fashion_mnist_denoise.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/pawanagr/anaconda3/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n",
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import keras \n",
    "from keras.datasets import fashion_mnist\n",
    "import matplotlib.pyplot as plt\n",
    "from keras.models import Sequential,Model\n",
    "from keras.layers import Input, Dense, Activation,Conv2D,MaxPooling2D,Dropout,Flatten,Reshape,UpSampling2D,Deconvolution2D,Conv2DTranspose\n",
    "import numpy as np\n",
    "from keras.callbacks import TensorBoard\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "(x_train, y_train), (x_test, y_test) = fashion_mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         (None, 28, 28, 1)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_1 (Conv2D)            (None, 28, 28, 16)        160       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 14, 14, 16)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 14, 14, 8)         1160      \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 7, 7, 8)           0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 7, 7, 8)           584       \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_1 (Conv2DTr (None, 7, 7, 8)           584       \n",
      "_________________________________________________________________\n",
      "up_sampling2d_1 (UpSampling2 (None, 14, 14, 8)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_2 (Conv2DTr (None, 14, 14, 16)        1168      \n",
      "_________________________________________________________________\n",
      "up_sampling2d_2 (UpSampling2 (None, 28, 28, 16)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_transpose_3 (Conv2DTr (None, 28, 28, 1)         145       \n",
      "=================================================================\n",
      "Total params: 3,801\n",
      "Trainable params: 3,801\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "input_img = Input(shape=(28,28,1))\n",
    "#encoder\n",
    "l1 = Conv2D(16, (3,3),strides = 1 , activation = 'relu',padding='same')(input_img)\n",
    "l1 = MaxPooling2D((2,2))(l1)\n",
    "l1 = Conv2D(8,(3,3),strides = 1, activation='relu',padding='same')(l1)\n",
    "l1 = MaxPooling2D((2,2),padding = 'same')(l1)\n",
    "#\n",
    "l1 = Conv2D(8,(3,3),strides = 1,activation = 'relu',padding='same')(l1)\n",
    "#decoder network\n",
    "l1 = Conv2DTranspose(8,(3,3),strides = 1,activation='relu',padding='same')(l1)\n",
    "l1 = UpSampling2D((2,2))(l1)\n",
    "l1 = Conv2DTranspose(16,(3,3),strides = 1,activation='relu',padding = 'same')(l1)\n",
    "l1 = UpSampling2D((2,2))(l1)\n",
    "l2 = Conv2DTranspose(1,(3,3),strides = 1, activation='sigmoid',padding  ='same')(l1)\n",
    "\n",
    "model2 = Model(input_img, l2)\n",
    "model2.compile(optimizer='Adam', loss='binary_crossentropy')\n",
    "model2.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_train_1=x_train/255.0\n",
    "from time import time\n",
    "tensorboard = TensorBoard(log_dir=\"logs/{}\".format(time()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 28, 28, 1)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train_ = x_train_1 + 0.3 * np.random.normal(loc=0.0, scale=1.0, size=x_train.shape)\n",
    "x_train_ = np.clip(x_train_, 0., 1.)\n",
    "x_ = x_train_[:,:,:,np.newaxis]\n",
    "x_.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.max(x_[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f5a9868f7f0>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(np.reshape(x_train_[500],(28,28)),cmap='Greys')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(60000, 28, 28)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "60000/60000 [==============================] - 53s 878us/step - loss: 0.2858\n",
      "Epoch 2/10\n",
      "60000/60000 [==============================] - 48s 794us/step - loss: 0.2855\n",
      "Epoch 3/10\n",
      "60000/60000 [==============================] - 47s 777us/step - loss: 0.2852\n",
      "Epoch 4/10\n",
      "60000/60000 [==============================] - 52s 861us/step - loss: 0.2849\n",
      "Epoch 5/10\n",
      "60000/60000 [==============================] - 51s 854us/step - loss: 0.2846\n",
      "Epoch 6/10\n",
      "60000/60000 [==============================] - 45s 742us/step - loss: 0.2843\n",
      "Epoch 7/10\n",
      "60000/60000 [==============================] - 45s 742us/step - loss: 0.2841\n",
      "Epoch 8/10\n",
      "60000/60000 [==============================] - 51s 849us/step - loss: 0.2838\n",
      "Epoch 9/10\n",
      "60000/60000 [==============================] - 49s 814us/step - loss: 0.2836\n",
      "Epoch 10/10\n",
      "60000/60000 [==============================] - 50s 832us/step - loss: 0.2834\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0x7f5a87ff5a90>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model2.fit(x_, x_train_1[:,:,:,np.newaxis], epochs = 10, batch_size = 256, callbacks = [tensorboard])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "pp = model2.predict(np.reshape(x_[0],(1,28,28,1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "pp = np.reshape(pp,(28,28))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## reconstructed Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f5a87f44278>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(pp,cmap='Greys')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f5a94088cf8>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(np.reshape(x_[0],(28,28)),cmap='Greys')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7f5a865cc4e0>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(x_train[0],cmap='Greys')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}