [
        "plt.plot(history.history['accuracy'], label='accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.ylim([0.5, 1])\n",
        "plt.legend(loc='lower right')\n",
        "\n",
        "test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "aT-gALjbYJS7",
        "colab": {}
      },
      "source": [
        "print(test_acc)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AY_iqoJosfib",
        "colab_type": "text"
      },
      "source": [
        "# For Loop"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "1HpCpCb2YJTC",
        "colab": {}
      },
      "source": [
        "import pandas as pd\n",
        "from tqdm import tqdm"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "G9vleVZfd1IO",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "84c1dad6-0fd5-4fef-fff3-21028d1d393c"
      },
      "source": [
        "meta = []\n",
        "for ks in [3,6]:\n",
        "    for f in [96,128,256]:\n",
        "        for s in [2,3]:\n",
        "            model = keras.Sequential([\n",
        "                # keras.layers.Conv2D(kernel_size=6,filters=64,use_bias=False,padding='same',input_shape=(32,32,3)),\n",
        "                # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                # keras.layers.Activation('relu'),\n",
        "                #keras.layers.Dropout(0.5),\n",
        "                #keras.layers.MaxPool2D(),\n",
        "\n",
        "                keras.layers.Conv2D(kernel_size=ks,filters=f,use_bias=False,padding='same',strides=s),\n",
        "                keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                keras.layers.Activation('relu'),\n",
        "                keras.layers.MaxPool2D(),\n",
        "                \n",
        "                keras.layers.Conv2D(kernel_size=ks,filters=f,use_bias=False,padding='same',strides=s),\n",
        "                keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                keras.layers.Activation('relu'),\n",
        "                keras.layers.MaxPool2D(),\n",
        "                \n",
        "                \n",
        "                keras.layers.Conv2D(kernel_size=ks,filters=f,use_bias=False,padding='same',strides=s),\n",
        "                keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                keras.layers.Activation('relu'),\n",
        "                \n",
        "                keras.layers.Flatten(),\n",
        "                \n",
        "                # keras.layers.Dense(128),\n",
        "                # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                # keras.layers.Activation('relu'),\n",
        "                # keras.layers.Dropout(0.5),\n",
        "\n",
        "                # keras.layers.Dense(200),\n",
        "                # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                # keras.layers.Activation('relu'),\n",
        "                # keras.layers.Dropout(0.5),\n",
        "\n",
        "                keras.layers.Dense(700),\n",
        "                keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                keras.layers.Activation('relu'),\n",
        "                keras.layers.Dropout(0.5),\n",
        "                \n",
        "                keras.layers.Dense(10, activation = tf.nn.softmax)    \n",
        "            ])\n",
        "\n",
        "            model.compile(optimizer=keras.optimizers.Adam(),\n",
        "                        loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "                        metrics=['accuracy'])\n",
        "\n",
        "            history = model.fit(train_images, train_labels, epochs=15, \n",
        "                                validation_data=(test_images, test_labels))\n",
        "\n",
        "            plt.plot(history.history['accuracy'], label='accuracy')\n",
        "            plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n",
        "            plt.xlabel('Epoch')\n",
        "            plt.ylabel('Accuracy')\n",
        "            plt.ylim([0.5, 1])\n",
        "            plt.legend(loc='lower right')\n",
        "            plt.show()\n",
        "\n",
        "            test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)\n",
        "            meta.append([ks,f,s,test_acc])\n",
        "            print(\"---------------------------------\",ks,f,s,test_acc,\"---------------------------\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 2.0302 - accuracy: 0.4252 - val_loss: 2.0269 - val_accuracy: 0.4296\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9474 - accuracy: 0.5090 - val_loss: 2.0681 - val_accuracy: 0.3856\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9058 - accuracy: 0.5521 - val_loss: 2.1080 - val_accuracy: 0.3456\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8831 - accuracy: 0.5747 - val_loss: 2.0118 - val_accuracy: 0.4446\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8672 - accuracy: 0.5914 - val_loss: 2.0549 - val_accuracy: 0.4034\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8489 - accuracy: 0.6097 - val_loss: 1.8801 - val_accuracy: 0.5774\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8363 - accuracy: 0.6225 - val_loss: 1.9085 - val_accuracy: 0.5490\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8239 - accuracy: 0.6354 - val_loss: 1.8516 - val_accuracy: 0.6073\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8212 - accuracy: 0.6378 - val_loss: 2.1216 - val_accuracy: 0.3373\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8113 - accuracy: 0.6476 - val_loss: 1.8632 - val_accuracy: 0.5947\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8027 - accuracy: 0.6566 - val_loss: 1.8470 - val_accuracy: 0.6122\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7936 - accuracy: 0.6653 - val_loss: 1.8798 - val_accuracy: 0.5784\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7867 - accuracy: 0.6725 - val_loss: 1.8353 - val_accuracy: 0.6243\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7817 - accuracy: 0.6777 - val_loss: 1.8977 - val_accuracy: 0.5612\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7748 - accuracy: 0.6850 - val_loss: 1.9477 - val_accuracy: 0.5111\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.9477 - accuracy: 0.5111\n",
            "--------------------------------- 3 96 2 0.5110999941825867 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 2.0532 - accuracy: 0.4021 - val_loss: 2.0459 - val_accuracy: 0.4093\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9806 - accuracy: 0.4753 - val_loss: 1.9715 - val_accuracy: 0.4845\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9504 - accuracy: 0.5065 - val_loss: 2.1400 - val_accuracy: 0.3157\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9310 - accuracy: 0.5271 - val_loss: 1.9423 - val_accuracy: 0.5150\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9147 - accuracy: 0.5431 - val_loss: 2.0477 - val_accuracy: 0.4099\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8993 - accuracy: 0.5583 - val_loss: 1.9657 - val_accuracy: 0.4908\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8945 - accuracy: 0.5635 - val_loss: 1.9234 - val_accuracy: 0.5341\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8829 - accuracy: 0.5756 - val_loss: 1.9663 - val_accuracy: 0.4909\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8702 - accuracy: 0.5876 - val_loss: 1.9595 - val_accuracy: 0.4989\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8687 - accuracy: 0.5896 - val_loss: 1.8900 - val_accuracy: 0.5688\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8634 - accuracy: 0.5951 - val_loss: 1.9282 - val_accuracy: 0.5298\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8559 - accuracy: 0.6036 - val_loss: 1.9132 - val_accuracy: 0.5451\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8516 - accuracy: 0.6070 - val_loss: 1.9527 - val_accuracy: 0.5048\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8452 - accuracy: 0.6141 - val_loss: 1.9084 - val_accuracy: 0.5496\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8420 - accuracy: 0.6172 - val_loss: 1.8810 - val_accuracy: 0.5779\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8810 - accuracy: 0.5779\n",
            "--------------------------------- 3 96 3 0.5778999924659729 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 2.0253 - accuracy: 0.4295 - val_loss: 2.0864 - val_accuracy: 0.3695\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9382 - accuracy: 0.5192 - val_loss: 2.1177 - val_accuracy: 0.3373\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9049 - accuracy: 0.5520 - val_loss: 1.9371 - val_accuracy: 0.5213\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8780 - accuracy: 0.5789 - val_loss: 1.8914 - val_accuracy: 0.5671\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8574 - accuracy: 0.6011 - val_loss: 1.9396 - val_accuracy: 0.5197\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8397 - accuracy: 0.6193 - val_loss: 1.8795 - val_accuracy: 0.5770\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8317 - accuracy: 0.6267 - val_loss: 1.9161 - val_accuracy: 0.5420\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8168 - accuracy: 0.6426 - val_loss: 1.8650 - val_accuracy: 0.5937\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8127 - accuracy: 0.6462 - val_loss: 1.8902 - val_accuracy: 0.5680\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8022 - accuracy: 0.6571 - val_loss: 1.8688 - val_accuracy: 0.5893\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7917 - accuracy: 0.6679 - val_loss: 1.8893 - val_accuracy: 0.5693\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7840 - accuracy: 0.6754 - val_loss: 1.8792 - val_accuracy: 0.5797\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7757 - accuracy: 0.6842 - val_loss: 1.8635 - val_accuracy: 0.5950\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7712 - accuracy: 0.6883 - val_loss: 1.8367 - val_accuracy: 0.6219\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.7625 - accuracy: 0.6971 - val_loss: 1.8423 - val_accuracy: 0.6166\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8423 - accuracy: 0.6166\n",
            "--------------------------------- 3 128 2 0.616599977016449 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 2.0487 - accuracy: 0.4065 - val_loss: 2.0237 - val_accuracy: 0.4326\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9778 - accuracy: 0.4780 - val_loss: 2.0705 - val_accuracy: 0.3847\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9499 - accuracy: 0.5068 - val_loss: 2.1156 - val_accuracy: 0.3400\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9295 - accuracy: 0.5281 - val_loss: 1.9481 - val_accuracy: 0.5095\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9108 - accuracy: 0.5466 - val_loss: 1.9894 - val_accuracy: 0.4668\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8992 - accuracy: 0.5591 - val_loss: 1.9587 - val_accuracy: 0.4992\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8899 - accuracy: 0.5683 - val_loss: 1.9518 - val_accuracy: 0.5056\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8814 - accuracy: 0.5765 - val_loss: 1.9239 - val_accuracy: 0.5346\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8709 - accuracy: 0.5878 - val_loss: 2.1057 - val_accuracy: 0.3524\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8625 - accuracy: 0.5965 - val_loss: 1.8972 - val_accuracy: 0.5615\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8554 - accuracy: 0.6031 - val_loss: 1.9234 - val_accuracy: 0.5348\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8511 - accuracy: 0.6077 - val_loss: 1.9690 - val_accuracy: 0.4894\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8411 - accuracy: 0.6183 - val_loss: 1.9386 - val_accuracy: 0.5200\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8416 - accuracy: 0.6169 - val_loss: 1.9200 - val_accuracy: 0.5377\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8339 - accuracy: 0.6251 - val_loss: 1.9042 - val_accuracy: 0.5558\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEKCAYAAAAfGVI8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de3xU1bn/8c+TCwSSAAkk4ZIQELkLCKR4a72gWKwU2loKaK21Vk971GO1PWptT6XW09pT+9PaWltsvVCttvXSY22rFcWqR7AEVETuIpdwSwgkJEDuz++PPcSICUwgk0nY3/frNa+ZvWfP3s8MYT17r7X2WubuiIhIeCXEOwAREYkvJQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQi1kiMLMHzKzYzFa08L6Z2T1mtt7MlpvZhFjFIiIiLYvlFcFDwNTDvH8BMDTyuAq4L4axiIhIC2KWCNz9FWD3YTaZAcz3wGKgl5n1i1U8IiLSvKQ4HnsAsKXJclFk3fZDNzSzqwiuGkhNTZ04YsSIdglQROR4sXTp0l3untXce/FMBFFz93nAPICCggIvLCyMc0QiIp2LmW1q6b149hraCuQ1Wc6NrBMRkXYUz0TwDPClSO+hU4Fyd/9ItZCIiMRWzKqGzOwx4Gygj5kVAbcCyQDu/ivgb8CngPXAfuDyWMUiIiIti1kicPc5R3jfgatjdXwREYmO7iwWEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJuZgmAjObamZrzGy9md3czPv5ZvaimS03s5fNLDeW8YiIyEfFLBGYWSJwL3ABMAqYY2ajDtnsTmC+u48FbgN+FKt4RESkebG8IpgErHf3De5eAzwOzDhkm1HAS5HXC5t5X0REYiyWiWAAsKXJclFkXVNvA5+LvP4skG5mvQ/dkZldZWaFZlZYUlISk2BFRMIq3o3F3wLOMrM3gbOArUD9oRu5+zx3L3D3gqysrPaOUUTkuJYUw31vBfKaLOdG1jVy921ErgjMLA24yN3LYhiTiIgcIpZXBEuAoWY22My6ALOBZ5puYGZ9zOxgDN8GHohhPCIi0oyYJQJ3rwOuAZ4HVgF/dPd3zew2M5se2exsYI2ZrQVygP+OVTwiItI8c/d4x9AqBQUFXlhYGO8wREQ6FTNb6u4Fzb0X78ZiERGJMyUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZBTIhARCTklAhGRkFMiEBEJOSUCEZGQUyIQEQk5JQIRkZCLaSIws6lmtsbM1pvZzc28P9DMFprZm2a23Mw+Fct4RETko2KWCMwsEbgXuAAYBcwxs1GHbPZd4I/uPh6YDfwyVvGIiEjzYnlFMAlY7+4b3L0GeByYccg2DvSIvO4JbIthPCIi0oxYJoIBwJYmy0WRdU3NBb5oZkXA34Brm9uRmV1lZoVmVlhSUhKLWEVEQivejcVzgIfcPRf4FPA7M/tITO4+z90L3L0gKyur3YMUETmeHTERmNmnmyuco7AVyGuynBtZ19QVwB8B3H0RkAL0OYpjiYjIUYqmgJ8FrDOz/zGzEa3Y9xJgqJkNNrMuBI3BzxyyzWbgXAAzG0mQCFT3IyLSjo6YCNz9i8B44D3gITNbFKmzTz/C5+qAa4DngVUEvYPeNbPbzGx6ZLNvAlea2dvAY8CX3d2P4fuIiEgrWbTlrpn1Bi4FvkFQsJ8I3OPuP49deB9VUFDghYWF7XlIEZFOz8yWuntBc+9F00Yw3cyeBl4GkoFJ7n4BMI7gjF5ERDqxpCi2uQi4y91fabrS3feb2RWxCUtERNpLNIlgLrD94IKZdQNy3H2ju78Yq8BERKR9RNNr6E9AQ5Pl+sg6ERE5DkSTCJIiQ0QAEHndJXYhiYhIe4omEZQ06e6Jmc0AdsUuJBERaU/RtBF8DXjUzH4BGMH4QV+KaVQiItJujpgI3P094FQzS4ssV8Y8KhERaTfRXBFgZhcCo4EUMwPA3W+LYVwiItJOormh7FcE4w1dS1A1NBPIj3FcIiLSTqJpLD7d3b8E7HH37wOnAcNiG5aIiLSXaBJBVeR5v5n1B2qBfrELSURE2lM0bQR/MbNewE+AZQTTS94f06hERKTdHDYRRCakedHdy4AnzexZIMXdy9slOhERibnDVg25ewNwb5PlaiUBEZHjSzRtBC+a2UV2sN+oiIgcV6JJBP9GMMhctZntNbMKM9sb47hERKSdRHNn8WGnpBQRkc7tiInAzM5sbv2hE9WIiEjnFE330f9s8joFmAQsBSbHJCIREWlX0VQNfbrpspnlAXfHLCIREWlX0TQWH6oIGNnWgYiISHxE00bwc4K7iSFIHCcT3GEsIiLHgWjaCAqbvK4DHnP3/4tRPCIi0s6iSQRPAFXuXg9gZolm1t3d98c2NBERaQ9R3VkMdGuy3A1YEJtwRESkvUWTCFKaTk8Zed09diGJiEh7iiYR7DOzCQcXzGwicCB2IYmISHuKpo3gG8CfzGwbwVSVfQmmrhQRkeNANDeULTGzEcDwyKo17l4b27BERKS9RDN5/dVAqruvcPcVQJqZ/XvsQxMRkfYQTRvBlZEZygBw9z3AlbELSURE2lM0iSCx6aQ0ZpYIdIldSCIi0p6iaSx+DviDmf06svxvwN9jF5KIiLSnaBLBTcBVwNciy8sJeg6JiMhx4IhVQ5EJ7N8ANhLMRTAZWBXNzs1sqpmtMbP1ZnZzM+/fZWZvRR5rzaysuf2IiEjstHhFYGbDgDmRxy7gDwDufk40O460JdwLTCEYunqJmT3j7isPbuPu1zfZ/lpg/FF8BxEROQaHuyJYTXD2P83dP+7uPwfqW7HvScB6d9/g7jXA48CMw2w/B3isFfsXEZE2cLhE8DlgO7DQzO43s3MJ7iyO1gBgS5Plosi6jzCzfGAw8FIL719lZoVmVlhSUtKKEERE5EhaTATu/md3nw2MABYSDDWRbWb3mdn5bRzHbOCJg0NdNxPLPHcvcPeCrKysNj60iEi4RdNYvM/dfx+ZuzgXeJOgJ9GRbAXymiznRtY1ZzaqFhIRiYtWzVns7nsiZ+fnRrH5EmComQ02sy4Ehf0zh24UGccoA1jUmlhERKRtHM3k9VFx9zrgGuB5gu6mf3T3d83sNjOb3mTT2cDj7u7N7UdERGIrmhvKjpq7/w342yHrvnfI8txYxiAiIocXsysCERHpHJQIRERCTolARCTklAhEREJOiUBEJOSUCEREQk6JQEQk5JQIRERCTolARCTklAhEREJOiUBEJOSUCEREQk6JQEQk5JQIRERCTolARCTklAhEREJOiUBEJOSUCEREQk6JQEQk5JQIRERCTolARCTklAhEREJOiUBEJOSUCEREQk6JQEQk5JLiHYCIiLSstr6BTaX7WF9cych+Pcjvndrmx1AiEBHpAKpq69lQso91xRW8V1zJushj46591DU4ALd+ehSXnzG4zY+tRCAi0o4qq+tYX1zJ+uJK1hVXsH5nJetLKtm8ez8elPckGAzqncqQ7DTOH5XD0Jw0TsxKZ0h2218NgBKBiIRQRVUt72wtZ1tZFcmJRtekBLokJZCcmECXxOB1l6TmXycnJtA1KQEzO+wx9uyrYX1JJet2Nin0iyvZXl7VuE2XxAROyErlpAE9+ez4AZyYncbQ7HQG9elO16TEWP8MjZQIROS4VlvfwOrtFbxVVMbbW4LH+pLKxrPvo5WcaB9KDk0Txq7KanZV1jRu271LIkOy0jjthN6cmJPGiVlpDM1JJy+jG0mJ8e+zo0QgIscNd2dT6X7eLirjrUihv2LbXmrqGgDITO3CyXm9mDa2P+PyejKodyp1DU5NXQM19Q3U1DVQG3mujqyrPeS96rqGxu2bvtd0H+NyewXVOdnBo3/PbiQkHP4KIp6UCESk0yqtrI4U+uXB2X5RGWX7awFISU5gzICefOnUfMbl9eLkvF7kZnQ7YpVOGCkRiEincKCmnhXbggL/rUihv2X3ASBoXB2Wk84nR/VtLPSH5aR1iGqXzkCJQETiyt2pqK5jV0U1JRVB3XpJRRW7KmvYVRms21p2gHXFldRHulEO6NWNcXk9ufTUfMbl9uKkAT1J7ari7GjF9Jczs6nAz4BE4Dfufkcz23wBmAs48La7XxzLmEQk9tydyuq6xoL9YIEeNKIGr0sqa4LCv7K6sQ6/qcQEIzO1C1lpXenXM4Upo3IYl9uLsXk9yU5PicO3On7FLBGYWSJwLzAFKAKWmNkz7r6yyTZDgW8DZ7j7HjPLjlU8ItJ2GhqcnRVVbC7dz+bdHzy27N7Pzr1BYV/dTOGeYNA7rSt90rrSJ60LQ7JSyYosZ6VH1qcHhX9G9y4duoH1eBLLK4JJwHp33wBgZo8DM4CVTba5ErjX3fcAuHtxDOMRkVbYV133oQK+aYFftPsANfUfFPSJCUb/XinkZXTnlMGZ9EnvGhTw6V0+VMhndO9Cogr3DieWiWAAsKXJchFwyiHbDAMws/8jqD6a6+7PHbojM7sKuApg4MCBMQlWJGzqG5yde6uaL+xL91O6r+ZD26enJJHfuzsj+qYzZVQOAzO7Nz769+pGshpmO614t64kAUOBs4Fc4BUzG+PuZU03cvd5wDyAgoKCY7wNRCRc6hucLbv3s3pHBWt3VrBmZwVrd1SwqXR/s2f1AzO7c/7oHPKaFPT5man07J4cx28hsRTLRLAVyGuynBtZ11QR8Ia71wLvm9lagsSwJIZxiRyX3J2de6sbC/qDBf+64gqqaj8o8AdmdmdYTjqTR2QzsLfO6iW2iWAJMNTMBhMkgNnAoT2C/gzMAR40sz4EVUUbYhiTyHGhbH8Naz50hl/Jmp0VlB+obdwmO70rw/umc8kp+QzPSWd433ROzE5TN0v5iJj9Rbh7nZldAzxPUP//gLu/a2a3AYXu/kzkvfPNbCVQD/ynu5fGKiaRzqaqtj4o7COF/sGz/J17qxu3SU9JYkTfdKaN7cfwvukMywkemald4hi5dCbmxzryUjsrKCjwwsLCeIchEhPFFVUs27SHpZHHiq17G+vxuyYlMDQnjWE56YyIFPjD+6bTt0dKXIdNqK2tpaioiKqqqiNvLDGXkpJCbm4uyckfbtMxs6XuXtDcZ3SNKBIn9Q3Omh0VLN28p7Hw37x7PwBdkhIYO6Anl58xiJPzejGiXw8GZnbvkF0vi4qKSE9PZ9CgQRrHJ87cndLSUoqKihg8OPoJbJQIRNpJ+YFa3tpSxtJNQcH/5uY97KupByArvSsF+Rl86bR8JuRnMLp/j3Ydj/5YVFVVKQl0EGZG7969KSkpadXnlAhEYsDd2Vi6v7GKZ9mmPawtrsA9uLt2ZL8eXDQxl4n5GUwYmNHpR8XszLEfb47m30KJQKQNlO+vZc3OCpZt/qDgP3hDVnpKEhMGZnDh2H5MzM9gXF4v0tRzRzoQ/TWKtEJpZXXjpOLrd1Y0TkVYXPFBL54T+qRyzohsJuZnMDE/gxOz0jRmjnRoSgQih3B3iiuqWbczmGc2KPSDCcZ3Nxl2IbVLIifmpHPmsCyGZqcxNCeNcbm96J3WNY7RSyzV1dWRlHT8FZvH3zcSiVJDg7O17ADrS4KCvmmhX1Fd17hdz27JDM1O45OjczgxO72x0I93t82O6Pt/eZeV2/a26T5H9e/BrZ8efcTtPvOZz7Blyxaqqqq47rrruOqqq3juuee45ZZbqK+vp0+fPrz44otUVlZy7bXXUlhYiJlx6623ctFFF5GWlkZlZSUATzzxBM8++ywPPfQQX/7yl0lJSeHNN9/kjDPOYPbs2Vx33XVUVVXRrVs3HnzwQYYPH059fT033XQTzz33HAkJCVx55ZWMHj2ae+65hz//+c8AvPDCC/zyl7/k6aefbtPf6FgpEUgo7N5Xw6rte1m5bS+rduxl3c5K1hdXcqC2vnGbPmldGZqdxmcnDGica3Zodjp90rqowO8EHnjgATIzMzlw4AAf+9jHmDFjBldeeSWvvPIKgwcPZvfu3QD84Ac/oGfPnrzzzjsA7Nmz54j7Lioq4vXXXycxMZG9e/fy6quvkpSUxIIFC7jlllt48sknmTdvHhs3buStt94iKSmJ3bt3k5GRwb//+79TUlJCVlYWDz74IF/5yldi+jscDSUCOa7UNzgbS/cFBf724LFy+94P3Ymb06Mrw3LSmTNp4AcTjGelkaE7cY9ZNGfusXLPPfc0nmlv2bKFefPmceaZZzb2p8/MzARgwYIFPP74442fy8jIOOK+Z86cSWJi0J23vLycyy67jHXr1mFm1NbWNu73a1/7WmPV0cHjXXrppTzyyCNcfvnlLFq0iPnz57fRN247SgTSaVVW17G6SWG/cnsFa3bsbRxgLSnBODE7jTOG9GFkvx6M6t+Dkf16aOiF49DLL7/MggULWLRoEd27d+fss8/m5JNPZvXq1VHvo+lV36F3Saempja+/q//+i/OOeccnn76aTZu3MjZZ5992P1efvnlfPrTnyYlJYWZM2d2yDaGjheRyCHcg7r8VdsrPjjT37GXTaX7G7fp1T2ZkX17cPGk/EiBHwyw1lluymp3y34HlTtg0r9BSo94R3PMysvLycjIoHv37qxevZrFixdTVVXFK6+8wvvvv99YNZSZmcmUKVO49957ufvuu4GgaigjI4OcnBxWrVrF8OHDefrpp0lPT2/xWAMGDADgoYcealw/ZcoUfv3rX3POOec0Vg1lZmbSv39/+vfvz+23386CBQti/lscDSUC6XD2VtWy5P3dvPH+bt7eUsaq7XvZWxU03prBoN6pnNS/JzMn5jae6avhthUaGuCF78GB3bD4PvjEt+BjV0BS5+3tNHXqVH71q18xcuRIhg8fzqmnnkpWVhbz5s3jc5/7HA0NDWRnZ/PCCy/w3e9+l6uvvpqTTjqJxMREbr31Vj73uc9xxx13MG3aNLKysigoKGhsOD7UjTfeyGWXXcbtt9/OhRde2Lj+q1/9KmvXrmXs2LEkJydz5ZVXcs011wBwySWXUFJSwsiRI9vl92gtDToncVd+oJbCjbtZvKGUxRt28+62chocuiQmMHpAD0b169FY4A/PSdcwyseqeDX88hQ47RrYuQI2vAw98+CcW2DsLEho3VXUqlWrOmwB11Fcc801jB8/niuuuOLoduAOtfshsSskHvnvv7l/Ew06Jx1K+YHgjH/xhlIWv1/Ku9v24h4MtDY+rxfXTh7KqSf0ZvzAXqQkq2qnzW1+PXgu+Ar0HgLvLYQFc+HPX4fXfw7nfg+GTQ0uv+SYTZw4kdTUVH7605+2/sN11XBgD+zfDfXV0GMApGW3eYxKBBJz5ftr+VfjGX8pK7d/UPBPGNiL684NCv6T81Twt4tNr0NqNmSeECwPOQcGnwWr/hde/AE8NhvyToXz5kL+afGM9LiwdOnS1n2goQ4OlAVVdzX7gnVd0oIE0K1X2weIEoHEQNn+Gv71/m4WbwgK/1U7Pij4Jw7M4BvnDuPUEzIZp4I/PjYtgvzTP3zGn5AAoz8LI6bBm7+Dl38MD04NrgzO/R7kxK9baCh4A1TtDQr/qr2AQ1IKpPeDbpmQFNuebkoEcsz2Vdfx2vpdLN5QyhsbdjcW/F2TEpiYn8H15w3jlMEq+DuEss2wtwjy/6P59xOTgyqjsbPhjV/Ba3fDfWcEbQfn3AIZ+e0b7/HMPTjjP7AneHg9JCRBap+g8E/u1m7Vc0oEclQaGpzF75fy5NKt/H3FdvbX1NM1KYGCQRnccN4wTjmhN+Pyeqr7ZkezaVHwPPAIVT5dusMnboCJX4bX7oJ/zYN3n4KCK+DMbwWFlRyduirYvyc4+6+vARKgW8+g8O+aHpe2GSUCaZX3d+3jqWVFPLVsK1vLDpDeNYnp4/oz4+QBTMjvpYK/o9v8OnTtEX1VT/dMOP8HcMrX4J93wL9+HVQdnX4tnHZ1UHDJkdXXQVWk0bc2cv9Ll3RI7wspvVrdU6utKRHIEZUfqOWvy7fz5LIilm7aQ4LBx4dmcePU4XxydF9V93Qmm16Hgae2vuDpOQCm/xxOuxZeug1e/hH863448z+hx5mxibWza2iA6vKg8K+uIKj37wY9+kO3DEjsOHe4KxFIs+rqG3h1/S6eXFrEP1bupKaugaHZadx8wQg+O34AOT1S4h2itNa+XbBrLYybc/T7yBoGsx6BoqWw4FZ47ia44CnY3zco3DpBl9Omo4y2OXeoqYzU+5dF6v2TITUruLpK7hab4x4jJQL5kDU7KnhyWRFPv7mVkopqMronM+djeVw0MZcxA3rq7t3ObHOkfSD/9GPfV+5EuOwv8N6LsDsByjZBZXFQdVSyBmjDv5O+Y+CCO9puf23NHeoOROr990BDLVhCUOXTLaPFev+ONLdBx4hC4qq0sppn3t7Gk8uKWLF1L0kJxjkjsrloQi6TR2TTJSkh3iF2bFXl8M//gXefDgrH3kPiHVHzNi0K7kztP75t9mcGJ54HK1dCr75QsT34LeqqI8NVtM9Jw80330xeXh5XX301AHPnziUpKYmFCxeyZ88eamtruf3225kxY8YR91VZWcmMGTOa/dz8+fO58847MTPGjh3L7x78DTs3reVr117Pho2bAbjvp/9N/0FDmTbzElasWAHAnXfeSWVlJXPnzm0cDO+1115jzpw5DBs2jNtvv52amhp69+7No48+Sk5OTrNzJpSXl7N8+fLGMZLuv/9+Vq5cyV133XXsP6K7d6rHxIkTXY5ddW29//2d7f7Vh5f4kG//1fNvetYvvOcVf+C1Db6roire4XUO9XXuhQ+5//gE91t7ut/aw/21u+MdVct+fZb7Axe0+W5XrlwZvGiody/f6r51mfuu9cHv0w6WLVvmZ555ZuPyyJEjffPmzV5eXu7u7iUlJT5kyBBvaGhwd/fU1NQW91VbW9vs51asWOFDhw71kh3b3CuLvXTNG+5bl/kXPj3F7/rBt90ri72uusrLysr8/fff99GjRzfu8yc/+Ynfeuut7u5+1lln+de//vXG93bv3t0Y1/333+833HCDu7vfeOONft11131ou4qKCj/hhBO8pqbG3d1PO+00X758ebPfo/HfpAmg0FsoV3VFECLuzoqte3li6RaeeXsbe/bXkpXela98fDAXTchleF/1AInapkXw9xthx/KgK+bUJ+Hpr8F7L8EZ18U7uo+qroDtb8Mnvhm7Y1hC0BCamAzlRVD6HvQ+IegbH0Pjx4+nuLiYbdu2UVJSQkZGBn379uX666/nlVdeISEhga1bt7Jz50769u172H25O7fccsuHP7d9Gy899xdmTjuPPvU7odzJ7JUO3TJ4adGbzP/TX6BrVxKBnl26HnGim1mzZjW+LioqYtasWWzfvp2amprGuRNamjNh8uTJPPvss4wcOZLa2lrGjBlzlL/ahykRhEBVbT1/Xb6dhxdtZHlROV2SEpgyKofPT8jlE0P7kJSoqp+olRcFI3eueDIY9+Wi38JJFwXVJEMmw5LfQM3+oB9+R7LlX8Hdq0e6f6AtpGYFhf+eTbBrXVBVFuMeMjNnzuSJJ55gx44dzJo1i0cffZSSkhKWLl1KcnIygwYN+sgcA81p/FzhEpIbqhg0bDRVRW8Hdf/1dZCW9cHNXi1ISkqioaGhcflwcxtce+213HDDDUyfPp2XX36ZuXPnHja+r371q/zwhz9kxIgRXH755Uf8PtFSCXAc215+gJ88v5oz7niJb/7pbfZV1/H96aNZcst53HvxBM4Zka0kEK3aA8GwCz8vgNV/hbNugmsKYcznP2gIPHFyMDDYptfjG2tzNi8KztjzJrXP8bplBAmgviZIBrVHLoSPxaxZs3j88cd54oknmDlzJuXl5WRnZ5OcnMzChQvZtGnTkXfiTnlpMdk9u5O8ey0L//IHNm3ZCik9mXzh5/nT31+mtDYFkrs1Tnt57rnnct999wFQX19PeXk5OTk5FBcXU1paSnV1Nc8++2yLh2w6t8HDDz/cuP7gnAkHHbzKOOWUU9iyZQu///3vmTPnGHp/HUKlwHHG3Vm8oZSvP7KUj/94Ib98+T0m5GfwyBWnsOCGs7js9EH07J4c7zA7D/egEfgXH4OXfwjDPgnXLAmGWzj0rH/g6UFj7HsvxSfWw9m0CPqObd8bwLqmQ++hwZXIrrUfDKAWA6NHj6aiooIBAwbQr18/LrnkEgoLCxkzZgzz589nxIgRLX+49gDs3QbFK7nkkwUULl3GmMmfZ/5f/hl8rscARo//GN/5znc466yzGDduHDfccAMAP/vZz1i4cCFjxoxh4sSJrFy5kuTkZL73ve8xadIkpkyZcthjz507l5kzZzJx4kT69Pngbu3vfve77Nmzh5NOOolx48axcOHCxve+8IUvcMYZZ0Q1xWa0NB/BceJATT1/fmsrD7++kdU7KujZLZnZH8vji6fmk5fZwaopOosd78Dfb4ZNr0FOpAvjoI8f/jPzZ0DFTrh6cfvEGI26arhjYDA8xNQftvnujzgfQV01lK4PRtXMGNwxZkRzD6p7KouDrp8QJK5umZDSM+53+h7OtGnTuP766zn33HNb3EbzEYTM5tL9/G7xRv6wZAt7q+oY2a8HP75oDNPHDaBbl477x9yh7SuFhbfD0oeCvuDT7oIJl0VXOAyZHLQhlG8N7sbtCLa9GYxvE68hpZO6Qp9hQePx7g3Qa2Bwc1W81B4I2npqKiN3+uYGwzsnduwr5bKyMiZNmopNM2cAAArxSURBVMS4ceMOmwSOhhJBJ+TuvLpuF/MXbeTF1cUkmDH1pL58+fRBFORn6Kavo1VfGzT2vvwjqK4M5vM9+6agvjtaQ84NEsGGhTD+i7GLtTUOtlm0R0NxSxKToc/QIBGUbQquDmIwwcphNdQH8zRXlvDO6ve49Bu3fqhHU9euXXnjjTfaN6ZW6NWrF2vXro3JvpUIOpGKqlqeWraVhxdtZEPJPvqkdeHac07k4lPy6dtTQz4ck/degue+DSWr4YRzYOodkH2YeuWW5IwOJn1576WOkwg2LwrOyGM4Yqi7H/kEJCERMocEiWDv1uAO3PT+7TMsxYGy4Jj1NdAtkzFnTeet5RfF/rhxcDTV/UoEncB7JZXMf30jTy7bSmV1HSfn9eKuWeP41Jh+rR/ts6EhmIREArs3wPPfgTV/C+qvZz8Gwy84+sLpYDfSdf/oGL91Qz1sfgNGfyZmh0hJSaG0tJTevXtHkQwSIGNQUDVTWRx0yeyVF/RoioW66uBY1XuDiV56D4WuabE5Vgfg7pSWlpKS0roTQyWCDqq+wVm4upiHF23k1XW76JKYwLSx/bjs9EGMyzvK6ep2rIAHPgkfvz64sSjMVUjVFfDKnbD4l0Ef9/O+D6d+PTI0wjEaMhmWPw7b34IBE459f8eieGUwAmb+GTE7RG5uLkVFRZSUlLTug1XVULUakjdB995tmwzcg3/j6nLAggbqLl2hdEvbHaODSklJITc3t1WfUSLoYNydF1bu5Ed/X837u/bRt0cK3zp/GLMnDaRP2jEWUu8+HTSQvfSD4Ex42t0xnwKvw2hoCMbCKdsU9AZ69adQuRNOviSYijH98HectsqQc4Ln916KfyI42D4Qw4bi5OTkxjtiW63wAfjrN2FAAVz8h7ZpRH7/1WCfu9YEU29e8GPo2bqCMWximgjMbCrwMyAR+I2733HI+18GfgJsjaz6hbv/JpYxdWRrdlRw27Pv8n/rSxmanca9F0/g/NE5JLfVTV9rnw/6up9wVtAgWrYZvjA/vj042op7MMxy2abgsafp82Yo3xKZDSpiQEFQDZQ7se1jScsORsx8b2Ewm1c8bXo96BXTa2B842hJwVegex948gp4YCpc+tTRF9qVxfCP78LyPwTf9+I/Bvd9yBHFLBGYWSJwLzAFKAKWmNkz7r7ykE3/4O7XxCqOzmDPvhruWrCWRxZvIj0lme9PH80lpwxs27t+y7fCzneCKpCPfwMyT4D/vRp+OyX4D9NRR8xs6kBZpKDf/NGCvmwz1B5yw1L33tArPyiUR04LXmfkQ69BwfeNZdXYkMmw6JdB9US8ZvFyDxqKB3fwiWNGTYduT8HjF8Nvz4dLn4as4dF/vqEelj4IL94WDO/xiW8FVZ8dbZiPDiyWVwSTgPXuvgHAzB4HZgCHJoLQqq1v4NHFm7hrwToqqmr54qn5XH/eMDJSY1Bds+754PngGdLYLwRnXo9fAr85D2b/Pn79zJvTUA/LHob1L0YK+k3BEMdNdUkPCvbME4LqmMaCfmDwiOc0ikMmw//9DDa+FjQ+x8PuDUH1Vzy7jUZr8Cfg8r/BIxcF7VgX/zG64TC2vQXPXg/blsGgT8CF/y+YPEdaJWZ3FpvZ54Gp7v7VyPKlwClNz/4jVUM/AkqAtcD17v6R1hwzuwq4KrI4HFhzlGH1AXYd5WfjoTPF25lihc4Vb2eKFTpXvJ0pVji2ePPdPau5N+LdWPwX4DF3rzazfwMeBiYfupG7zwPmHevBzKywpVusO6LOFG9nihU6V7ydKVboXPF2plghdvHGspPzViCvyXIuHzQKA+Dupe5eHVn8DRCDljsRETmcWCaCJcBQMxtsZl2A2cAzTTcws35NFqcDq2IYj4iINCNmVUPuXmdm1wDPE3QffcDd3zWz2wimTHsG+A8zmw7UAbuBL8cqnohjrl5qZ50p3s4UK3SueDtTrNC54u1MsUKM4u10w1CLiEjb0qAzIiIhp0QgIhJyoUkEZjbVzNaY2Xozuzne8bTEzPLMbKGZrTSzd83sunjHFA0zSzSzN82s5QlaOwAz62VmT5jZajNbZWYd+m4rM7s+8newwsweM7MONd64mT1gZsVmtqLJukwze8HM1kWe225OxWPQQqw/ifwtLDezp83sKEd0bFvNxdrkvW+amZtZm40rHopE0GS4iwuAUcAcMxsV36haVAd8091HAacCV3fgWJu6js7R6+tnwHPuPgIYRweO2cwGAP8BFLj7SQSdLmbHN6qPeAiYesi6m4EX3X0o8GJkuSN4iI/G+gJwkruPJbip9dvtHVQLHuKjsWJmecD5wOa2PFgoEgFNhrtw9xrg4HAXHY67b3f3ZZHXFQQFVQeZ87B5ZpYLXEhwL0iHZWY9gTOB3wK4e427l8U3qiNKArqZWRLQHdgW53g+xN1fIejx19QMgptDiTzHbjKEVmguVnf/h7vXRRYXE9zvFHct/K4AdwE3Am3ayycsiWAA0HToiiI6eOEKYGaDgPFAx50/L3A3wR9nQ7wDOYLBBMOZPBipxvqNmaXGO6iWuPtW4E6Cs7/tQLm7/yO+UUUlx923R17vAHLiGUwrfAX4e7yDaImZzQC2uvvbbb3vsCSCTsfM0oAngW+4+954x9MSM5sGFLv70njHEoUkYAJwn7uPB/bRcaotPiJStz6DIIH1B1LNrIPMfxkdD/qnd/g+6mb2HYJq2UfjHUtzzKw7cAvwvVjsPyyJ4IjDXXQkZpZMkAQedfen4h3PEZwBTDezjQRVbpPN7JH4htSiIqDI3Q9eYT1BkBg6qvOA9929xN1rgaeA0+McUzR2Hhw1IPJcHOd4Disy+OU04BLvuDdWDSE4IXg78n8tF1hmZm0yo1JYEsERh7voKCyY9PW3wCp3/3/xjudI3P3b7p7r7oMIfteX3L1DnrW6+w5gi5kdHOz+XDr2sOibgVPNrHvk7+JcOnDjdhPPAJdFXl8G/G8cYzmsyORZNwLT3X1/vONpibu/4+7Z7j4o8n+tCJgQ+Zs+ZqFIBJHGoIPDXawC/uju78Y3qhadAVxKcGb9VuTxqXgHdRy5FnjUzJYDJwM/jHM8LYpcuTwBLAPeIfj/2qGGRDCzx4BFwHAzKzKzK4A7gClmto7gquaOw+2jvbQQ6y+AdOCFyP+1X8U1yIgWYo3d8TrulZCIiLSHUFwRiIhIy5QIRERCTolARCTklAhEREJOiUBEJOSUCEQOYWb1TbruvtWWo9Wa2aDmRpQUiaeYTVUp0okdcPeT4x2ESHvRFYFIlMxso5n9j5m9Y2b/MrMTI+sHmdlLkTHtXzSzgZH1OZEx7t+OPA4OD5FoZvdH5hn4h5l1i9uXEkGJQKQ53Q6pGprV5L1ydx9DcEfq3ZF1Pwcejoxp/yhwT2T9PcA/3X0cwZhGB+9mHwrc6+6jgTLgohh/H5HD0p3FIocws0p3T2tm/UZgsrtviAwMuMPde5vZLqCfu9dG1m939z5mVgLkunt1k30MAl6ITNqCmd0EJLv77bH/ZiLN0xWBSOt4C69bo7rJ63rUVidxpkQg0jqzmjwvirx+nQ+mkLwEeDXy+kXg69A4p3PP9gpSpDV0JiLyUd3M7K0my8+5+8EupBmRkUurgTmRddcSzHr2nwQzoF0eWX8dMC8ycmQ9QVLYjkgHozYCkShF2ggK3H1XvGMRaUuqGhIRCTldEYiIhJyuCEREQk6JQEQk5JQIRERCTolARCTklAhERELu/wPy6NTD1WTL4AAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.9042 - accuracy: 0.5558\n",
            "--------------------------------- 3 128 3 0.5558000206947327 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 2.0259 - accuracy: 0.4289 - val_loss: 2.0329 - val_accuracy: 0.4215\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.9380 - accuracy: 0.5195 - val_loss: 1.9432 - val_accuracy: 0.5147\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8951 - accuracy: 0.5631 - val_loss: 2.0001 - val_accuracy: 0.4567\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8669 - accuracy: 0.5912 - val_loss: 1.8732 - val_accuracy: 0.5838\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8467 - accuracy: 0.6119 - val_loss: 1.8633 - val_accuracy: 0.5954\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8297 - accuracy: 0.6293 - val_loss: 1.9499 - val_accuracy: 0.5082\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8134 - accuracy: 0.6454 - val_loss: 1.9078 - val_accuracy: 0.5500\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8061 - accuracy: 0.6527 - val_loss: 1.8959 - val_accuracy: 0.5608\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7866 - accuracy: 0.6727 - val_loss: 1.8934 - val_accuracy: 0.5650\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7792 - accuracy: 0.6797 - val_loss: 1.8509 - val_accuracy: 0.6072\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7679 - accuracy: 0.6912 - val_loss: 1.8430 - val_accuracy: 0.6150\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7610 - accuracy: 0.6986 - val_loss: 1.8538 - val_accuracy: 0.6042\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7506 - accuracy: 0.7096 - val_loss: 1.8025 - val_accuracy: 0.6561\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7424 - accuracy: 0.7174 - val_loss: 1.8109 - val_accuracy: 0.6480\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7356 - accuracy: 0.7241 - val_loss: 1.7915 - val_accuracy: 0.6670\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.7915 - accuracy: 0.6670\n",
            "--------------------------------- 3 256 2 0.6669999957084656 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 2.0493 - accuracy: 0.4057 - val_loss: 1.9871 - val_accuracy: 0.4676\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.9788 - accuracy: 0.4775 - val_loss: 2.0876 - val_accuracy: 0.3669\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.9450 - accuracy: 0.5120 - val_loss: 2.1144 - val_accuracy: 0.3418\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.9237 - accuracy: 0.5341 - val_loss: 1.9858 - val_accuracy: 0.4714\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.9108 - accuracy: 0.5464 - val_loss: 1.9741 - val_accuracy: 0.4814\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8942 - accuracy: 0.5642 - val_loss: 1.9320 - val_accuracy: 0.5261\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8786 - accuracy: 0.5806 - val_loss: 1.9564 - val_accuracy: 0.5012\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8676 - accuracy: 0.5913 - val_loss: 1.8964 - val_accuracy: 0.5621\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8606 - accuracy: 0.5973 - val_loss: 1.8836 - val_accuracy: 0.5739\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8547 - accuracy: 0.6032 - val_loss: 1.9134 - val_accuracy: 0.5440\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8459 - accuracy: 0.6119 - val_loss: 1.8839 - val_accuracy: 0.5749\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8365 - accuracy: 0.6228 - val_loss: 1.8677 - val_accuracy: 0.5916\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8282 - accuracy: 0.6305 - val_loss: 1.9146 - val_accuracy: 0.5427\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8208 - accuracy: 0.6383 - val_loss: 1.9082 - val_accuracy: 0.5497\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8174 - accuracy: 0.6421 - val_loss: 1.8801 - val_accuracy: 0.5777\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8801 - accuracy: 0.5777\n",
            "--------------------------------- 3 256 3 0.5777000188827515 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 2.0492 - accuracy: 0.4057 - val_loss: 2.0384 - val_accuracy: 0.4161\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.9545 - accuracy: 0.5031 - val_loss: 2.0534 - val_accuracy: 0.4030\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.9062 - accuracy: 0.5509 - val_loss: 1.9652 - val_accuracy: 0.4916\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8785 - accuracy: 0.5800 - val_loss: 1.9088 - val_accuracy: 0.5491\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8597 - accuracy: 0.5986 - val_loss: 1.9620 - val_accuracy: 0.4942\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8411 - accuracy: 0.6178 - val_loss: 1.9386 - val_accuracy: 0.5185\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8259 - accuracy: 0.6333 - val_loss: 1.9176 - val_accuracy: 0.5404\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8152 - accuracy: 0.6438 - val_loss: 1.8975 - val_accuracy: 0.5600\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.8030 - accuracy: 0.6564 - val_loss: 1.8534 - val_accuracy: 0.6058\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.7964 - accuracy: 0.6630 - val_loss: 1.8734 - val_accuracy: 0.5860\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.7834 - accuracy: 0.6758 - val_loss: 1.8669 - val_accuracy: 0.5912\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.7801 - accuracy: 0.6790 - val_loss: 1.8341 - val_accuracy: 0.6243\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.7678 - accuracy: 0.6913 - val_loss: 1.8433 - val_accuracy: 0.6154\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.7662 - accuracy: 0.6930 - val_loss: 1.9379 - val_accuracy: 0.5210\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 1.7546 - accuracy: 0.7056 - val_loss: 1.8132 - val_accuracy: 0.6458\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8132 - accuracy: 0.6458\n",
            "--------------------------------- 6 96 2 0.645799994468689 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 7s 4ms/step - loss: 2.0666 - accuracy: 0.3877 - val_loss: 2.2193 - val_accuracy: 0.2360\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9969 - accuracy: 0.4597 - val_loss: 2.0623 - val_accuracy: 0.3937\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9594 - accuracy: 0.4973 - val_loss: 1.9865 - val_accuracy: 0.4697\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9354 - accuracy: 0.5206 - val_loss: 2.0942 - val_accuracy: 0.3615\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9169 - accuracy: 0.5403 - val_loss: 1.9280 - val_accuracy: 0.5300\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.9009 - accuracy: 0.5572 - val_loss: 1.9453 - val_accuracy: 0.5123\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8920 - accuracy: 0.5658 - val_loss: 1.9604 - val_accuracy: 0.4987\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8793 - accuracy: 0.5793 - val_loss: 1.8985 - val_accuracy: 0.5593\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8675 - accuracy: 0.5908 - val_loss: 2.0415 - val_accuracy: 0.4168\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8625 - accuracy: 0.5955 - val_loss: 1.8940 - val_accuracy: 0.5633\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8539 - accuracy: 0.6049 - val_loss: 1.9028 - val_accuracy: 0.5566\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8487 - accuracy: 0.6097 - val_loss: 1.9325 - val_accuracy: 0.5261\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8395 - accuracy: 0.6194 - val_loss: 1.9634 - val_accuracy: 0.4940\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8408 - accuracy: 0.6181 - val_loss: 1.8665 - val_accuracy: 0.5917\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 6s 4ms/step - loss: 1.8279 - accuracy: 0.6312 - val_loss: 1.8788 - val_accuracy: 0.5806\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8788 - accuracy: 0.5806\n",
            "--------------------------------- 6 96 3 0.5806000232696533 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 2.0490 - accuracy: 0.4047 - val_loss: 2.1714 - val_accuracy: 0.2829\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.9590 - accuracy: 0.4976 - val_loss: 2.0684 - val_accuracy: 0.3895\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.9057 - accuracy: 0.5521 - val_loss: 1.9935 - val_accuracy: 0.4631\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8759 - accuracy: 0.5827 - val_loss: 1.9435 - val_accuracy: 0.5134\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8573 - accuracy: 0.6012 - val_loss: 1.9124 - val_accuracy: 0.5450\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8393 - accuracy: 0.6193 - val_loss: 1.8736 - val_accuracy: 0.5849\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8213 - accuracy: 0.6371 - val_loss: 1.8922 - val_accuracy: 0.5653\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8119 - accuracy: 0.6465 - val_loss: 1.8927 - val_accuracy: 0.5657\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7979 - accuracy: 0.6614 - val_loss: 1.8581 - val_accuracy: 0.6001\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7909 - accuracy: 0.6682 - val_loss: 1.8974 - val_accuracy: 0.5589\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7777 - accuracy: 0.6816 - val_loss: 1.8789 - val_accuracy: 0.5802\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7693 - accuracy: 0.6890 - val_loss: 1.8271 - val_accuracy: 0.6307\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7561 - accuracy: 0.7035 - val_loss: 1.8101 - val_accuracy: 0.6472\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7496 - accuracy: 0.7102 - val_loss: 1.8843 - val_accuracy: 0.5753\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7439 - accuracy: 0.7155 - val_loss: 1.8673 - val_accuracy: 0.5914\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8673 - accuracy: 0.5914\n",
            "--------------------------------- 6 128 2 0.5914000272750854 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 2.0637 - accuracy: 0.3899 - val_loss: 2.0664 - val_accuracy: 0.3899\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.9915 - accuracy: 0.4646 - val_loss: 2.0171 - val_accuracy: 0.4377\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.9544 - accuracy: 0.5017 - val_loss: 2.0062 - val_accuracy: 0.4511\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.9304 - accuracy: 0.5271 - val_loss: 1.9832 - val_accuracy: 0.4753\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.9164 - accuracy: 0.5411 - val_loss: 1.9267 - val_accuracy: 0.5309\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8972 - accuracy: 0.5613 - val_loss: 2.0394 - val_accuracy: 0.4193\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8862 - accuracy: 0.5712 - val_loss: 2.0111 - val_accuracy: 0.4474\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8734 - accuracy: 0.5846 - val_loss: 1.9487 - val_accuracy: 0.5085\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8610 - accuracy: 0.5981 - val_loss: 2.0994 - val_accuracy: 0.3598\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8558 - accuracy: 0.6027 - val_loss: 1.9160 - val_accuracy: 0.5421\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8485 - accuracy: 0.6101 - val_loss: 1.9484 - val_accuracy: 0.5093\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8405 - accuracy: 0.6180 - val_loss: 1.9656 - val_accuracy: 0.4933\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8313 - accuracy: 0.6272 - val_loss: 2.0187 - val_accuracy: 0.4397\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8279 - accuracy: 0.6309 - val_loss: 1.9118 - val_accuracy: 0.5470\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 7s 5ms/step - loss: 1.8205 - accuracy: 0.6381 - val_loss: 1.8995 - val_accuracy: 0.5599\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8995 - accuracy: 0.5599\n",
            "--------------------------------- 6 128 3 0.5598999857902527 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 2.0464 - accuracy: 0.4094 - val_loss: 2.0637 - val_accuracy: 0.3916\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.9543 - accuracy: 0.5019 - val_loss: 1.9504 - val_accuracy: 0.5060\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.9046 - accuracy: 0.5530 - val_loss: 1.8956 - val_accuracy: 0.5616\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8727 - accuracy: 0.5857 - val_loss: 1.9167 - val_accuracy: 0.5402\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8505 - accuracy: 0.6077 - val_loss: 1.9732 - val_accuracy: 0.4821\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8258 - accuracy: 0.6330 - val_loss: 1.8620 - val_accuracy: 0.5959\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8131 - accuracy: 0.6455 - val_loss: 1.9666 - val_accuracy: 0.4929\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7976 - accuracy: 0.6617 - val_loss: 1.9022 - val_accuracy: 0.5552\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7830 - accuracy: 0.6762 - val_loss: 1.8543 - val_accuracy: 0.6046\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7710 - accuracy: 0.6884 - val_loss: 1.9382 - val_accuracy: 0.5195\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7595 - accuracy: 0.6994 - val_loss: 1.8302 - val_accuracy: 0.6277\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7460 - accuracy: 0.7141 - val_loss: 1.8847 - val_accuracy: 0.5740\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7372 - accuracy: 0.7226 - val_loss: 1.8121 - val_accuracy: 0.6465\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7262 - accuracy: 0.7333 - val_loss: 1.8429 - val_accuracy: 0.6158\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7193 - accuracy: 0.7403 - val_loss: 1.8302 - val_accuracy: 0.6293\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8302 - accuracy: 0.6293\n",
            "--------------------------------- 6 256 2 0.6292999982833862 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 2.0639 - accuracy: 0.3896 - val_loss: 2.2255 - val_accuracy: 0.2301\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.9957 - accuracy: 0.4604 - val_loss: 1.9936 - val_accuracy: 0.4625\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.9656 - accuracy: 0.4903 - val_loss: 1.9703 - val_accuracy: 0.4867\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.9409 - accuracy: 0.5166 - val_loss: 2.0244 - val_accuracy: 0.4303\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.9180 - accuracy: 0.5400 - val_loss: 1.9574 - val_accuracy: 0.5001\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8978 - accuracy: 0.5601 - val_loss: 1.9656 - val_accuracy: 0.4905\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8833 - accuracy: 0.5748 - val_loss: 1.9268 - val_accuracy: 0.5327\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8738 - accuracy: 0.5844 - val_loss: 2.0002 - val_accuracy: 0.4566\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8643 - accuracy: 0.5948 - val_loss: 1.9381 - val_accuracy: 0.5200\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8541 - accuracy: 0.6046 - val_loss: 1.9926 - val_accuracy: 0.4649\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8449 - accuracy: 0.6148 - val_loss: 1.9272 - val_accuracy: 0.5310\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8369 - accuracy: 0.6223 - val_loss: 1.9819 - val_accuracy: 0.4772\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8334 - accuracy: 0.6255 - val_loss: 1.8995 - val_accuracy: 0.5599\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8267 - accuracy: 0.6326 - val_loss: 1.8934 - val_accuracy: 0.5647\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 13s 8ms/step - loss: 1.8205 - accuracy: 0.6390 - val_loss: 1.8798 - val_accuracy: 0.5806\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8798 - accuracy: 0.5806\n",
            "--------------------------------- 6 256 3 0.5806000232696533 ---------------------------\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3R3cH0oNsjhG",
        "colab_type": "text"
      },
      "source": [
        "## Comparison Table"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "2M3TNtVqd1Iv",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 221
        },
        "outputId": "86e9c676-e54c-4700-b353-bdcb24ec3745"
      },
      "source": [
        "meta"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[3, 96, 2, 0.5110999941825867],\n",
              " [3, 96, 3, 0.5778999924659729],\n",
              " [3, 128, 2, 0.616599977016449],\n",
              " [3, 128, 3, 0.5558000206947327],\n",
              " [3, 256, 2, 0.6669999957084656],\n",
              " [3, 256, 3, 0.5777000188827515],\n",
              " [6, 96, 2, 0.645799994468689],\n",
              " [6, 96, 3, 0.5806000232696533],\n",
              " [6, 128, 2, 0.5914000272750854],\n",
              " [6, 128, 3, 0.5598999857902527],\n",
              " [6, 256, 2, 0.6292999982833862],\n",
              " [6, 256, 3, 0.5806000232696533]]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bw4YF1UCoMFw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "meta.sort(key = lambda k: k[3])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EnUsYcRcstxo",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 221
        },
        "outputId": "62379a53-1f99-40e8-997d-24671f695bf5"
      },
      "source": [
        "meta"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[3, 96, 2, 0.5110999941825867],\n",
              " [3, 128, 3, 0.5558000206947327],\n",
              " [6, 128, 3, 0.5598999857902527],\n",
              " [3, 256, 3, 0.5777000188827515],\n",
              " [3, 96, 3, 0.5778999924659729],\n",
              " [6, 96, 3, 0.5806000232696533],\n",
              " [6, 256, 3, 0.5806000232696533],\n",
              " [6, 128, 2, 0.5914000272750854],\n",
              " [3, 128, 2, 0.616599977016449],\n",
              " [6, 256, 2, 0.6292999982833862],\n",
              " [6, 96, 2, 0.645799994468689],\n",
              " [3, 256, 2, 0.6669999957084656]]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ogvB15hjMPfz",
        "colab_type": "text"
      },
      "source": [
        "3 strides always worse than 2 strides. No clear trend between 3 kernel size and 6 kernel size.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WoICbmtjMcjb",
        "colab_type": "text"
      },
      "source": [
        "# For Loop #2"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eyPzl6kQMDDM",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "b8a78d62-f99e-475c-8de3-389a4ae6f8f7"
      },
      "source": [
        "meta2 = []\n",
        "for ks in [3,6]:\n",
        "    for f in [256,512,1024]:\n",
        "        for s in [2]:\n",
        "            model = keras.Sequential([\n",
        "                # keras.layers.Conv2D(kernel_size=6,filters=64,use_bias=False,padding='same',input_shape=(32,32,3)),\n",
        "                # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                # keras.layers.Activation('relu'),\n",
        "                #keras.layers.Dropout(0.5),\n",
        "                #keras.layers.MaxPool2D(),\n",
        "\n",
        "                keras.layers.Conv2D(kernel_size=ks,filters=f,use_bias=False,padding='same',strides=s),\n",
        "                keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                keras.layers.Activation('relu'),\n",
        "                keras.layers.MaxPool2D(),\n",
        "                \n",
        "                keras.layers.Conv2D(kernel_size=ks,filters=f,use_bias=False,padding='same',strides=s),\n",
        "                keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                keras.layers.Activation('relu'),\n",
        "                keras.layers.MaxPool2D(),\n",
        "                \n",
        "                \n",
        "                keras.layers.Conv2D(kernel_size=ks,filters=f,use_bias=False,padding='same',strides=s),\n",
        "                keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                keras.layers.Activation('relu'),\n",
        "                \n",
        "                keras.layers.Flatten(),\n",
        "                \n",
        "                # keras.layers.Dense(128),\n",
        "                # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                # keras.layers.Activation('relu'),\n",
        "                # keras.layers.Dropout(0.5),\n",
        "\n",
        "                # keras.layers.Dense(200),\n",
        "                # keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                # keras.layers.Activation('relu'),\n",
        "                # keras.layers.Dropout(0.5),\n",
        "\n",
        "                keras.layers.Dense(700),\n",
        "                keras.layers.BatchNormalization(center=True,scale=False),\n",
        "                keras.layers.Activation('relu'),\n",
        "                keras.layers.Dropout(0.5),\n",
        "                \n",
        "                keras.layers.Dense(10, activation = tf.nn.softmax)    \n",
        "            ])\n",
        "\n",
        "            model.compile(optimizer=keras.optimizers.Adam(),\n",
        "                        loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
        "                        metrics=['accuracy'])\n",
        "\n",
        "            history = model.fit(train_images, train_labels, epochs=15, \n",
        "                                validation_data=(test_images, test_labels))\n",
        "\n",
        "            plt.plot(history.history['accuracy'], label='accuracy')\n",
        "            plt.plot(history.history['val_accuracy'], label = 'val_accuracy')\n",
        "            plt.xlabel('Epoch')\n",
        "            plt.ylabel('Accuracy')\n",
        "            plt.ylim([0.5, 1])\n",
        "            plt.legend(loc='lower right')\n",
        "            plt.show()\n",
        "\n",
        "            test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)\n",
        "            meta.append([ks,f,s,test_acc])\n",
        "            print(\"---------------------------------\",ks,f,s,test_acc,\"---------------------------\")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 2.0287 - accuracy: 0.4254 - val_loss: 2.0601 - val_accuracy: 0.3941\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.9320 - accuracy: 0.5250 - val_loss: 2.0377 - val_accuracy: 0.4195\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8917 - accuracy: 0.5666 - val_loss: 2.0896 - val_accuracy: 0.3651\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8629 - accuracy: 0.5957 - val_loss: 1.8774 - val_accuracy: 0.5823\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8407 - accuracy: 0.6178 - val_loss: 1.8663 - val_accuracy: 0.5927\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8226 - accuracy: 0.6358 - val_loss: 2.0709 - val_accuracy: 0.3872\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.8086 - accuracy: 0.6503 - val_loss: 1.8344 - val_accuracy: 0.6249\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7959 - accuracy: 0.6632 - val_loss: 1.8260 - val_accuracy: 0.6327\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7827 - accuracy: 0.6766 - val_loss: 1.8514 - val_accuracy: 0.6085\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7745 - accuracy: 0.6846 - val_loss: 1.8361 - val_accuracy: 0.6231\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7629 - accuracy: 0.6966 - val_loss: 1.8142 - val_accuracy: 0.6447\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7553 - accuracy: 0.7041 - val_loss: 1.9935 - val_accuracy: 0.4650\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7475 - accuracy: 0.7125 - val_loss: 1.8452 - val_accuracy: 0.6122\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7344 - accuracy: 0.7261 - val_loss: 1.7948 - val_accuracy: 0.6647\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 8s 5ms/step - loss: 1.7321 - accuracy: 0.7275 - val_loss: 1.8582 - val_accuracy: 0.6003\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8582 - accuracy: 0.6003\n",
            "--------------------------------- 3 256 2 0.6003000140190125 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 2.0317 - accuracy: 0.4237 - val_loss: 2.0634 - val_accuracy: 0.3893\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.9434 - accuracy: 0.5133 - val_loss: 1.9760 - val_accuracy: 0.4811\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8985 - accuracy: 0.5587 - val_loss: 2.0968 - val_accuracy: 0.3595\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8678 - accuracy: 0.5905 - val_loss: 1.9098 - val_accuracy: 0.5486\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8455 - accuracy: 0.6133 - val_loss: 1.9680 - val_accuracy: 0.4901\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8257 - accuracy: 0.6332 - val_loss: 1.9409 - val_accuracy: 0.5157\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8064 - accuracy: 0.6527 - val_loss: 1.8637 - val_accuracy: 0.5942\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7961 - accuracy: 0.6629 - val_loss: 1.9549 - val_accuracy: 0.5045\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7821 - accuracy: 0.6766 - val_loss: 1.8391 - val_accuracy: 0.6193\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7670 - accuracy: 0.6923 - val_loss: 1.8489 - val_accuracy: 0.6104\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7572 - accuracy: 0.7026 - val_loss: 2.0150 - val_accuracy: 0.4414\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7443 - accuracy: 0.7152 - val_loss: 1.8431 - val_accuracy: 0.6151\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7384 - accuracy: 0.7211 - val_loss: 1.7974 - val_accuracy: 0.6616\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7289 - accuracy: 0.7308 - val_loss: 1.8262 - val_accuracy: 0.6323\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7188 - accuracy: 0.7407 - val_loss: 1.8063 - val_accuracy: 0.6524\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8063 - accuracy: 0.6524\n",
            "--------------------------------- 3 512 2 0.652400016784668 ---------------------------\n",
            "Epoch 1/15\n",
            "   1/1563 [..............................] - ETA: 0s - loss: 2.3224 - accuracy: 0.0625WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0082s vs `on_train_batch_end` time: 0.0147s). Check your callbacks.\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 2.0434 - accuracy: 0.4120 - val_loss: 2.1008 - val_accuracy: 0.3551\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.9501 - accuracy: 0.5072 - val_loss: 1.9661 - val_accuracy: 0.4898\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.9053 - accuracy: 0.5520 - val_loss: 1.8970 - val_accuracy: 0.5608\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.8740 - accuracy: 0.5847 - val_loss: 1.9563 - val_accuracy: 0.5001\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.8506 - accuracy: 0.6079 - val_loss: 2.0501 - val_accuracy: 0.4072\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.8272 - accuracy: 0.6317 - val_loss: 1.8360 - val_accuracy: 0.6228\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.8122 - accuracy: 0.6467 - val_loss: 1.9264 - val_accuracy: 0.5317\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.7963 - accuracy: 0.6630 - val_loss: 1.8881 - val_accuracy: 0.5691\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.7786 - accuracy: 0.6803 - val_loss: 1.8716 - val_accuracy: 0.5874\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.7666 - accuracy: 0.6932 - val_loss: 1.8661 - val_accuracy: 0.5914\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.7536 - accuracy: 0.7057 - val_loss: 2.0281 - val_accuracy: 0.4303\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.7429 - accuracy: 0.7167 - val_loss: 1.8228 - val_accuracy: 0.6344\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.7336 - accuracy: 0.7264 - val_loss: 1.8366 - val_accuracy: 0.6233\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.7247 - accuracy: 0.7347 - val_loss: 1.8117 - val_accuracy: 0.6469\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 36s 23ms/step - loss: 1.7182 - accuracy: 0.7416 - val_loss: 1.8597 - val_accuracy: 0.5984\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 2s - loss: 1.8597 - accuracy: 0.5984\n",
            "--------------------------------- 3 1024 2 0.5983999967575073 ---------------------------\n",
            "Epoch 1/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 2.0538 - accuracy: 0.3989 - val_loss: 2.0641 - val_accuracy: 0.3903\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.9540 - accuracy: 0.5021 - val_loss: 1.9515 - val_accuracy: 0.5049\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.9025 - accuracy: 0.5554 - val_loss: 2.0002 - val_accuracy: 0.4562\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8673 - accuracy: 0.5914 - val_loss: 1.9604 - val_accuracy: 0.4956\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8443 - accuracy: 0.6136 - val_loss: 1.9391 - val_accuracy: 0.5204\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8276 - accuracy: 0.6316 - val_loss: 1.9170 - val_accuracy: 0.5402\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.8059 - accuracy: 0.6536 - val_loss: 1.9109 - val_accuracy: 0.5468\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7942 - accuracy: 0.6653 - val_loss: 1.8424 - val_accuracy: 0.6159\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7774 - accuracy: 0.6823 - val_loss: 1.8675 - val_accuracy: 0.5899\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7692 - accuracy: 0.6904 - val_loss: 2.0052 - val_accuracy: 0.4535\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7591 - accuracy: 0.7000 - val_loss: 1.8630 - val_accuracy: 0.5936\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7443 - accuracy: 0.7158 - val_loss: 1.8808 - val_accuracy: 0.5782\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7357 - accuracy: 0.7241 - val_loss: 1.9007 - val_accuracy: 0.5586\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7290 - accuracy: 0.7305 - val_loss: 1.8229 - val_accuracy: 0.6373\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 14s 9ms/step - loss: 1.7169 - accuracy: 0.7428 - val_loss: 1.8030 - val_accuracy: 0.6554\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 1s - loss: 1.8030 - accuracy: 0.6554\n",
            "--------------------------------- 6 256 2 0.6553999781608582 ---------------------------\n",
            "Epoch 1/15\n",
            "   1/1563 [..............................] - ETA: 0s - loss: 2.3034 - accuracy: 0.0938WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0087s vs `on_train_batch_end` time: 0.0150s). Check your callbacks.\n",
            "1563/1563 [==============================] - 40s 25ms/step - loss: 2.0481 - accuracy: 0.4062 - val_loss: 2.1506 - val_accuracy: 0.3061\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 39s 25ms/step - loss: 1.9542 - accuracy: 0.5018 - val_loss: 2.1710 - val_accuracy: 0.2840\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 39s 25ms/step - loss: 1.9065 - accuracy: 0.5519 - val_loss: 2.0789 - val_accuracy: 0.3762\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 40s 25ms/step - loss: 1.8757 - accuracy: 0.5827 - val_loss: 1.9462 - val_accuracy: 0.5116\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 39s 25ms/step - loss: 1.8479 - accuracy: 0.6116 - val_loss: 1.8949 - val_accuracy: 0.5624\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 39s 25ms/step - loss: 1.8337 - accuracy: 0.6247 - val_loss: 1.9182 - val_accuracy: 0.5391\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 39s 25ms/step - loss: 1.8187 - accuracy: 0.6407 - val_loss: 1.8743 - val_accuracy: 0.5841\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 40s 25ms/step - loss: 1.8001 - accuracy: 0.6592 - val_loss: 1.8575 - val_accuracy: 0.6008\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 40s 25ms/step - loss: 1.7821 - accuracy: 0.6775 - val_loss: 1.9541 - val_accuracy: 0.5034\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 39s 25ms/step - loss: 1.7653 - accuracy: 0.6941 - val_loss: 1.8486 - val_accuracy: 0.6111\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 39s 25ms/step - loss: 1.7519 - accuracy: 0.7076 - val_loss: 1.8223 - val_accuracy: 0.6363\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 40s 25ms/step - loss: 1.7448 - accuracy: 0.7151 - val_loss: 1.8468 - val_accuracy: 0.6112\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 40s 25ms/step - loss: 1.7336 - accuracy: 0.7263 - val_loss: 1.9333 - val_accuracy: 0.5253\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 40s 26ms/step - loss: 1.7225 - accuracy: 0.7373 - val_loss: 1.7944 - val_accuracy: 0.6653\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 41s 26ms/step - loss: 1.7146 - accuracy: 0.7453 - val_loss: 1.8001 - val_accuracy: 0.6588\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 2s - loss: 1.8001 - accuracy: 0.6588\n",
            "--------------------------------- 6 512 2 0.6588000059127808 ---------------------------\n",
            "Epoch 1/15\n",
            "   2/1563 [..............................] - ETA: 1:17 - loss: 2.2930 - accuracy: 0.0938WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0329s vs `on_train_batch_end` time: 0.0660s). Check your callbacks.\n",
            "1563/1563 [==============================] - 157s 101ms/step - loss: 2.0600 - accuracy: 0.3937 - val_loss: 2.1273 - val_accuracy: 0.3263\n",
            "Epoch 2/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.9640 - accuracy: 0.4923 - val_loss: 2.0358 - val_accuracy: 0.4219\n",
            "Epoch 3/15\n",
            "1563/1563 [==============================] - 158s 101ms/step - loss: 1.9160 - accuracy: 0.5409 - val_loss: 1.9445 - val_accuracy: 0.5147\n",
            "Epoch 4/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.8823 - accuracy: 0.5762 - val_loss: 1.9296 - val_accuracy: 0.5286\n",
            "Epoch 5/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.8559 - accuracy: 0.6026 - val_loss: 2.0179 - val_accuracy: 0.4396\n",
            "Epoch 6/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.8331 - accuracy: 0.6252 - val_loss: 2.0510 - val_accuracy: 0.4056\n",
            "Epoch 7/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.8086 - accuracy: 0.6499 - val_loss: 2.0095 - val_accuracy: 0.4480\n",
            "Epoch 8/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.7940 - accuracy: 0.6655 - val_loss: 1.8752 - val_accuracy: 0.5842\n",
            "Epoch 9/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.7747 - accuracy: 0.6840 - val_loss: 1.8396 - val_accuracy: 0.6178\n",
            "Epoch 10/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.7609 - accuracy: 0.6977 - val_loss: 1.8502 - val_accuracy: 0.6073\n",
            "Epoch 11/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.7480 - accuracy: 0.7120 - val_loss: 1.8880 - val_accuracy: 0.5708\n",
            "Epoch 12/15\n",
            "1563/1563 [==============================] - 157s 101ms/step - loss: 1.7350 - accuracy: 0.7249 - val_loss: 1.9112 - val_accuracy: 0.5469\n",
            "Epoch 13/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.7206 - accuracy: 0.7394 - val_loss: 1.7986 - val_accuracy: 0.6594\n",
            "Epoch 14/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.7117 - accuracy: 0.7478 - val_loss: 1.8162 - val_accuracy: 0.6429\n",
            "Epoch 15/15\n",
            "1563/1563 [==============================] - 157s 100ms/step - loss: 1.6980 - accuracy: 0.7622 - val_loss: 1.7863 - val_accuracy: 0.6735\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "313/313 - 8s - loss: 1.7863 - accuracy: 0.6735\n",
            "--------------------------------- 6 1024 2 0.6735000014305115 ---------------------------\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SDWtt92QMlUb",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        },
        "outputId": "ecdd1ef0-5cf8-4935-b2e7-7b167a8e98a8"
      },
      "source": [
        "meta"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[3, 256, 2, 0.6003000140190125],\n",
              " [3, 512, 2, 0.652400016784668],\n",
              " [3, 1024, 2, 0.5983999967575073],\n",
              " [6, 256, 2, 0.6553999781608582],\n",
              " [6, 512, 2, 0.6588000059127808],\n",
              " [6, 1024, 2, 0.6735000014305115]]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3-coOGJmjSb0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}