TF_Forum_14434.ipynb

tf_forum_14434.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOg0LdoAtojI0vZDxxCvg9I",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/kiransair/bf447740ee1dec5e9f6c3a9524b41680/tf_forum_14434.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf"
      ],
      "metadata": {
        "id": "Oa5VmmLa91Ho"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class AnomalyDetector(tf.keras.models.Model):\n",
        "    def __init__(self):\n",
        "        super(AnomalyDetector, self).__init__()\n",
        "        \n",
        "        self.encoder = tf.keras.Sequential([\n",
        "            tf.keras.layers.Input(shape=(300, 300, 3)),\n",
        "            tf.keras.layers.Conv2D(64, (3, 3), activation='relu', padding='same'),\n",
        "            tf.keras.layers.MaxPooling2D((2, 2), padding='same'),\n",
        "            tf.keras.layers.BatchNormalization(),\n",
        "            tf.keras.layers.Conv2D(32, (3, 3), activation='relu', padding='same'),\n",
        "            tf.keras.layers.MaxPooling2D((2, 2), padding='same'),\n",
        "            tf.keras.layers.BatchNormalization(),\n",
        "            tf.keras.layers.Conv2D(16, (3, 3), activation='relu', padding='same'),\n",
        "            tf.keras.layers.MaxPooling2D((2, 2), padding='same')\n",
        "            ]) # Smallest Layer Defined Here\n",
        "    \n",
        "        self.decoder = tf.keras.Sequential([\n",
        "            tf.keras.layers.Conv2D(64, (3, 3), activation='relu', padding='same'),\n",
        "            tf.keras.layers.UpSampling2D((2, 2)),\n",
        "            tf.keras.layers.Conv2D(32, (3, 3), activation='relu', padding='same'),\n",
        "            tf.keras.layers.UpSampling2D((2, 2)),\n",
        "            tf.keras.layers.Conv2D(16, (3, 3), activation='relu'),\n",
        "            tf.keras.layers.UpSampling2D((2, 2)),\n",
        "            tf.keras.layers.Conv2D(3, (3, 3), activation='sigmoid', padding='same')\n",
        "            ])\n",
        "    \n",
        "    def call(self, x):\n",
        "        encoded = self.encoder(x)\n",
        "        decoded = self.decoder(encoded)\n",
        "        return decoded\n",
        "\n",
        "auto_encoder = AnomalyDetector()"
      ],
      "metadata": {
        "id": "Z9ZDIfi9_FdU"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "auto_encoder.build((None,300,300,3))"
      ],
      "metadata": {
        "id": "6Bs-l8hiAY2_"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "auto_encoder.summary(expand_nested=True)\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "skRepOs-_MfO",
        "outputId": "f34a71f2-3e80-4d8d-8408-75d4347930cf"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"anomaly_detector\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " sequential_4 (Sequential)   (None, 38, 38, 16)        25264     \n",
            "|¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯|\n",
            "| conv2d_14 (Conv2D)        (None, 300, 300, 64)      1792      |\n",
            "|                                                               |\n",
            "| max_pooling2d_6 (MaxPooling  (None, 150, 150, 64)   0         |\n",
            "| 2D)                                                           |\n",
            "|                                                               |\n",
            "| batch_normalization_4 (Batc  (None, 150, 150, 64)   256       |\n",
            "| hNormalization)                                               |\n",
            "|                                                               |\n",
            "| conv2d_15 (Conv2D)        (None, 150, 150, 32)      18464     |\n",
            "|                                                               |\n",
            "| max_pooling2d_7 (MaxPooling  (None, 75, 75, 32)     0         |\n",
            "| 2D)                                                           |\n",
            "|                                                               |\n",
            "| batch_normalization_5 (Batc  (None, 75, 75, 32)     128       |\n",
            "| hNormalization)                                               |\n",
            "|                                                               |\n",
            "| conv2d_16 (Conv2D)        (None, 75, 75, 16)        4624      |\n",
            "|                                                               |\n",
            "| max_pooling2d_8 (MaxPooling  (None, 38, 38, 16)     0         |\n",
            "| 2D)                                                           |\n",
            "¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯\n",
            " sequential_5 (Sequential)   (None, 300, 300, 3)       32803     \n",
            "|¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯|\n",
            "| conv2d_17 (Conv2D)        (None, 38, 38, 64)        9280      |\n",
            "|                                                               |\n",
            "| up_sampling2d_6 (UpSampling  (None, 76, 76, 64)     0         |\n",
            "| 2D)                                                           |\n",
            "|                                                               |\n",
            "| conv2d_18 (Conv2D)        (None, 76, 76, 32)        18464     |\n",
            "|                                                               |\n",
            "| up_sampling2d_7 (UpSampling  (None, 152, 152, 32)   0         |\n",
            "| 2D)                                                           |\n",
            "|                                                               |\n",
            "| conv2d_19 (Conv2D)        (None, 150, 150, 16)      4624      |\n",
            "|                                                               |\n",
            "| up_sampling2d_8 (UpSampling  (None, 300, 300, 16)   0         |\n",
            "| 2D)                                                           |\n",
            "|                                                               |\n",
            "| conv2d_20 (Conv2D)        (None, 300, 300, 3)       435       |\n",
            "¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯\n",
            "=================================================================\n",
            "Total params: 58,067\n",
            "Trainable params: 57,875\n",
            "Non-trainable params: 192\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "Al9vhak8BJqv"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}