network.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import hyperas\n",
    "import hyperas.distributions\n",
    "import hyperopt\n",
    "import keras.layers\n",
    "import keras.models\n",
    "import keras.optimizers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "shape = (1, 128, 128)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "x = keras.layers.Input(shape)\n",
    "\n",
    "y = keras.layers.Convolution2D(64, 3, 3, activation=\"relu\", border_mode=\"same\")(x)\n",
    "y = keras.layers.Convolution2D(64, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "\n",
    "y = keras.layers.MaxPooling2D((2, 2), (2, 2))(y)\n",
    "\n",
    "y = keras.layers.Convolution2D(128, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(128, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "\n",
    "y = keras.layers.MaxPooling2D((2, 2), (2, 2))(y)\n",
    "\n",
    "y = keras.layers.Convolution2D(256, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(256, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(256, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(256, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "\n",
    "y = keras.layers.MaxPooling2D((2, 2), (2, 2))(y)\n",
    "\n",
    "y = keras.layers.Convolution2D(512, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(512, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(512, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(512, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "\n",
    "y = keras.layers.MaxPooling2D((2, 2), (2, 2))(y)\n",
    "\n",
    "y = keras.layers.Convolution2D(512, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(512, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(512, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "y = keras.layers.Convolution2D(512, 3, 3, activation=\"relu\", border_mode=\"same\")(y)\n",
    "\n",
    "# y = keras.layers.Convolution2D(4096, 8, 8, activation=\"relu\", border_mode=\"same\")(y)\n",
    "# y = keras.layers.Convolution2D(4096, 1, 1, activation=\"relu\", border_mode=\"same\")(y)\n",
    "\n",
    "# y = keras.layers.Deconvolution2D(512, 8, 8, (1, 512, 8, 8), subsample=(2, 2))(y)\n",
    "\n",
    "# y = keras.layers.UpSampling2D()(y)\n",
    "\n",
    "# y = keras.layers.Deconvolution2D(512, 3, 3, (1, 512, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(512, 3, 3, (1, 512, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(512, 3, 3, (1, 512, 16, 16), border_mode=\"same\")(y)\n",
    "\n",
    "# y = keras.layers.UpSampling2D()(y)\n",
    "\n",
    "# y = keras.layers.Deconvolution2D(512, 3, 3, (1, 512, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(512, 3, 3, (1, 512, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(512, 3, 3, (1, 512, 16, 16), border_mode=\"same\")(y)\n",
    "\n",
    "# y = keras.layers.UpSampling2D()(y)\n",
    "\n",
    "# y = keras.layers.Deconvolution2D(512, 3, 3, (1, 512, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(512, 3, 3, (1, 512, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(256, 3, 3, (1, 256, 16, 16), border_mode=\"same\")(y)\n",
    "\n",
    "# y = keras.layers.UpSampling2D()(y)\n",
    "\n",
    "# y = keras.layers.Deconvolution2D(256, 3, 3, (1, 256, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(256, 3, 3, (1, 256, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(128, 3, 3, (1, 128, 16, 16), border_mode=\"same\")(y)\n",
    "\n",
    "# y = keras.layers.UpSampling2D()(y)\n",
    "\n",
    "# y = keras.layers.Deconvolution2D(128, 3, 3, (1, 128, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D( 64, 3, 3, (1,  64, 16, 16), border_mode=\"same\")(y)\n",
    "\n",
    "# y = keras.layers.UpSampling2D()(y)\n",
    "\n",
    "# y = keras.layers.Deconvolution2D(64, 3, 3, (1, 64, 16, 16), border_mode=\"same\")(y)\n",
    "# y = keras.layers.Deconvolution2D(64, 3, 3, (1, 64, 16, 16), border_mode=\"same\")(y)\n",
    "\n",
    "# y = keras.layers.Convolution2D(3, 1, 1, activation=\"relu\")\n",
    "\n",
    "model = keras.models.Model(x, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}