Simple_Autoencoders.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Simple_Autoencoders.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyPXoOXKR6wgUozRomtynFno",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Neel7317/Generative-Models/blob/main/Simple_Autoencoders.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "P0suXqKq1OD7"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "from tensorflow.keras.layers import Input,Dense"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.models import Model\n"
      ],
      "metadata": {
        "id": "HN-woD661dbr"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.optimizers import Adam\n",
        "import tensorflow as tf"
      ],
      "metadata": {
        "id": "a4q6TcUd2DES"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "XLQtCWL03YDP"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "np.random.seed(42)\n",
        "#tf.random.seed(42)"
      ],
      "metadata": {
        "id": "v86wLj0F2SgZ"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "datasets=tf.keras.datasets.mnist"
      ],
      "metadata": {
        "id": "R2NXiIwF3OCS"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "(x_train,y_train), (x_test,y_test) = tf.keras.datasets.mnist.load_data()"
      ],
      "metadata": {
        "id": "qqWZ-3Uj_93g"
      },
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x_train=x_train/255\n",
        "x_test=x_test/255"
      ],
      "metadata": {
        "id": "2zjNGWyZAGvJ"
      },
      "execution_count": 46,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x_train.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "djWG-30SAx2r",
        "outputId": "98a0d7bf-57c2-4889-e41b-5eedc697a33d"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(60000, 28, 28)"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "x_train=np.reshape(x_train,(-1,28*28*1))# height*width*channel\n",
        "x_test=np.reshape(x_test,(-1,28*28*1))"
      ],
      "metadata": {
        "id": "DLqccnmUBl37"
      },
      "execution_count": 47,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "x_train.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "j-uoKfQOCJOX",
        "outputId": "0f49fa2c-8be1-4ec0-c3e0-7d5aad03072e"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(60000, 784)"
            ]
          },
          "metadata": {},
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "laten_dim=32"
      ],
      "metadata": {
        "id": "rkrndcEyCLqG"
      },
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#simple autoencoder..\n",
        "input=Input(shape=(28*28,))\n",
        "h=Dense(laten_dim,activation='relu')(input)\n",
        "output=Dense(28*28,activation='sigmoid')(h)"
      ],
      "metadata": {
        "id": "9g5W98vqIPAV"
      },
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder=Model(input,output)\n",
        "autoencoder.summary()#in summary you can clearly see that input and output datashape are identical"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T6jgjf-UJG4l",
        "outputId": "7074741b-8324-40e6-ddce-fe2bb05fa369"
      },
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_3 (InputLayer)        [(None, 784)]             0         \n",
            "                                                                 \n",
            " dense_4 (Dense)             (None, 32)                25120     \n",
            "                                                                 \n",
            " dense_5 (Dense)             (None, 784)               25872     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 50,992\n",
            "Trainable params: 50,992\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder.compile(optimizer=Adam(1e-3),loss='binary_crossentropy')"
      ],
      "metadata": {
        "id": "zL_TRJosJXss"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "autoencoder.fit(x_train,x_train,batch_size=256,epochs=50,validation_data=(x_test,x_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S0vxYhNZKujO",
        "outputId": "40161d46-fd86-41c7-b7bd-38f5bd14804a"
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/50\n",
            "235/235 [==============================] - 3s 13ms/step - loss: 0.2771 - val_loss: 0.1928\n",
            "Epoch 2/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.1731 - val_loss: 0.1554\n",
            "Epoch 3/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.1456 - val_loss: 0.1349\n",
            "Epoch 4/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.1294 - val_loss: 0.1219\n",
            "Epoch 5/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.1190 - val_loss: 0.1137\n",
            "Epoch 6/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.1119 - val_loss: 0.1077\n",
            "Epoch 7/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.1069 - val_loss: 0.1036\n",
            "Epoch 8/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.1032 - val_loss: 0.1004\n",
            "Epoch 9/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.1004 - val_loss: 0.0979\n",
            "Epoch 10/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0983 - val_loss: 0.0962\n",
            "Epoch 11/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0969 - val_loss: 0.0951\n",
            "Epoch 12/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0959 - val_loss: 0.0942\n",
            "Epoch 13/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0953 - val_loss: 0.0938\n",
            "Epoch 14/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0948 - val_loss: 0.0933\n",
            "Epoch 15/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0945 - val_loss: 0.0931\n",
            "Epoch 16/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0943 - val_loss: 0.0928\n",
            "Epoch 17/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0940 - val_loss: 0.0927\n",
            "Epoch 18/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0939 - val_loss: 0.0925\n",
            "Epoch 19/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0937 - val_loss: 0.0925\n",
            "Epoch 20/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0936 - val_loss: 0.0923\n",
            "Epoch 21/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0935 - val_loss: 0.0922\n",
            "Epoch 22/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0934 - val_loss: 0.0921\n",
            "Epoch 23/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0934 - val_loss: 0.0920\n",
            "Epoch 24/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0933 - val_loss: 0.0920\n",
            "Epoch 25/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0932 - val_loss: 0.0919\n",
            "Epoch 26/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0932 - val_loss: 0.0919\n",
            "Epoch 27/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0931 - val_loss: 0.0920\n",
            "Epoch 28/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0931 - val_loss: 0.0919\n",
            "Epoch 29/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0930 - val_loss: 0.0918\n",
            "Epoch 30/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0930 - val_loss: 0.0918\n",
            "Epoch 31/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0930 - val_loss: 0.0918\n",
            "Epoch 32/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0929 - val_loss: 0.0918\n",
            "Epoch 33/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0929 - val_loss: 0.0917\n",
            "Epoch 34/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0929 - val_loss: 0.0916\n",
            "Epoch 35/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0928 - val_loss: 0.0917\n",
            "Epoch 36/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0928 - val_loss: 0.0916\n",
            "Epoch 37/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0928 - val_loss: 0.0916\n",
            "Epoch 38/50\n",
            "235/235 [==============================] - 3s 11ms/step - loss: 0.0928 - val_loss: 0.0915\n",
            "Epoch 39/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0928 - val_loss: 0.0916\n",
            "Epoch 40/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0927 - val_loss: 0.0915\n",
            "Epoch 41/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0927 - val_loss: 0.0915\n",
            "Epoch 42/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0927 - val_loss: 0.0915\n",
            "Epoch 43/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0927 - val_loss: 0.0915\n",
            "Epoch 44/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0927 - val_loss: 0.0915\n",
            "Epoch 45/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0927 - val_loss: 0.0915\n",
            "Epoch 46/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0926 - val_loss: 0.0916\n",
            "Epoch 47/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0926 - val_loss: 0.0915\n",
            "Epoch 48/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0926 - val_loss: 0.0915\n",
            "Epoch 49/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0926 - val_loss: 0.0914\n",
            "Epoch 50/50\n",
            "235/235 [==============================] - 3s 12ms/step - loss: 0.0926 - val_loss: 0.0915\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f46c3f5b050>"
            ]
          },
          "metadata": {},
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_pred_y=autoencoder.predict(x_test)"
      ],
      "metadata": {
        "id": "R0mAK-b7LCFL"
      },
      "execution_count": 54,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_pred_y[0].shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BiukHehCNw-x",
        "outputId": "e7c354a3-e1ae-4037-a429-d09fd4e5ebfc"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(784,)"
            ]
          },
          "metadata": {},
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Reconstructed Image** "
      ],
      "metadata": {
        "id": "lYXicRvZO3VW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(test_pred_y[0].reshape(28,28))\n",
        "plt.gray()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "vSyKY333N5D_",
        "outputId": "caa85b90-9605-496e-c528-8a076cf0ff1c"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# **Original Image**"
      ],
      "metadata": {
        "id": "UwKj0dftPBAz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.imshow(x_test[0].reshape(28,28))\n",
        "plt.gray()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "xzRjjbM6Ools",
        "outputId": "ed48079f-07a4-43d1-82f1-24677b6a86d0"
      },
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "n=10\n",
        "plt.figure(figsize=(20,4))\n",
        "for i in range(1,11):\n",
        "  #original Image\n",
        "  ax=plt.subplot(2,n,i)\n",
        "  ax.set_title('Original Images')\n",
        "  plt.imshow(x_test[i].reshape(28,28))\n",
        "  plt.gray()\n",
        "  ax.get_xaxis().set_visible(False)\n",
        "  ax.get_yaxis().set_visible(False)\n",
        "  #Reconstructed Image\n",
        "  ax=plt.subplot(2,n,i+n)\n",
        "  ax.set_title('Reconstructed')\n",
        "  plt.imshow(test_pred_y[i].reshape(28,28))\n",
        "  plt.gray()\n",
        "  ax.get_xaxis().set_visible(False)\n",
        "  ax.get_yaxis().set_visible(False)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 261
        },
        "id": "NPgVDQv_OzYe",
        "outputId": "f0b4ba34-e4aa-4e6a-d55f-1dbe8ab10eb1"
      },
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x288 with 20 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "OL12p23pQBiI"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}