uedge_training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from __future__ import print_function\n",
    "\n",
    "import warnings\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from imutils.paths import list_images\n",
    "from keras import backend as K\n",
    "from keras.callbacks import CSVLogger\n",
    "from keras.layers import Input, concatenate, Conv2D, MaxPooling2D, UpSampling2D\n",
    "from keras.models import Model\n",
    "from keras.optimizers import Adam\n",
    "from scipy.io import loadmat\n",
    "from scipy.misc import imread\n",
    "from skimage.io import imsave\n",
    "from skimage.transform import resize\n",
    "\n",
    "from datasetutils.bsds500 import BSDS500\n",
    "\n",
    "target_size = (256, 256)\n",
    "\n",
    "warnings.filterwarnings('ignore')\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "K.set_image_data_format('channels_last') \n",
    "\n",
    "\n",
    "\n",
    "def get_unet(img_rows, img_cols, channels):\n",
    "    inputs = Input((img_rows, img_cols, channels))\n",
    "    conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(inputs)\n",
    "    conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv1)\n",
    "    pool1 = MaxPooling2D(pool_size=(2, 2))(conv1)\n",
    "\n",
    "    conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(pool1)\n",
    "    conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv2)\n",
    "    pool2 = MaxPooling2D(pool_size=(2, 2))(conv2)\n",
    "\n",
    "    conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(pool2)\n",
    "    conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv3)\n",
    "    pool3 = MaxPooling2D(pool_size=(2, 2))(conv3)\n",
    "\n",
    "    conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(pool3)\n",
    "    conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv4)\n",
    "    pool4 = MaxPooling2D(pool_size=(2, 2))(conv4)\n",
    "\n",
    "    conv5 = Conv2D(512, (3, 3), activation='relu', padding='same')(pool4)\n",
    "    conv5 = Conv2D(512, (3, 3), activation='relu', padding='same')(conv5)\n",
    "\n",
    "    up6 = concatenate([UpSampling2D(size=(2, 2))(conv5), conv4], axis=3)\n",
    "    conv6 = Conv2D(256, (3, 3), activation='relu', padding='same')(up6)\n",
    "    conv6 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv6)\n",
    "\n",
    "    up7 = concatenate([UpSampling2D(size=(2, 2))(conv6), conv3], axis=3)\n",
    "    conv7 = Conv2D(128, (3, 3), activation='relu', padding='same')(up7)\n",
    "    conv7 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv7)\n",
    "\n",
    "    up8 = concatenate([UpSampling2D(size=(2, 2))(conv7), conv2], axis=3)\n",
    "    conv8 = Conv2D(64, (3, 3), activation='relu', padding='same')(up8)\n",
    "    conv8 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv8)\n",
    "\n",
    "    up9 = concatenate([UpSampling2D(size=(2, 2))(conv8), conv1], axis=3)\n",
    "    conv9 = Conv2D(32, (3, 3), activation='relu', padding='same')(up9)\n",
    "    conv9 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv9)\n",
    "\n",
    "    conv10 = Conv2D(1, (1, 1), activation='sigmoid')(conv9)\n",
    "\n",
    "    model = Model(inputs=[inputs], outputs=[conv10])\n",
    "    \n",
    "    \n",
    "    model.compile(optimizer=Adam(lr=1e-3), \n",
    "                      loss='binary_crossentropy',\n",
    "                      metrics=['binary_crossentropy'])\n",
    "\n",
    "\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/i008/BSR\n"
     ]
    }
   ],
   "source": [
    "from datasetutils.bsds500 import BSDS500\n",
    "TARGET_SHAPE = (256, 256)\n",
    "\n",
    "bsds = BSDS500(target_size=TARGET_SHAPE)\n",
    "ids, contours_train, sgmnts, train_images = bsds.get_train()\n",
    "ids, contours_test, sgmnts, test_images = bsds.get_train()\n",
    "\n",
    "C = np.concatenate([contours_test, contours_train])\n",
    "I = np.concatenate([test_images, train_images])\n",
    "\n",
    "csv_callback = CSVLogger('history.log', append=True)\n",
    "unet = get_unet(256, 256, 3)\n",
    "history = unet.fit(I, C, verbose=2, epochs=200, validation_split=0.1, callbacks=[csv_callback])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "csv_callback = CSVLogger('history.log', append=True)\n",
    "unet = get_unet(256, 256, 3)\n",
    "H = unet.fit(I, C, verbose=2, epochs=200, validation_split=0.1, callbacks=[csv_callback])\n",
    "\n",
    "# unet = load_model('unet.keras')\n",
    "# save_model(unet, 'unet2.keras')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x7fe70c1c2240>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtQAAAIlCAYAAAD1429MAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBbUQQgghhBAj8P8AxTyuJHx5thsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe4d894e828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn\n",
    "f = pd.DataFrame(H.history)[['loss','val_loss']].plot(figsize=(12,9))\n",
    "plt.title('model training')\n",
    "plt.ylabel('binary_crossentropy_loss')\n",
    "plt.xlabel('epochs')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3.0
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}