main.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# imports\n",
    "from __future__ import division\n",
    "import numpy as np\n",
    "from collections import namedtuple\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/x/alexburka/20170221/stickcam/2/surface.png\n",
      "/x/alexburka/20170221/optocam/2/surface.png\n",
      "/x/alexburka/20170222/biocam/2/surface.png\n",
      "/x/alexburka/20170221/stickcam/3/surface.png\n",
      "/x/alexburka/20170221/optocam/3/surface.png\n",
      "/x/alexburka/20170222/biocam/3/surface.png\n",
      "/x/alexburka/20170221/stickcam/48/surface.png\n",
      "/x/alexburka/20170222/optocam/1/surface.png\n",
      "/x/alexburka/20170223/biocam/20/surface.png\n",
      "/x/alexburka/20170221/stickcam/52/surface.png\n",
      "/x/alexburka/20170222/optocam/5/surface.png\n",
      "/x/alexburka/20170223/biocam/24/surface.png\n",
      "/x/alexburka/20170221/stickcam/34/surface.png\n",
      "/x/alexburka/20170221/optocam/35/surface.png\n",
      "/x/alexburka/20170223/biocam/3/surface.png\n",
      "/x/alexburka/20170221/stickcam/31/surface.png\n",
      "/x/alexburka/20170221/optocam/32/surface.png\n",
      "/x/alexburka/20170222/biocam/30/surface.png\n",
      "/x/alexburka/20170221/stickcam/15/surface.png\n",
      "/x/alexburka/20170221/optocam/14/surface.png\n",
      "/x/alexburka/20170222/biocam/14/surface.png\n",
      "/x/alexburka/20170221/stickcam/30/surface.png\n",
      "/x/alexburka/20170221/optocam/31/surface.png\n",
      "/x/alexburka/20170222/biocam/29/surface.png\n",
      "/x/alexburka/20170221/stickcam/35/surface.png\n",
      "/x/alexburka/20170221/optocam/36/surface.png\n",
      "/x/alexburka/20170223/biocam/4/surface.png\n",
      "/x/alexburka/20170221/stickcam/22/surface.png\n",
      "/x/alexburka/20170221/optocam/21/surface.png\n",
      "/x/alexburka/20170222/biocam/21/surface.png\n",
      "/x/alexburka/20170221/stickcam/57/surface.png\n",
      "/x/alexburka/20170222/optocam/11/surface.png\n",
      "/x/alexburka/20170223/biocam/30/surface.png\n",
      "/x/alexburka/20170221/stickcam/26/surface.png\n",
      "/x/alexburka/20170222/biocam/25/surface.png\n",
      "/x/alexburka/20170221/stickcam/43/surface.png\n",
      "/x/alexburka/20170221/optocam/45/surface.png\n",
      "/x/alexburka/20170223/biocam/15/surface.png\n",
      "/x/alexburka/20170221/stickcam/46/surface.png\n",
      "/x/alexburka/20170221/optocam/48/surface.png\n",
      "/x/alexburka/20170223/biocam/18/surface.png\n",
      "/x/alexburka/20170221/stickcam/29/surface.png\n",
      "/x/alexburka/20170221/optocam/30/surface.png\n",
      "/x/alexburka/20170222/biocam/28/surface.png\n",
      "/x/alexburka/20170221/stickcam/8/surface.png\n",
      "/x/alexburka/20170221/optocam/7/surface.png\n",
      "/x/alexburka/20170222/biocam/7/surface.png\n",
      "/x/alexburka/20170221/stickcam/23/surface.png\n",
      "/x/alexburka/20170221/optocam/22/surface.png\n",
      "/x/alexburka/20170222/biocam/22/surface.png\n",
      "/x/alexburka/20170221/stickcam/36/surface.png\n",
      "/x/alexburka/20170221/optocam/37/surface.png\n",
      "/x/alexburka/20170223/biocam/5/surface.png\n",
      "/x/alexburka/20170221/stickcam/12/surface.png\n",
      "/x/alexburka/20170221/optocam/11/surface.png\n",
      "/x/alexburka/20170222/biocam/11/surface.png\n",
      "/x/alexburka/20170221/stickcam/28/surface.png\n",
      "/x/alexburka/20170221/optocam/28/surface.png\n",
      "/x/alexburka/20170222/biocam/27/surface.png\n",
      "/x/alexburka/20170221/stickcam/33/surface.png\n",
      "/x/alexburka/20170221/optocam/34/surface.png\n",
      "/x/alexburka/20170223/biocam/2/surface.png\n",
      "/x/alexburka/20170221/stickcam/51/surface.png\n",
      "/x/alexburka/20170222/optocam/4/surface.png\n",
      "/x/alexburka/20170223/biocam/23/surface.png\n",
      "/x/alexburka/20170221/stickcam/11/surface.png\n",
      "/x/alexburka/20170221/optocam/10/surface.png\n",
      "/x/alexburka/20170222/biocam/10/surface.png\n",
      "/x/alexburka/20170221/stickcam/1/surface.png\n",
      "/x/alexburka/20170221/optocam/1/surface.png\n",
      "/x/alexburka/20170222/biocam/1/surface.png\n",
      "/x/alexburka/20170221/stickcam/39/surface.png\n",
      "/x/alexburka/20170221/optocam/41/surface.png\n",
      "/x/alexburka/20170223/biocam/8/surface.png\n",
      "/x/alexburka/20170221/stickcam/59/surface.png\n",
      "/x/alexburka/20170222/optocam/13/surface.png\n",
      "/x/alexburka/20170223/biocam/32/surface.png\n",
      "/x/alexburka/20170221/stickcam/58/surface.png\n",
      "/x/alexburka/20170222/optocam/12/surface.png\n",
      "/x/alexburka/20170223/biocam/31/surface.png\n",
      "/x/alexburka/20170221/stickcam/24/surface.png\n",
      "/x/alexburka/20170221/optocam/23/surface.png\n",
      "/x/alexburka/20170222/biocam/23/surface.png\n",
      "/x/alexburka/20170221/stickcam/14/surface.png\n",
      "/x/alexburka/20170221/optocam/13/surface.png\n",
      "/x/alexburka/20170222/biocam/13/surface.png\n",
      "/x/alexburka/20170221/stickcam/42/surface.png\n",
      "/x/alexburka/20170221/optocam/44/surface.png\n",
      "/x/alexburka/20170223/biocam/14/surface.png\n",
      "/x/alexburka/20170221/stickcam/13/surface.png\n",
      "/x/alexburka/20170221/optocam/12/surface.png\n",
      "/x/alexburka/20170222/biocam/12/surface.png\n",
      "/x/alexburka/20170221/optocam/33/surface.png\n",
      "/x/alexburka/20170223/biocam/1/surface.png\n",
      "/x/alexburka/20170221/optocam/4/surface.png\n",
      "/x/alexburka/20170222/biocam/4/surface.png\n",
      "/x/alexburka/20170221/optocam/15/surface.png\n",
      "/x/alexburka/20170222/biocam/15/surface.png\n",
      "/x/alexburka/20170221/optocam/16/surface.png\n",
      "/x/alexburka/20170222/biocam/16/surface.png\n",
      "/x/alexburka/20170221/optocam/17/surface.png\n",
      "/x/alexburka/20170222/biocam/17/surface.png\n",
      "/x/alexburka/20170221/optocam/18/surface.png\n",
      "/x/alexburka/20170222/biocam/18/surface.png\n",
      "/x/alexburka/20170221/optocam/19/surface.png\n",
      "/x/alexburka/20170222/biocam/19/surface.png\n",
      "/x/alexburka/20170221/optocam/20/surface.png\n",
      "/x/alexburka/20170222/biocam/20/surface.png\n",
      "/x/alexburka/20170221/optocam/5/surface.png\n",
      "/x/alexburka/20170222/biocam/5/surface.png\n",
      "/x/alexburka/20170221/optocam/6/surface.png\n",
      "/x/alexburka/20170222/biocam/6/surface.png\n",
      "/x/alexburka/20170221/optocam/40/surface.png\n",
      "/x/alexburka/20170223/biocam/7/surface.png\n",
      "/x/alexburka/20170221/optocam/39/surface.png\n",
      "/x/alexburka/20170223/biocam/6/surface.png\n",
      "/x/alexburka/20170221/optocam/9/surface.png\n",
      "/x/alexburka/20170222/biocam/9/surface.png\n",
      "/x/alexburka/20170221/optocam/27/surface.png\n",
      "/x/alexburka/20170222/biocam/26/surface.png\n",
      "/x/alexburka/20170221/optocam/8/surface.png\n",
      "/x/alexburka/20170222/biocam/8/surface.png\n",
      "/x/alexburka/20170221/optocam/24/surface.png\n",
      "/x/alexburka/20170222/biocam/24/surface.png\n",
      "/x/alexburka/20170222/optocam/8/surface.png\n",
      "/x/alexburka/20170223/biocam/26/surface.png\n",
      "/x/alexburka/20170221/optocam/46/surface.png\n",
      "/x/alexburka/20170223/biocam/16/surface.png\n",
      "/x/alexburka/20170222/optocam/10/surface.png\n",
      "/x/alexburka/20170223/biocam/28/surface.png\n",
      "/x/alexburka/20170221/optocam/43/surface.png\n",
      "/x/alexburka/20170223/biocam/12/surface.png\n",
      "/x/alexburka/20170222/optocam/2/surface.png\n",
      "/x/alexburka/20170223/biocam/21/surface.png\n",
      "/x/alexburka/20170222/optocam/7/surface.png\n",
      "/x/alexburka/20170223/biocam/25/surface.png\n",
      "/x/alexburka/20170222/optocam/9/surface.png\n",
      "/x/alexburka/20170223/biocam/27/surface.png\n",
      "/x/alexburka/20170221/optocam/42/surface.png\n",
      "/x/alexburka/20170223/biocam/9/surface.png\n",
      "/x/alexburka/20170222/optocam/3/surface.png\n",
      "/x/alexburka/20170223/biocam/22/surface.png\n",
      "/x/alexburka/20170221/optocam/47/surface.png\n",
      "/x/alexburka/20170223/biocam/17/surface.png\n",
      "/x/alexburka/20170221/optocam/49/surface.png\n",
      "/x/alexburka/20170223/biocam/19/surface.png\n"
     ]
    }
   ],
   "source": [
    "# Original MINC network on our images\n",
    "\n",
    "import minc_model.copy_minc_labeled_surface_images\n",
    "import minc_model.test\n",
    "reload(minc_model)\n",
    "reload(minc_model.copy_minc_labeled_surface_images)\n",
    "\n",
    "minc_model.copy_minc_labeled_surface_images.go()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "googlenet images/brick/20170303_biocam_8.png stone 0.937407 False\n",
      "googlenet images/brick/20170303_optocam_9.png stone 0.634154 False\n",
      "googlenet images/brick/20170303_stickcam_15.png polishedstone 0.831786 False\n",
      "googlenet images/brick/20170407_optocam_10.png stone 0.774483 False\n",
      "googlenet images/brick/20170413_stickcam_10.png stone 0.641102 False\n",
      "googlenet images/brick/20170413_stickcam_12.png stone 0.69984 False\n",
      "googlenet images/brick/20170418_optocam_12.png painted 0.341726 False\n",
      "googlenet images/brick/20170418_optocam_15.png stone 0.227097 False\n",
      "googlenet images/brick/20170418_optocam_17.png stone 0.854124 False\n",
      "googlenet images/brick/20170418_optocam_18.png stone 0.360229 False\n",
      "googlenet images/brick/20170418_optocam_4.png stone 0.767872 False\n",
      "googlenet images/brick/20170419_stickcam_3.png stone 0.683206 False\n",
      "googlenet images/brick/20170419_stickcam_4.png stone 0.755649 False\n",
      "googlenet images/brick/20170419_stickcam_6.png carpet 0.919335 False\n",
      "googlenet images/brick/20170419_stickcam_7.png stone 0.714372 False\n",
      "googlenet images/brick/20170420_biocam_18.png stone 0.951562 False\n",
      "googlenet images/brick/20170420_biocam_6.png carpet 0.918977 False\n",
      "googlenet images/brick/20170420_biocam_7.png wallpaper 0.721795 False\n",
      "googlenet images/brick/20170420_biocam_8.png wood 0.663887 False\n",
      "googlenet images/brick/20170420_biocam_9.png carpet 0.373368 False\n",
      "googlenet images/carpet/20170221_optocam_41.png hair 0.918619 False\n",
      "googlenet images/carpet/20170221_stickcam_39.png hair 0.742087 False\n",
      "googlenet images/carpet/20170221_stickcam_48.png carpet 0.437515 True\n",
      "googlenet images/carpet/20170221_stickcam_52.png food 0.530583 False\n",
      "googlenet images/carpet/20170222_optocam_1.png food 0.648994 False\n",
      "googlenet images/carpet/20170222_optocam_5.png food 0.191521 False\n",
      "googlenet images/carpet/20170223_biocam_20.png food 0.778819 False\n",
      "googlenet images/carpet/20170223_biocam_24.png food 0.904417 False\n",
      "googlenet images/carpet/20170223_biocam_8.png hair 0.820276 False\n",
      "googlenet images/carpet/20170308_optocam_10.png stone 0.688181 False\n",
      "googlenet images/carpet/20170312_biocam_9.png stone 0.667016 False\n",
      "googlenet images/carpet/20170312_optocam_10.png stone 0.902265 False\n",
      "googlenet images/carpet/20170323_biocam_9.png carpet 0.797239 True\n",
      "googlenet images/carpet/20170323_optocam_5.png carpet 0.947855 True\n",
      "googlenet images/carpet/20170323_optocam_7.png stone 0.237229 False\n",
      "googlenet images/carpet/20170323_stickcam_7.png carpet 0.898793 True\n",
      "googlenet images/carpet/20170404_optocam_3.png carpet 0.940466 True\n",
      "googlenet images/carpet/20170404_stickcam_13.png fabric 0.704665 False\n",
      "googlenet images/carpet/20170406_biocam_8.png hair 0.203078 False\n",
      "googlenet images/carpet/20170418_optocam_1.png carpet 0.519426 True\n",
      "googlenet images/carpet/20170421_biocam_9.png other 0.630807 False\n",
      "googlenet images/carpet/20170421_optocam_9.png other 0.567235 False\n",
      "googlenet images/carpet/20170421_stickcam_11.png fabric 0.344217 False\n",
      "googlenet images/carpet/20170501_optocam_10.png stone 0.716393 False\n",
      "googlenet images/carpet/20170501_optocam_9.png glass 0.404863 False\n",
      "googlenet images/carpet/20170501_stickcam_8.png carpet 0.75828 True\n",
      "googlenet images/carpet/20170503_biocam_4.png carpet 0.546214 True\n",
      "googlenet images/carpet/20170503_biocam_5.png polishedstone 0.515694 False\n",
      "googlenet images/carpet/20170503_biocam_8.png stone 0.961872 False\n",
      "googlenet images/carpet/20170503_optocam_7.png foliage 0.832228 False\n",
      "googlenet images/carpet/20170503_stickcam_1.png glass 0.700285 False\n",
      "googlenet images/carpet/20170504_stickcam_10.png carpet 0.501892 True\n",
      "googlenet images/carpet/20170504_stickcam_7.png fabric 0.288537 False\n",
      "googlenet images/ceramic/20170303_stickcam_5.png glass 0.741875 False\n",
      "googlenet images/ceramic/20170312_biocam_6.png painted 0.451582 False\n",
      "googlenet images/ceramic/20170312_optocam_4.png painted 0.699576 False\n",
      "googlenet images/ceramic/20170504_stickcam_12.png carpet 0.657453 False\n",
      "googlenet images/fabric/20170221_optocam_11.png polishedstone 0.509256 False\n",
      "googlenet images/fabric/20170221_optocam_14.png carpet 0.467442 False\n",
      "googlenet images/fabric/20170221_optocam_27.png painted 0.769497 False\n",
      "googlenet images/fabric/20170221_optocam_3.png wallpaper 0.628532 False\n",
      "googlenet images/fabric/20170221_optocam_32.png carpet 0.74906 False\n",
      "googlenet images/fabric/20170221_optocam_33.png glass 0.725001 False\n",
      "googlenet images/fabric/20170221_optocam_34.png plastic 0.585343 False\n",
      "googlenet images/fabric/20170221_optocam_40.png wallpaper 0.528323 False\n",
      "googlenet images/fabric/20170221_optocam_44.png carpet 0.609234 False\n",
      "googlenet images/fabric/20170221_optocam_46.png carpet 0.418505 False\n",
      "googlenet images/fabric/20170221_optocam_48.png fabric 0.234019 True\n",
      "googlenet images/fabric/20170221_optocam_49.png food 0.735947 False\n",
      "googlenet images/fabric/20170221_optocam_7.png plastic 0.563125 False\n",
      "googlenet images/fabric/20170221_stickcam_12.png water 0.233916 False\n",
      "googlenet images/fabric/20170221_stickcam_15.png wallpaper 0.667112 False\n",
      "googlenet images/fabric/20170221_stickcam_3.png fabric 0.455064 True\n",
      "googlenet images/fabric/20170221_stickcam_31.png water 0.692121 False\n",
      "googlenet images/fabric/20170221_stickcam_33.png water 0.87698 False\n",
      "googlenet images/fabric/20170221_stickcam_42.png foliage 0.427924 False\n",
      "googlenet images/fabric/20170221_stickcam_46.png fabric 0.501508 True\n",
      "googlenet images/fabric/20170221_stickcam_8.png fabric 0.430878 True\n",
      "googlenet images/fabric/20170222_biocam_11.png plastic 0.513772 False\n",
      "googlenet images/fabric/20170222_biocam_14.png fabric 0.425003 True\n",
      "googlenet images/fabric/20170222_biocam_26.png painted 0.520331 False\n",
      "googlenet images/fabric/20170222_biocam_3.png fabric 0.668325 True\n",
      "googlenet images/fabric/20170222_biocam_30.png water 0.3351 False\n",
      "googlenet images/fabric/20170222_biocam_7.png plastic 0.868063 False\n",
      "googlenet images/fabric/20170222_optocam_10.png fabric 0.627353 True\n",
      "googlenet images/fabric/20170223_biocam_1.png plastic 0.746981 False\n",
      "googlenet images/fabric/20170223_biocam_14.png carpet 0.38988 False\n",
      "googlenet images/fabric/20170223_biocam_16.png hair 0.27863 False\n",
      "googlenet images/fabric/20170223_biocam_18.png fabric 0.381463 True\n",
      "googlenet images/fabric/20170223_biocam_19.png food 0.807433 False\n",
      "googlenet images/fabric/20170223_biocam_2.png fabric 0.500264 True\n",
      "googlenet images/fabric/20170223_biocam_28.png glass 0.279192 False\n",
      "googlenet images/fabric/20170223_biocam_7.png wallpaper 0.50991 False\n",
      "googlenet images/fabric/20170303_biocam_2.png tile 0.44912 False\n",
      "googlenet images/fabric/20170303_biocam_3.png wallpaper 0.53065 False\n",
      "googlenet images/fabric/20170303_optocam_1.png wallpaper 0.629615 False\n",
      "googlenet images/fabric/20170303_optocam_2.png fabric 0.782268 True\n",
      "googlenet images/fabric/20170303_stickcam_3.png wallpaper 0.635654 False\n",
      "googlenet images/fabric/20170303_stickcam_4.png fabric 0.772483 True\n",
      "googlenet images/fabric/20170304_biocam_2.png fabric 0.922374 True\n",
      "googlenet images/fabric/20170304_biocam_4.png carpet 0.691 False\n",
      "googlenet images/fabric/20170304_biocam_5.png wallpaper 0.990719 False\n",
      "googlenet images/fabric/20170304_biocam_8.png food 0.627399 False\n",
      "googlenet images/fabric/20170304_optocam_1.png fabric 0.785547 True\n",
      "googlenet images/fabric/20170304_optocam_2.png wallpaper 0.49117 False\n",
      "googlenet images/fabric/20170304_optocam_3.png wallpaper 0.764142 False\n",
      "googlenet images/fabric/20170304_optocam_5.png wallpaper 0.3764 False\n",
      "googlenet images/fabric/20170304_stickcam_1.png fabric 0.844537 True\n",
      "googlenet images/fabric/20170304_stickcam_5.png wallpaper 0.984598 False\n",
      "googlenet images/fabric/20170304_stickcam_6.png carpet 0.820368 False\n",
      "googlenet images/fabric/20170304_stickcam_9.png brick 0.239337 False\n",
      "googlenet images/fabric/20170306_biocam_5.png fabric 0.538066 True\n",
      "googlenet images/fabric/20170307_optocam_6.png fabric 0.794223 True\n",
      "googlenet images/fabric/20170307_stickcam_3.png fabric 0.927086 True\n",
      "googlenet images/fabric/20170308_optocam_16.png carpet 0.840928 False\n",
      "googlenet images/fabric/20170308_optocam_17.png carpet 0.611369 False\n",
      "googlenet images/fabric/20170308_optocam_18.png carpet 0.334032 False\n",
      "googlenet images/fabric/20170312_biocam_10.png fabric 0.757588 True\n",
      "googlenet images/fabric/20170312_biocam_4.png carpet 0.630168 False\n",
      "googlenet images/fabric/20170312_biocam_7.png carpet 0.959878 False\n",
      "googlenet images/fabric/20170312_biocam_8.png carpet 0.484052 False\n",
      "googlenet images/fabric/20170312_optocam_5.png carpet 0.993496 False\n",
      "googlenet images/fabric/20170312_optocam_7.png carpet 0.751006 False\n",
      "googlenet images/fabric/20170312_optocam_8.png fabric 0.729622 True\n",
      "googlenet images/fabric/20170312_stickcam_10.png carpet 0.99137 False\n",
      "googlenet images/fabric/20170312_stickcam_8.png wallpaper 0.958237 False\n",
      "googlenet images/fabric/20170323_biocam_7.png carpet 0.699621 False\n",
      "googlenet images/fabric/20170323_optocam_2.png carpet 0.752274 False\n",
      "googlenet images/fabric/20170323_stickcam_9.png carpet 0.845104 False\n",
      "googlenet images/fabric/20170403_stickcam_14.png paper 0.354549 False\n",
      "googlenet images/fabric/20170403_stickcam_15.png paper 0.960596 False\n",
      "googlenet images/fabric/20170403_stickcam_17.png carpet 0.666712 False\n",
      "googlenet images/fabric/20170404_stickcam_5.png plastic 0.369058 False\n",
      "googlenet images/fabric/20170404_stickcam_7.png paper 0.513509 False\n",
      "googlenet images/fabric/20170405_optocam_1.png carpet 0.913844 False\n",
      "googlenet images/fabric/20170405_optocam_2.png carpet 0.459498 False\n",
      "googlenet images/fabric/20170405_optocam_3.png wallpaper 0.258611 False\n",
      "googlenet images/fabric/20170405_optocam_4.png food 0.767367 False\n",
      "googlenet images/fabric/20170405_stickcam_6.png carpet 0.971227 False\n",
      "googlenet images/fabric/20170405_stickcam_7.png water 0.645287 False\n",
      "googlenet images/fabric/20170405_stickcam_8.png carpet 0.454525 False\n",
      "googlenet images/fabric/20170405_stickcam_9.png painted 0.26219 False\n",
      "googlenet images/fabric/20170421_biocam_5.png fabric 0.387568 True\n",
      "googlenet images/fabric/20170421_optocam_7.png carpet 0.969625 False\n",
      "googlenet images/fabric/20170421_stickcam_5.png carpet 0.971838 False\n",
      "googlenet images/fabric/20170426_stickcam_8.png fabric 0.94945 True\n",
      "googlenet images/fabric/20170426_stickcam_9.png carpet 0.731396 False\n",
      "googlenet images/fabric/20170427_biocam_4.png metal 0.515288 False\n",
      "googlenet images/fabric/20170427_biocam_5.png carpet 0.927288 False\n",
      "googlenet images/fabric/20170427_optocam_3.png fabric 0.741972 True\n",
      "googlenet images/fabric/20170427_optocam_4.png fabric 0.235678 True\n",
      "googlenet images/fabric/20170501_optocam_7.png wallpaper 0.965248 False\n",
      "googlenet images/fabric/20170501_optocam_8.png other 0.275224 False\n",
      "googlenet images/fabric/20170501_stickcam_6.png wallpaper 0.995927 False\n",
      "googlenet images/fabric/20170501_stickcam_7.png fabric 0.901466 True\n",
      "googlenet images/fabric/20170503_biocam_2.png fabric 0.933416 True\n",
      "googlenet images/fabric/20170503_biocam_3.png wallpaper 0.902557 False\n",
      "googlenet images/fabric/20170504_stickcam_8.png carpet 0.935697 False\n",
      "googlenet images/fabric/20170519_biocam_1.png fabric 0.747399 True\n",
      "googlenet images/fabric/20170519_biocam_2.png wallpaper 0.487002 False\n",
      "googlenet images/fabric/20170519_optocam_3.png fabric 0.569662 True\n",
      "googlenet images/fabric/20170519_optocam_4.png fabric 0.802265 True\n",
      "googlenet images/fabric/20170519_stickcam_8.png fabric 0.957911 True\n",
      "googlenet images/fabric/20170519_stickcam_9.png wallpaper 0.960854 False\n",
      "googlenet images/leather/20170221_optocam_10.png hair 0.329435 False\n",
      "googlenet images/leather/20170221_optocam_12.png fabric 0.619016 False\n",
      "googlenet images/leather/20170221_optocam_21.png carpet 0.897559 False\n",
      "googlenet images/leather/20170221_optocam_37.png wallpaper 0.840624 False\n",
      "googlenet images/leather/20170221_stickcam_11.png metal 0.175212 False\n",
      "googlenet images/leather/20170221_stickcam_13.png fabric 0.769335 False\n",
      "googlenet images/leather/20170221_stickcam_22.png food 0.358781 False\n",
      "googlenet images/leather/20170221_stickcam_26.png hair 0.283755 False\n",
      "googlenet images/leather/20170221_stickcam_36.png metal 0.307425 False\n",
      "googlenet images/leather/20170221_stickcam_57.png carpet 0.491693 False\n",
      "googlenet images/leather/20170222_biocam_10.png stone 0.453363 False\n",
      "googlenet images/leather/20170222_biocam_12.png plastic 0.43568 False\n",
      "googlenet images/leather/20170222_biocam_21.png water 0.774892 False\n",
      "googlenet images/leather/20170222_biocam_25.png hair 0.883368 False\n",
      "googlenet images/leather/20170222_optocam_11.png carpet 0.388584 False\n",
      "googlenet images/leather/20170222_optocam_2.png metal 0.325653 False\n",
      "googlenet images/leather/20170223_biocam_21.png metal 0.46179 False\n",
      "googlenet images/leather/20170223_biocam_30.png foliage 0.235768 False\n",
      "googlenet images/leather/20170223_biocam_5.png carpet 0.650742 False\n",
      "googlenet images/leather/20170304_biocam_3.png metal 0.28044 False\n",
      "googlenet images/leather/20170304_optocam_4.png metal 0.275711 False\n",
      "googlenet images/leather/20170304_stickcam_4.png leather 0.774153 True\n",
      "googlenet images/leather/20170312_biocam_3.png carpet 0.40854 False\n",
      "googlenet images/leather/20170312_optocam_2.png carpet 0.718289 False\n",
      "googlenet images/leather/20170404_stickcam_6.png plastic 0.687386 False\n",
      "googlenet images/leather/20170421_optocam_2.png plastic 0.414608 False\n",
      "googlenet images/leather/20170421_stickcam_14.png painted 0.575669 False\n",
      "googlenet images/leather/20170508_biocam_8.png carpet 0.292829 False\n",
      "googlenet images/leather/20170508_optocam_8.png stone 0.834878 False\n",
      "googlenet images/metal/20170222_optocam_3.png paper 0.758342 False\n",
      "googlenet images/metal/20170223_biocam_22.png plastic 0.270887 False\n",
      "googlenet images/metal/20170227_stickcam_11.png tile 0.223524 False\n",
      "googlenet images/metal/20170227_stickcam_12.png wood 0.931628 False\n",
      "googlenet images/metal/20170301_biocam_10.png wood 0.986814 False\n",
      "googlenet images/metal/20170301_biocam_11.png stone 0.620543 False\n",
      "googlenet images/metal/20170301_optocam_11.png tile 0.466 False\n",
      "googlenet images/metal/20170301_optocam_12.png wood 0.967267 False\n",
      "googlenet images/metal/20170303_biocam_10.png stone 0.742705 False\n",
      "googlenet images/metal/20170303_biocam_7.png glass 0.445856 False\n",
      "googlenet images/metal/20170303_optocam_4.png glass 0.82401 False\n",
      "googlenet images/metal/20170303_optocam_7.png metal 0.240644 True\n",
      "googlenet images/metal/20170303_stickcam_11.png glass 0.549075 False\n",
      "googlenet images/metal/20170303_stickcam_13.png plastic 0.319386 False\n",
      "googlenet images/metal/20170404_stickcam_4.png plastic 0.462313 False\n",
      "googlenet images/metal/20170406_biocam_13.png metal 0.571811 True\n",
      "googlenet images/metal/20170418_optocam_2.png metal 0.424445 True\n",
      "googlenet images/metal/20170421_biocam_2.png metal 0.650853 True\n",
      "googlenet images/metal/20170421_optocam_10.png metal 0.671287 True\n",
      "googlenet images/metal/20170421_stickcam_3.png painted 0.883779 False\n",
      "googlenet images/metal/20170421_stickcam_8.png plastic 0.352336 False\n",
      "googlenet images/metal/20170426_stickcam_13.png fabric 0.259609 False\n",
      "googlenet images/metal/20170427_biocam_8.png fabric 0.58354 False\n",
      "googlenet images/metal/20170427_optocam_8.png wallpaper 0.354734 False\n",
      "googlenet images/metal/20170501_optocam_6.png painted 0.296849 False\n",
      "googlenet images/metal/20170501_stickcam_5.png hair 0.280123 False\n",
      "googlenet images/metal/20170503_biocam_1.png carpet 0.629827 False\n",
      "googlenet images/metal/20170503_biocam_11.png glass 0.35521 False\n",
      "googlenet images/metal/20170503_stickcam_5.png carpet 0.247611 False\n",
      "googlenet images/other/20170221_optocam_20.png fabric 0.585984 False\n",
      "googlenet images/other/20170221_optocam_28.png tile 0.752186 False\n",
      "googlenet images/other/20170221_optocam_30.png metal 0.391819 False\n",
      "googlenet images/other/20170221_optocam_35.png water 0.347551 False\n",
      "googlenet images/other/20170221_optocam_42.png fabric 0.430259 False\n",
      "googlenet images/other/20170221_optocam_45.png fabric 0.831473 False\n",
      "googlenet images/other/20170221_optocam_5.png wallpaper 0.923411 False\n",
      "googlenet images/other/20170221_optocam_6.png glass 0.374183 False\n",
      "googlenet images/other/20170221_stickcam_28.png tile 0.688395 False\n",
      "googlenet images/other/20170221_stickcam_29.png metal 0.488512 False\n",
      "googlenet images/other/20170221_stickcam_34.png glass 0.237483 False\n",
      "googlenet images/other/20170221_stickcam_43.png water 0.48404 False\n",
      "googlenet images/other/20170221_stickcam_58.png plastic 0.720497 False\n",
      "googlenet images/other/20170222_biocam_20.png glass 0.480391 False\n",
      "googlenet images/other/20170222_biocam_27.png tile 0.570401 False\n",
      "googlenet images/other/20170222_biocam_28.png fabric 0.499768 False\n",
      "googlenet images/other/20170222_biocam_5.png wallpaper 0.981174 False\n",
      "googlenet images/other/20170222_biocam_6.png stone 0.306737 False\n",
      "googlenet images/other/20170222_optocam_12.png plastic 0.545157 False\n",
      "googlenet images/other/20170222_optocam_8.png tile 0.241356 False\n",
      "googlenet images/other/20170223_biocam_15.png fabric 0.741765 False\n",
      "googlenet images/other/20170223_biocam_26.png fabric 0.444864 False\n",
      "googlenet images/other/20170223_biocam_3.png plastic 0.757442 False\n",
      "googlenet images/other/20170223_biocam_31.png plastic 0.418348 False\n",
      "googlenet images/other/20170223_biocam_9.png fabric 0.414017 False\n",
      "googlenet images/other/20170227_stickcam_10.png fabric 0.916834 False\n",
      "googlenet images/other/20170301_biocam_12.png fabric 0.871237 False\n",
      "googlenet images/other/20170301_optocam_10.png fabric 0.835007 False\n",
      "googlenet images/other/20170303_biocam_1.png fabric 0.879122 False\n",
      "googlenet images/other/20170303_biocam_5.png painted 0.33225 False\n",
      "googlenet images/other/20170303_optocam_10.png hair 0.645626 False\n",
      "googlenet images/other/20170303_optocam_3.png fabric 0.518218 False\n",
      "googlenet images/other/20170303_optocam_6.png metal 0.407291 False\n",
      "googlenet images/other/20170303_stickcam_1.png carpet 0.37463 False\n",
      "googlenet images/other/20170303_stickcam_12.png mirror 0.390946 False\n",
      "googlenet images/other/20170303_stickcam_2.png metal 0.269738 False\n",
      "googlenet images/other/20170303_stickcam_9.png fabric 0.480747 False\n",
      "googlenet images/other/20170304_biocam_1.png stone 0.775679 False\n",
      "googlenet images/other/20170304_biocam_6.png food 0.539028 False\n",
      "googlenet images/other/20170304_biocam_7.png food 0.9274 False\n",
      "googlenet images/other/20170304_optocam_6.png food 0.952876 False\n",
      "googlenet images/other/20170304_optocam_7.png foliage 0.51328 False\n",
      "googlenet images/other/20170304_stickcam_7.png foliage 0.517448 False\n",
      "googlenet images/other/20170304_stickcam_8.png tile 0.540175 False\n",
      "googlenet images/other/20170306_biocam_1.png polishedstone 0.399475 False\n",
      "googlenet images/other/20170306_biocam_2.png polishedstone 0.853739 False\n",
      "googlenet images/other/20170306_biocam_4.png tile 0.528299 False\n",
      "googlenet images/other/20170306_biocam_8.png polishedstone 0.690923 False\n",
      "googlenet images/other/20170307_optocam_1.png polishedstone 0.522918 False\n",
      "googlenet images/other/20170307_optocam_2.png food 0.378887 False\n",
      "googlenet images/other/20170307_optocam_8.png tile 0.740344 False\n",
      "googlenet images/other/20170307_optocam_9.png foliage 0.535006 False\n",
      "googlenet images/other/20170307_stickcam_1.png tile 0.834732 False\n",
      "googlenet images/other/20170307_stickcam_6.png polishedstone 0.265076 False\n",
      "googlenet images/other/20170307_stickcam_7.png polishedstone 0.463227 False\n",
      "googlenet images/other/20170307_stickcam_8.png fabric 0.410002 False\n",
      "googlenet images/other/20170312_biocam_2.png fabric 0.26818 False\n",
      "googlenet images/other/20170312_optocam_9.png skin 0.729202 False\n",
      "googlenet images/other/20170312_stickcam_1.png skin 0.835169 False\n",
      "googlenet images/other/20170321_optocam_1.png wallpaper 0.830385 False\n",
      "googlenet images/other/20170323_biocam_1.png carpet 0.607927 False\n",
      "googlenet images/other/20170323_stickcam_5.png carpet 0.680532 False\n",
      "googlenet images/other/20170403_stickcam_12.png fabric 0.916672 False\n",
      "googlenet images/other/20170403_stickcam_2.png stone 0.446326 False\n",
      "googlenet images/other/20170403_stickcam_3.png paper 0.39124 False\n",
      "googlenet images/other/20170404_stickcam_1.png carpet 0.687867 False\n",
      "googlenet images/other/20170404_stickcam_9.png paper 0.781599 False\n",
      "googlenet images/other/20170421_biocam_8.png stone 0.236738 False\n",
      "googlenet images/other/20170421_stickcam_18.png polishedstone 0.583412 False\n",
      "googlenet images/other/20170503_biocam_9.png other 0.687632 True\n",
      "googlenet images/other/20170503_stickcam_2.png wallpaper 0.254634 False\n",
      "googlenet images/other/20170508_biocam_5.png carpet 0.409424 False\n",
      "googlenet images/other/20170508_optocam_4.png fabric 0.845164 False\n",
      "googlenet images/other/20170508_optocam_9.png wood 0.8116 False\n",
      "googlenet images/other/20170519_biocam_3.png carpet 0.583471 False\n",
      "googlenet images/other/20170519_optocam_2.png polishedstone 0.443507 False\n",
      "googlenet images/other/20170519_stickcam_10.png carpet 0.460724 False\n",
      "googlenet images/painted/20170221_optocam_43.png painted 0.292167 True\n",
      "googlenet images/painted/20170223_biocam_12.png plastic 0.656168 False\n",
      "googlenet images/painted/20170227_stickcam_3.png fabric 0.439271 False\n",
      "googlenet images/painted/20170227_stickcam_5.png painted 0.719299 True\n",
      "googlenet images/painted/20170301_biocam_4.png fabric 0.840657 False\n",
      "googlenet images/painted/20170301_biocam_7.png painted 0.602946 True\n",
      "googlenet images/painted/20170301_optocam_1.png wood 0.437164 False\n",
      "googlenet images/painted/20170301_optocam_5.png painted 0.935702 True\n",
      "googlenet images/painted/20170312_biocam_5.png carpet 0.996806 False\n",
      "googlenet images/painted/20170312_optocam_3.png wood 0.703147 False\n",
      "googlenet images/painted/20170312_stickcam_5.png painted 0.575091 True\n",
      "googlenet images/painted/20170321_optocam_2.png other 0.231779 False\n",
      "googlenet images/painted/20170323_biocam_3.png plastic 0.348535 False\n",
      "googlenet images/painted/20170323_biocam_5.png metal 0.36569 False\n",
      "googlenet images/painted/20170323_optocam_1.png plastic 0.263505 False\n",
      "googlenet images/painted/20170323_optocam_9.png plastic 0.564555 False\n",
      "googlenet images/painted/20170323_stickcam_1.png plastic 0.621431 False\n",
      "googlenet images/painted/20170323_stickcam_11.png hair 0.338697 False\n",
      "googlenet images/painted/20170323_stickcam_4.png plastic 0.521886 False\n",
      "googlenet images/painted/20170404_optocam_2.png leather 0.320048 False\n",
      "googlenet images/painted/20170404_optocam_6.png plastic 0.361648 False\n",
      "googlenet images/painted/20170404_stickcam_12.png glass 0.478778 False\n",
      "googlenet images/painted/20170404_stickcam_2.png paper 0.991028 False\n",
      "googlenet images/painted/20170405_stickcam_3.png fabric 0.397326 False\n",
      "googlenet images/painted/20170421_biocam_10.png tile 0.204556 False\n",
      "googlenet images/painted/20170421_optocam_11.png painted 0.737719 True\n",
      "googlenet images/painted/20170421_optocam_6.png paper 0.283417 False\n",
      "googlenet images/painted/20170421_stickcam_12.png painted 0.681117 True\n",
      "googlenet images/painted/20170421_stickcam_13.png fabric 0.41651 False\n",
      "googlenet images/painted/20170426_stickcam_10.png painted 0.363098 True\n",
      "googlenet images/painted/20170427_biocam_7.png painted 0.775267 True\n",
      "googlenet images/painted/20170427_optocam_7.png plastic 0.365263 False\n",
      "googlenet images/painted/20170503_biocam_12.png water 0.259755 False\n",
      "googlenet images/painted/20170503_optocam_3.png water 0.528017 False\n",
      "googlenet images/painted/20170503_stickcam_6.png stone 0.77106 False\n",
      "googlenet images/painted/20170516_stickcam_10.png stone 0.453417 False\n",
      "googlenet images/painted/20170516_stickcam_9.png wallpaper 0.34782 False\n",
      "googlenet images/painted/20170518_biocam_4.png stone 0.359 False\n",
      "googlenet images/painted/20170518_optocam_2.png tile 0.29423 False\n",
      "googlenet images/painted/20170518_stickcam_1.png stone 0.682733 False\n",
      "googlenet images/painted/20170519_biocam_4.png glass 0.978031 False\n",
      "googlenet images/painted/20170519_optocam_1.png stone 0.981566 False\n",
      "googlenet images/painted/20170519_stickcam_11.png stone 0.796715 False\n",
      "googlenet images/paper/20170221_optocam_22.png stone 0.377539 False\n",
      "googlenet images/paper/20170221_optocam_23.png paper 0.571694 True\n",
      "googlenet images/paper/20170221_optocam_24.png sky 0.571121 False\n",
      "googlenet images/paper/20170221_optocam_36.png fabric 0.852779 False\n",
      "googlenet images/paper/20170221_optocam_39.png sky 0.660432 False\n",
      "googlenet images/paper/20170221_optocam_4.png glass 0.658171 False\n",
      "googlenet images/paper/20170221_optocam_8.png sky 0.668355 False\n",
      "googlenet images/paper/20170221_optocam_9.png sky 0.581469 False\n",
      "googlenet images/paper/20170221_stickcam_23.png stone 0.278175 False\n",
      "googlenet images/paper/20170221_stickcam_24.png paper 0.467708 True\n",
      "googlenet images/paper/20170221_stickcam_35.png fabric 0.732173 False\n",
      "googlenet images/paper/20170221_stickcam_51.png wallpaper 0.566838 False\n",
      "googlenet images/paper/20170222_biocam_22.png stone 0.346648 False\n",
      "googlenet images/paper/20170222_biocam_23.png paper 0.720021 True\n",
      "googlenet images/paper/20170222_biocam_24.png painted 0.235313 False\n",
      "googlenet images/paper/20170222_biocam_4.png tile 0.374952 False\n",
      "googlenet images/paper/20170222_biocam_8.png plastic 0.258144 False\n",
      "googlenet images/paper/20170222_biocam_9.png plastic 0.328591 False\n",
      "googlenet images/paper/20170222_optocam_4.png wallpaper 0.281338 False\n",
      "googlenet images/paper/20170222_optocam_7.png painted 0.796668 False\n",
      "googlenet images/paper/20170222_optocam_9.png glass 0.89644 False\n",
      "googlenet images/paper/20170223_biocam_23.png wallpaper 0.829336 False\n",
      "googlenet images/paper/20170223_biocam_25.png plastic 0.70798 False\n",
      "googlenet images/paper/20170223_biocam_27.png glass 0.571918 False\n",
      "googlenet images/paper/20170223_biocam_4.png fabric 0.764395 False\n",
      "googlenet images/paper/20170223_biocam_6.png sky 0.613228 False\n",
      "googlenet images/paper/20170306_biocam_6.png plastic 0.483544 False\n",
      "googlenet images/paper/20170306_biocam_7.png plastic 0.410961 False\n",
      "googlenet images/paper/20170307_optocam_4.png fabric 0.291681 False\n",
      "googlenet images/paper/20170307_optocam_5.png plastic 0.369034 False\n",
      "googlenet images/paper/20170307_stickcam_4.png carpet 0.536219 False\n",
      "googlenet images/paper/20170307_stickcam_5.png plastic 0.27191 False\n",
      "googlenet images/paper/20170403_stickcam_10.png paper 0.641842 True\n",
      "googlenet images/paper/20170403_stickcam_11.png paper 0.363912 True\n",
      "googlenet images/paper/20170403_stickcam_13.png paper 0.606235 True\n",
      "googlenet images/paper/20170403_stickcam_16.png paper 0.576896 True\n",
      "googlenet images/paper/20170403_stickcam_4.png paper 0.630342 True\n",
      "googlenet images/paper/20170404_stickcam_10.png paper 0.975176 True\n",
      "googlenet images/paper/20170404_stickcam_8.png paper 0.734693 True\n",
      "googlenet images/paper/20170426_stickcam_6.png wallpaper 0.507239 False\n",
      "googlenet images/paper/20170426_stickcam_7.png wallpaper 0.463823 False\n",
      "googlenet images/paper/20170427_biocam_1.png stone 0.916456 False\n",
      "googlenet images/paper/20170427_optocam_1.png wallpaper 0.590581 False\n",
      "googlenet images/paper/20170427_optocam_2.png carpet 0.818145 False\n",
      "googlenet images/paper/20170508_biocam_1.png metal 0.251025 False\n",
      "googlenet images/paper/20170508_biocam_11.png fabric 0.846964 False\n",
      "googlenet images/paper/20170508_biocam_12.png paper 0.914311 True\n",
      "googlenet images/paper/20170508_biocam_2.png fabric 0.33515 False\n",
      "googlenet images/paper/20170508_biocam_3.png fabric 0.864311 False\n",
      "googlenet images/paper/20170508_biocam_6.png plastic 0.45231 False\n",
      "googlenet images/paper/20170508_biocam_7.png metal 0.296638 False\n",
      "googlenet images/paper/20170508_biocam_9.png painted 0.506916 False\n",
      "googlenet images/paper/20170508_optocam_1.png painted 0.697963 False\n",
      "googlenet images/paper/20170508_optocam_10.png fabric 0.735127 False\n",
      "googlenet images/paper/20170508_optocam_11.png fabric 0.335792 False\n",
      "googlenet images/paper/20170508_optocam_12.png plastic 0.719693 False\n",
      "googlenet images/paper/20170508_optocam_2.png skin 0.858353 False\n",
      "googlenet images/paper/20170508_optocam_3.png wallpaper 0.825333 False\n",
      "googlenet images/paper/20170508_optocam_5.png water 0.521983 False\n",
      "googlenet images/paper/20170508_optocam_6.png plastic 0.638383 False\n",
      "googlenet images/paper/20170508_optocam_7.png plastic 0.323852 False\n",
      "googlenet images/plastic/20170221_optocam_1.png carpet 0.495575 False\n",
      "googlenet images/plastic/20170221_optocam_2.png metal 0.53782 False\n",
      "googlenet images/plastic/20170221_stickcam_1.png food 0.396737 False\n",
      "googlenet images/plastic/20170221_stickcam_2.png metal 0.518282 False\n",
      "googlenet images/plastic/20170222_biocam_1.png plastic 0.534373 True\n",
      "googlenet images/plastic/20170222_biocam_2.png metal 0.524859 False\n",
      "googlenet images/plastic/20170227_stickcam_1.png plastic 0.204394 True\n",
      "googlenet images/plastic/20170227_stickcam_6.png painted 0.716373 False\n",
      "googlenet images/plastic/20170227_stickcam_8.png painted 0.359764 False\n",
      "googlenet images/plastic/20170301_biocam_14.png fabric 0.230302 False\n",
      "googlenet images/plastic/20170301_biocam_17.png painted 0.497266 False\n",
      "googlenet images/plastic/20170301_biocam_5.png fabric 0.327971 False\n",
      "googlenet images/plastic/20170301_biocam_8.png carpet 0.738056 False\n",
      "googlenet images/plastic/20170301_optocam_3.png painted 0.345827 False\n",
      "googlenet images/plastic/20170301_optocam_4.png painted 0.273282 False\n",
      "googlenet images/plastic/20170301_optocam_7.png hair 0.292099 False\n",
      "googlenet images/plastic/20170308_optocam_13.png food 0.553811 False\n",
      "googlenet images/plastic/20170308_optocam_8.png water 0.233015 False\n",
      "googlenet images/plastic/20170404_optocam_1.png carpet 0.653327 False\n",
      "googlenet images/plastic/20170404_stickcam_11.png plastic 0.427841 True\n",
      "googlenet images/plastic/20170406_biocam_11.png stone 0.45382 False\n",
      "googlenet images/plastic/20170413_stickcam_15.png polishedstone 0.544963 False\n",
      "googlenet images/plastic/20170418_optocam_7.png hair 0.217288 False\n",
      "googlenet images/plastic/20170420_biocam_15.png plastic 0.468516 True\n",
      "googlenet images/plastic/20170420_optocam_1.png painted 0.640878 False\n",
      "googlenet images/plastic/20170421_biocam_12.png tile 0.366837 False\n",
      "googlenet images/plastic/20170421_biocam_13.png painted 0.708834 False\n",
      "googlenet images/plastic/20170421_biocam_7.png painted 0.405539 False\n",
      "googlenet images/plastic/20170421_optocam_1.png wallpaper 0.346151 False\n",
      "googlenet images/plastic/20170421_optocam_13.png ceramic 0.436356 False\n",
      "googlenet images/plastic/20170421_optocam_16.png polishedstone 0.500608 False\n",
      "googlenet images/plastic/20170421_optocam_5.png painted 0.92115 False\n",
      "googlenet images/plastic/20170421_optocam_8.png stone 0.98476 False\n",
      "googlenet images/plastic/20170421_stickcam_10.png stone 0.685717 False\n",
      "googlenet images/plastic/20170421_stickcam_16.png painted 0.918544 False\n",
      "googlenet images/plastic/20170421_stickcam_17.png plastic 0.431244 True\n",
      "googlenet images/plastic/20170421_stickcam_4.png painted 0.943399 False\n",
      "googlenet images/plastic/20170421_stickcam_6.png stone 0.634961 False\n",
      "googlenet images/plastic/20170426_stickcam_12.png plastic 0.731988 True\n",
      "googlenet images/plastic/20170427_biocam_6.png fabric 0.525225 False\n",
      "googlenet images/plastic/20170427_optocam_6.png glass 0.316249 False\n",
      "googlenet images/plastic/20170503_biocam_7.png stone 0.785311 False\n",
      "googlenet images/plastic/20170503_optocam_5.png stone 0.290307 False\n",
      "googlenet images/plastic/20170503_stickcam_3.png stone 0.671197 False\n",
      "googlenet images/plastic/20170504_stickcam_11.png painted 0.294955 False\n",
      "googlenet images/plastic/20170504_stickcam_9.png tile 0.307636 False\n",
      "googlenet images/plastic/20170516_stickcam_5.png wallpaper 0.587529 False\n",
      "googlenet images/plastic/20170518_biocam_1.png hair 0.393439 False\n",
      "googlenet images/plastic/20170518_optocam_5.png leather 0.289848 False\n",
      "googlenet images/plastic/20170518_stickcam_4.png hair 0.398296 False\n",
      "googlenet images/plastic/20170519_biocam_11.png wood 0.533934 False\n",
      "googlenet images/plastic/20170519_biocam_6.png tile 0.968526 False\n",
      "googlenet images/plastic/20170519_biocam_9.png ceramic 0.82519 False\n",
      "googlenet images/plastic/20170519_optocam_11.png fabric 0.553564 False\n",
      "googlenet images/plastic/20170519_optocam_6.png tile 0.949965 False\n",
      "googlenet images/plastic/20170519_optocam_9.png fabric 0.622502 False\n",
      "googlenet images/plastic/20170519_stickcam_3.png wallpaper 0.977144 False\n",
      "googlenet images/plastic/20170519_stickcam_5.png wood 0.525135 False\n",
      "googlenet images/plastic/20170519_stickcam_7.png tile 0.943178 False\n",
      "googlenet images/polishedstone/20170308_optocam_6.png carpet 0.2045 False\n",
      "googlenet images/polishedstone/20170405_optocam_10.png hair 0.541083 False\n",
      "googlenet images/polishedstone/20170405_optocam_8.png stone 0.44569 False\n",
      "googlenet images/polishedstone/20170405_optocam_9.png stone 0.684656 False\n",
      "googlenet images/polishedstone/20170406_biocam_12.png stone 0.398743 False\n",
      "googlenet images/polishedstone/20170406_biocam_5.png polishedstone 0.35107 True\n",
      "googlenet images/polishedstone/20170406_biocam_6.png stone 0.687518 False\n",
      "googlenet images/polishedstone/20170406_biocam_7.png hair 0.258166 False\n",
      "googlenet images/polishedstone/20170406_stickcam_12.png tile 0.54552 False\n",
      "googlenet images/polishedstone/20170406_stickcam_13.png stone 0.951894 False\n",
      "googlenet images/polishedstone/20170406_stickcam_14.png leather 0.18669 False\n",
      "googlenet images/polishedstone/20170413_stickcam_11.png polishedstone 0.71753 True\n",
      "googlenet images/polishedstone/20170418_optocam_10.png stone 0.580218 False\n",
      "googlenet images/polishedstone/20170418_optocam_16.png polishedstone 0.597809 True\n",
      "googlenet images/polishedstone/20170418_optocam_3.png polishedstone 0.951769 True\n",
      "googlenet images/polishedstone/20170418_optocam_8.png polishedstone 0.808073 True\n",
      "googlenet images/polishedstone/20170419_stickcam_18.png polishedstone 0.770337 True\n",
      "googlenet images/polishedstone/20170419_stickcam_2.png polishedstone 0.382344 True\n",
      "googlenet images/polishedstone/20170419_stickcam_5.png carpet 0.446418 False\n",
      "googlenet images/polishedstone/20170420_biocam_10.png stone 0.725938 False\n",
      "googlenet images/polishedstone/20170420_biocam_13.png polishedstone 0.769451 True\n",
      "googlenet images/polishedstone/20170420_biocam_17.png stone 0.397292 False\n",
      "googlenet images/polishedstone/20170420_biocam_3.png stone 0.756063 False\n",
      "googlenet images/polishedstone/20170420_biocam_5.png stone 0.617751 False\n",
      "googlenet images/polishedstone/20170421_biocam_3.png carpet 0.35923 False\n",
      "googlenet images/polishedstone/20170421_optocam_3.png wallpaper 0.591401 False\n",
      "googlenet images/polishedstone/20170421_stickcam_2.png stone 0.902958 False\n",
      "googlenet images/polishedstone/20170516_stickcam_2.png food 0.697785 False\n",
      "googlenet images/polishedstone/20170516_stickcam_4.png glass 0.434658 False\n",
      "googlenet images/polishedstone/20170516_stickcam_8.png stone 0.63919 False\n",
      "googlenet images/polishedstone/20170519_biocam_5.png stone 0.628565 False\n",
      "googlenet images/polishedstone/20170519_optocam_5.png stone 0.874122 False\n",
      "googlenet images/polishedstone/20170519_stickcam_6.png stone 0.824665 False\n",
      "googlenet images/stone/20170303_stickcam_8.png metal 0.836132 False\n",
      "googlenet images/stone/20170405_optocam_7.png stone 0.916329 True\n",
      "googlenet images/stone/20170406_biocam_4.png stone 0.940946 True\n",
      "googlenet images/stone/20170406_stickcam_11.png stone 0.917035 True\n",
      "googlenet images/stone/20170407_optocam_6.png stone 0.453647 True\n",
      "googlenet images/stone/20170407_optocam_7.png carpet 0.580682 False\n",
      "googlenet images/stone/20170407_optocam_8.png stone 0.723534 True\n",
      "googlenet images/stone/20170407_optocam_9.png food 0.907336 False\n",
      "googlenet images/stone/20170413_stickcam_13.png stone 0.263171 True\n",
      "googlenet images/stone/20170413_stickcam_6.png stone 0.486323 True\n",
      "googlenet images/stone/20170413_stickcam_7.png stone 0.823754 True\n",
      "googlenet images/stone/20170413_stickcam_8.png carpet 0.959323 False\n",
      "googlenet images/stone/20170413_stickcam_9.png food 0.665595 False\n",
      "googlenet images/stone/20170418_optocam_11.png stone 0.994656 True\n",
      "googlenet images/stone/20170418_optocam_13.png stone 0.958996 True\n",
      "googlenet images/stone/20170418_optocam_14.png stone 0.867406 True\n",
      "googlenet images/stone/20170418_optocam_5.png stone 0.59646 True\n",
      "googlenet images/stone/20170418_optocam_9.png stone 0.501989 True\n",
      "googlenet images/stone/20170419_stickcam_1.png stone 0.772772 True\n",
      "googlenet images/stone/20170420_biocam_14.png polishedstone 0.615123 False\n",
      "googlenet images/stone/20170420_biocam_22.png wallpaper 0.32627 False\n",
      "googlenet images/stone/20170420_biocam_23.png polishedstone 0.490341 False\n",
      "googlenet images/stone/20170421_biocam_6.png stone 0.554819 True\n",
      "googlenet images/stone/20170421_optocam_4.png stone 0.379895 True\n",
      "googlenet images/stone/20170421_stickcam_15.png carpet 0.363735 False\n",
      "googlenet images/stone/20170501_optocam_4.png wallpaper 0.45924 False\n",
      "googlenet images/stone/20170501_stickcam_10.png glass 0.737573 False\n",
      "googlenet images/stone/20170502_biocam_3.png stone 0.710217 True\n",
      "googlenet images/stone/20170516_stickcam_3.png wallpaper 0.47183 False\n",
      "googlenet images/stone/20170518_biocam_5.png wood 0.428361 False\n",
      "googlenet images/stone/20170518_optocam_1.png plastic 0.381097 False\n",
      "googlenet images/stone/20170518_stickcam_5.png glass 0.330719 False\n",
      "googlenet images/tile/20170221_optocam_15.png glass 0.854532 False\n",
      "googlenet images/tile/20170221_optocam_16.png stone 0.476802 False\n",
      "googlenet images/tile/20170221_optocam_17.png glass 0.71066 False\n",
      "googlenet images/tile/20170221_optocam_18.png plastic 0.5733 False\n",
      "googlenet images/tile/20170221_optocam_19.png wallpaper 0.800007 False\n",
      "googlenet images/tile/20170221_optocam_31.png polishedstone 0.44108 False\n",
      "googlenet images/tile/20170221_stickcam_30.png stone 0.296956 False\n",
      "googlenet images/tile/20170221_stickcam_59.png food 0.330639 False\n",
      "googlenet images/tile/20170222_biocam_15.png glass 0.855324 False\n",
      "googlenet images/tile/20170222_biocam_16.png glass 0.435635 False\n",
      "googlenet images/tile/20170222_biocam_17.png water 0.490546 False\n",
      "googlenet images/tile/20170222_biocam_18.png wallpaper 0.384382 False\n",
      "googlenet images/tile/20170222_biocam_19.png glass 0.748776 False\n",
      "googlenet images/tile/20170222_biocam_29.png stone 0.548992 False\n",
      "googlenet images/tile/20170222_optocam_13.png water 0.474286 False\n",
      "googlenet images/tile/20170223_biocam_32.png water 0.463402 False\n",
      "googlenet images/tile/20170227_stickcam_13.png food 0.362631 False\n",
      "googlenet images/tile/20170301_biocam_13.png polishedstone 0.52182 False\n",
      "googlenet images/tile/20170301_optocam_13.png polishedstone 0.712446 False\n",
      "googlenet images/tile/20170303_stickcam_6.png tile 0.290958 True\n",
      "googlenet images/tile/20170303_stickcam_7.png polishedstone 0.433763 False\n",
      "googlenet images/tile/20170308_optocam_11.png stone 0.78904 False\n",
      "googlenet images/tile/20170308_optocam_12.png stone 0.720886 False\n",
      "googlenet images/tile/20170308_optocam_14.png carpet 0.816423 False\n",
      "googlenet images/tile/20170308_optocam_15.png stone 0.678878 False\n",
      "googlenet images/tile/20170403_stickcam_5.png polishedstone 0.871127 False\n",
      "googlenet images/tile/20170403_stickcam_6.png stone 0.453163 False\n",
      "googlenet images/tile/20170403_stickcam_7.png polishedstone 0.382585 False\n",
      "googlenet images/tile/20170403_stickcam_8.png polishedstone 0.683025 False\n",
      "googlenet images/tile/20170403_stickcam_9.png polishedstone 0.4861 False\n",
      "googlenet images/tile/20170404_optocam_4.png stone 0.639413 False\n",
      "googlenet images/tile/20170404_optocam_5.png carpet 0.83903 False\n",
      "googlenet images/tile/20170405_optocam_5.png tile 0.888935 True\n",
      "googlenet images/tile/20170405_optocam_6.png metal 0.206737 False\n",
      "googlenet images/tile/20170405_stickcam_4.png wallpaper 0.223934 False\n",
      "googlenet images/tile/20170405_stickcam_5.png wallpaper 0.54029 False\n",
      "googlenet images/tile/20170406_biocam_14.png stone 0.906933 False\n",
      "googlenet images/tile/20170406_biocam_2.png polishedstone 0.556875 False\n",
      "googlenet images/tile/20170406_biocam_3.png stone 0.568254 False\n",
      "googlenet images/tile/20170406_stickcam_5.png polishedstone 0.50012 False\n",
      "googlenet images/tile/20170406_stickcam_6.png stone 0.488351 False\n",
      "googlenet images/tile/20170407_optocam_3.png carpet 0.687705 False\n",
      "googlenet images/tile/20170407_optocam_4.png stone 0.498768 False\n",
      "googlenet images/tile/20170407_optocam_5.png stone 0.457103 False\n",
      "googlenet images/tile/20170413_stickcam_3.png tile 0.421265 True\n",
      "googlenet images/tile/20170413_stickcam_4.png stone 0.559479 False\n",
      "googlenet images/tile/20170413_stickcam_5.png tile 0.594884 True\n",
      "googlenet images/tile/20170419_optocam_1.png stone 0.891387 False\n",
      "googlenet images/tile/20170419_optocam_2.png hair 0.510038 False\n",
      "googlenet images/tile/20170420_biocam_1.png stone 0.807019 False\n",
      "googlenet images/tile/20170420_biocam_19.png stone 0.50065 False\n",
      "googlenet images/tile/20170420_biocam_2.png stone 0.951939 False\n",
      "googlenet images/tile/20170420_biocam_20.png carpet 0.387656 False\n",
      "googlenet images/tile/20170420_biocam_21.png carpet 0.374707 False\n",
      "googlenet images/tile/20170420_optocam_2.png stone 0.822639 False\n",
      "googlenet images/tile/20170501_optocam_12.png glass 0.602813 False\n",
      "googlenet images/tile/20170501_optocam_2.png stone 0.305217 False\n",
      "googlenet images/tile/20170501_optocam_3.png tile 0.532894 True\n",
      "googlenet images/tile/20170501_optocam_5.png glass 0.579677 False\n",
      "googlenet images/tile/20170501_stickcam_11.png glass 0.616332 False\n",
      "googlenet images/tile/20170501_stickcam_12.png stone 0.331953 False\n",
      "googlenet images/tile/20170501_stickcam_14.png carpet 0.63902 False\n",
      "googlenet images/tile/20170501_stickcam_4.png fabric 0.378443 False\n",
      "googlenet images/tile/20170502_biocam_2.png wallpaper 0.377771 False\n",
      "googlenet images/tile/20170502_biocam_4.png stone 0.505968 False\n",
      "googlenet images/tile/20170502_biocam_5.png plastic 0.240207 False\n",
      "googlenet images/tile/20170502_biocam_6.png polishedstone 0.273665 False\n",
      "googlenet images/tile/20170503_biocam_6.png stone 0.659926 False\n",
      "googlenet images/tile/20170504_stickcam_3.png polishedstone 0.703577 False\n",
      "googlenet images/tile/20170504_stickcam_4.png polishedstone 0.48851 False\n",
      "googlenet images/tile/20170504_stickcam_5.png polishedstone 0.572939 False\n",
      "googlenet images/tile/20170504_stickcam_6.png glass 0.785471 False\n",
      "googlenet images/wood/20170221_optocam_13.png hair 0.304219 False\n",
      "googlenet images/wood/20170221_optocam_47.png wallpaper 0.384737 False\n",
      "googlenet images/wood/20170221_stickcam_14.png polishedstone 0.262667 False\n",
      "googlenet images/wood/20170222_biocam_13.png wood 0.30052 True\n",
      "googlenet images/wood/20170223_biocam_17.png ceramic 0.746645 False\n",
      "googlenet images/wood/20170227_stickcam_2.png carpet 0.839484 False\n",
      "googlenet images/wood/20170227_stickcam_4.png fabric 0.486064 False\n",
      "googlenet images/wood/20170227_stickcam_7.png fabric 0.579408 False\n",
      "googlenet images/wood/20170301_biocam_1.png fabric 0.57595 False\n",
      "googlenet images/wood/20170301_biocam_16.png wallpaper 0.822245 False\n",
      "googlenet images/wood/20170301_biocam_3.png carpet 0.455285 False\n",
      "googlenet images/wood/20170301_optocam_15.png fabric 0.742206 False\n",
      "googlenet images/wood/20170301_optocam_2.png wallpaper 0.620085 False\n",
      "googlenet images/wood/20170301_optocam_6.png carpet 0.408697 False\n",
      "googlenet images/wood/20170303_biocam_9.png polishedstone 0.981429 False\n",
      "googlenet images/wood/20170303_optocam_5.png polishedstone 0.968919 False\n",
      "googlenet images/wood/20170303_stickcam_14.png polishedstone 0.962406 False\n",
      "googlenet images/wood/20170306_biocam_3.png stone 0.317097 False\n",
      "googlenet images/wood/20170307_optocam_7.png stone 0.796209 False\n",
      "googlenet images/wood/20170308_optocam_7.png wood 0.305214 True\n",
      "googlenet images/wood/20170312_biocam_1.png stone 0.496451 False\n",
      "googlenet images/wood/20170323_biocam_10.png painted 0.324097 False\n",
      "googlenet images/wood/20170323_biocam_2.png carpet 0.309084 False\n",
      "googlenet images/wood/20170323_biocam_6.png carpet 0.800766 False\n",
      "googlenet images/wood/20170323_optocam_3.png carpet 0.933462 False\n",
      "googlenet images/wood/20170323_optocam_4.png plastic 0.588724 False\n",
      "googlenet images/wood/20170323_optocam_6.png painted 0.355619 False\n",
      "googlenet images/wood/20170323_optocam_8.png stone 0.889436 False\n",
      "googlenet images/wood/20170323_stickcam_10.png tile 0.477885 False\n",
      "googlenet images/wood/20170323_stickcam_2.png wallpaper 0.400407 False\n",
      "googlenet images/wood/20170323_stickcam_6.png plastic 0.363379 False\n",
      "googlenet images/wood/20170323_stickcam_8.png carpet 0.988409 False\n",
      "googlenet images/wood/20170404_optocam_7.png wood 0.995824 True\n",
      "googlenet images/wood/20170404_stickcam_3.png paper 0.591818 False\n",
      "googlenet images/wood/20170405_stickcam_2.png stone 0.392406 False\n",
      "googlenet images/wood/20170406_biocam_10.png wallpaper 0.482663 False\n",
      "googlenet images/wood/20170413_stickcam_14.png carpet 0.513314 False\n",
      "googlenet images/wood/20170418_optocam_19.png carpet 0.737744 False\n",
      "googlenet images/wood/20170418_optocam_20.png carpet 0.692745 False\n",
      "googlenet images/wood/20170418_optocam_6.png carpet 0.696619 False\n",
      "googlenet images/wood/20170419_stickcam_8.png carpet 0.595364 False\n",
      "googlenet images/wood/20170419_stickcam_9.png carpet 0.894821 False\n",
      "googlenet images/wood/20170420_biocam_11.png stone 0.390423 False\n",
      "googlenet images/wood/20170420_biocam_12.png stone 0.630682 False\n",
      "googlenet images/wood/20170420_biocam_16.png fabric 0.481608 False\n",
      "googlenet images/wood/20170420_biocam_4.png carpet 0.969619 False\n",
      "googlenet images/wood/20170420_optocam_4.png carpet 0.919404 False\n",
      "googlenet images/wood/20170421_biocam_1.png wood 0.924349 True\n",
      "googlenet images/wood/20170421_biocam_11.png stone 0.95224 False\n",
      "googlenet images/wood/20170421_optocam_14.png wood 0.950436 True\n",
      "googlenet images/wood/20170421_optocam_15.png stone 0.79739 False\n",
      "googlenet images/wood/20170421_stickcam_1.png wood 0.994704 True\n",
      "googlenet images/wood/20170421_stickcam_7.png wood 0.6255 True\n",
      "googlenet images/wood/20170421_stickcam_9.png stone 0.863095 False\n",
      "googlenet images/wood/20170426_stickcam_11.png fabric 0.391577 False\n",
      "googlenet images/wood/20170427_biocam_3.png fabric 0.701137 False\n",
      "googlenet images/wood/20170427_optocam_5.png fabric 0.694214 False\n",
      "googlenet images/wood/20170501_optocam_11.png carpet 0.540187 False\n",
      "googlenet images/wood/20170501_stickcam_3.png carpet 0.297997 False\n",
      "googlenet images/wood/20170502_biocam_1.png wood 0.442204 True\n",
      "googlenet images/wood/20170503_biocam_10.png wood 0.368568 True\n",
      "googlenet images/wood/20170503_biocam_13.png stone 0.79027 False\n",
      "googlenet images/wood/20170503_optocam_1.png wood 0.971558 True\n",
      "googlenet images/wood/20170503_optocam_4.png carpet 0.629341 False\n",
      "googlenet images/wood/20170503_stickcam_4.png wood 0.940685 True\n",
      "googlenet images/wood/20170503_stickcam_7.png other 0.384015 False\n",
      "googlenet images/wood/20170504_stickcam_2.png fabric 0.992974 False\n",
      "googlenet images/wood/20170516_stickcam_1.png other 0.946076 False\n",
      "googlenet images/wood/20170516_stickcam_7.png carpet 0.938069 False\n",
      "googlenet images/wood/20170518_biocam_2.png wood 0.925068 True\n",
      "googlenet images/wood/20170518_biocam_3.png wood 0.779912 True\n",
      "googlenet images/wood/20170518_optocam_3.png fabric 0.587261 False\n",
      "googlenet images/wood/20170518_optocam_4.png wood 0.632229 True\n",
      "googlenet images/wood/20170518_stickcam_2.png wood 0.894749 True\n",
      "googlenet images/wood/20170518_stickcam_3.png stone 0.894852 False\n",
      "googlenet images/wood/20170519_biocam_10.png stone 0.905455 False\n",
      "googlenet images/wood/20170519_biocam_7.png sky 0.905381 False\n",
      "googlenet images/wood/20170519_biocam_8.png fabric 0.615636 False\n",
      "googlenet images/wood/20170519_optocam_10.png tile 0.226899 False\n",
      "googlenet images/wood/20170519_optocam_7.png fabric 0.649458 False\n",
      "googlenet images/wood/20170519_optocam_8.png carpet 0.479209 False\n",
      "googlenet images/wood/20170519_stickcam_1.png carpet 0.562458 False\n",
      "googlenet images/wood/20170519_stickcam_2.png carpet 0.791916 False\n",
      "googlenet images/wood/20170519_stickcam_4.png metal 0.249714 False\n"
     ]
    }
   ],
   "source": [
    "answers, confusion, categories, net = minc_model.test.classify_images('googlenet')\n",
    "accuracy = np.sum(np.diag(np.array(confusion.values())))/np.sum(np.array(confusion.values()))\n",
    "\n",
    "OrigMinc = namedtuple('OrigMinc', ['acc', 'net', 'answers', 'confusion', 'materials'])\n",
    "orig_minc = OrigMinc(accuracy, net, answers, confusion, categories)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'confusion' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-2-ee8a4c0f3839>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mget_ipython\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmagic\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mu'matplotlib inline'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mmatplotlib\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpyplot\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mimshow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marray\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfusion\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalues\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minterpolation\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'nearest'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcmap\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'Greys'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'confusion' is not defined"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "plt.imshow(np.array(confusion.values(), dtype=np.float), interpolation='nearest', cmap='Greys')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.insert(0, '/home/alexburka/caffe/python')\n",
    "import caffe\n",
    "import numpy as np\n",
    "\n",
    "caffe.set_device(0)\n",
    "caffe.set_mode_gpu()\n",
    "\n",
    "def resize_image(image, new_shape):\n",
    "    old_shape = image.shape[0:2]\n",
    "    if abs(old_shape[0] - new_shape[0]) > abs(old_shape[1] - new_shape[1]):\n",
    "        intermediate_shape = (round(old_shape[0] * new_shape[1] / old_shape[1]), new_shape[1])\n",
    "        intermediate = caffe.io.resize_image(image, intermediate_shape)\n",
    "        start = int(round((intermediate_shape[0] - new_shape[0])/2))\n",
    "        return intermediate[start:(start+new_shape[0]),:,:]\n",
    "    else:\n",
    "        intermediate_shape = (new_shape[0], round(old_shape[1] * new_shape[0] / old_shape[0]))\n",
    "        intermediate = caffe.io.resize_image(image, intermediate_shape)\n",
    "        start = int(round((intermediate_shape[1] - new_shape[1])/2))\n",
    "        return intermediate[:,start:(start+new_shape[1]),:]\n",
    "    \n",
    "def top_k(net, img, k):\n",
    "    outputs = minc_net.predict([img*255], oversample=True, all_outputs=True)\n",
    "    probs = outputs['prob'][0]\n",
    "    top = (-probs).argsort()[:k]\n",
    "    print [(100*probs[i], categories[i]) for i in top]\n",
    "    for p in ['w', 'h', 'r', 'ss', 'so', 'sb']:\n",
    "        out = outputs['prob-%s' % p][0]\n",
    "        print p, out.argmax()+1\n",
    "\n",
    "categories=[x.strip() for x in open('minc_model/categories.txt').readlines()]\n",
    "\n",
    "arch = 'alexnet'\n",
    "minc_net = caffe.Classifier('minc_model/deploy-{}.finetune.prototxt'.format(arch),'minc_model/minc-{}.caffemodel'.format(arch),channel_swap=(2,1,0),mean=np.array([104,117,124]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(72.891616821289062, 'polishedstone'), (13.115638494491577, 'carpet'), (12.743878364562988, 'stone'), (0.35939167719334364, 'tile'), (0.26277503930032253, 'fabric')]\n",
      "w 1\n",
      "h 1\n",
      "r 1\n",
      "ss 1\n",
      "so 1\n",
      "sb 1\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAC/CAYAAADjJVfXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4b39beb490>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = caffe.io.load_image('minc_model/images/polishedstone/20170418_optocam_3.png')\n",
    "resized = resize_image(img, minc_net.image_dims)\n",
    "\n",
    "plt.subplot(121)\n",
    "plt.imshow(img)\n",
    "plt.subplot(122)\n",
    "plt.imshow(resized)\n",
    "top_k(minc_net, resized, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "ename": "ImportError",
     "evalue": "libhdf5.so.7: cannot open shared object file: No such file or directory",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mImportError\u001b[0m                               Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-26-e81388eeaef8>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mimport\u001b[0m \u001b[0mh5py\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/usr/local/lib/python2.7/dist-packages/h5py/__init__.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     11\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     12\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 13\u001b[0;31m     \u001b[0;32mfrom\u001b[0m \u001b[0;34m.\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0m_errors\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     14\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mImportError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     15\u001b[0m     \u001b[0;32mimport\u001b[0m \u001b[0mos\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0m_op\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mImportError\u001b[0m: libhdf5.so.7: cannot open shared object file: No such file or directory"
     ]
    }
   ],
   "source": [
    "import h5py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "## copied from http://nbviewer.jupyter.org/github/BVLC/caffe/blob/master/examples/02-fine-tuning.ipynb\n",
    "\n",
    "from caffe.proto import caffe_pb2\n",
    "\n",
    "def solver(train_net_path, test_net_path=None, base_lr=0.001):\n",
    "    s = caffe_pb2.SolverParameter()\n",
    "\n",
    "    # Specify locations of the train and (maybe) test networks.\n",
    "    s.train_net = train_net_path\n",
    "    if test_net_path is not None:\n",
    "        s.test_net.append(test_net_path)\n",
    "        s.test_interval = 1000  # Test after every 1000 training iterations.\n",
    "        s.test_iter.append(100) # Test on 100 batches each time we test.\n",
    "\n",
    "    # The number of iterations over which to average the gradient.\n",
    "    # Effectively boosts the training batch size by the given factor, without\n",
    "    # affecting memory utilization.\n",
    "    s.iter_size = 1\n",
    "    \n",
    "    s.max_iter = 100000     # # of times to update the net (training iterations)\n",
    "    \n",
    "    # Solve using the stochastic gradient descent (SGD) algorithm.\n",
    "    # Other choices include 'Adam' and 'RMSProp'.\n",
    "    s.type = 'SGD'\n",
    "\n",
    "    # Set the initial learning rate for SGD.\n",
    "    s.base_lr = base_lr\n",
    "\n",
    "    # Set `lr_policy` to define how the learning rate changes during training.\n",
    "    # Here, we 'step' the learning rate by multiplying it by a factor `gamma`\n",
    "    # every `stepsize` iterations.\n",
    "    s.lr_policy = 'step'\n",
    "    s.gamma = 0.1\n",
    "    s.stepsize = 20000\n",
    "\n",
    "    # Set other SGD hyperparameters. Setting a non-zero `momentum` takes a\n",
    "    # weighted average of the current gradient and previous gradients to make\n",
    "    # learning more stable. L2 weight decay regularizes learning, to help prevent\n",
    "    # the model from overfitting.\n",
    "    s.momentum = 0.9\n",
    "    s.weight_decay = 5e-4\n",
    "\n",
    "    # Display the current training loss and accuracy every 1000 iterations.\n",
    "    s.display = 1000\n",
    "\n",
    "    # Snapshots are files used to store networks we've trained.  Here, we'll\n",
    "    # snapshot every 10K iterations -- ten times during training.\n",
    "    #s.snapshot = 10000\n",
    "    #s.snapshot_prefix = caffe_root + 'models/finetune_flickr_style/finetune_flickr_style'\n",
    "    \n",
    "    # Train on the GPU.  Using the CPU to train large networks is very slow.\n",
    "    s.solver_mode = caffe_pb2.SolverParameter.GPU\n",
    "    \n",
    "    # Write the solver to a temporary file and return its filename.\n",
    "    with tempfile.NamedTemporaryFile(delete=False) as f:\n",
    "        f.write(str(s))\n",
    "        return f.name\n",
    "    \n",
    "def run_solvers(niter, solvers, disp_interval=10):\n",
    "    \"\"\"Run solvers for niter iterations,\n",
    "       returning the loss and accuracy recorded each iteration.\n",
    "       `solvers` is a list of (name, solver) tuples.\"\"\"\n",
    "    blobs = ('loss', 'acc')\n",
    "    loss, acc = ({name: np.zeros(niter) for name, _ in solvers}\n",
    "                 for _ in blobs)\n",
    "    for it in range(niter):\n",
    "        for name, s in solvers:\n",
    "            s.step(1)  # run a single SGD step in Caffe\n",
    "            loss[name][it], acc[name][it] = (s.net.blobs[b].data.copy()\n",
    "                                             for b in blobs)\n",
    "        if it % disp_interval == 0 or it + 1 == niter:\n",
    "            loss_disp = '; '.join('%s: loss=%.3f, acc=%2d%%' %\n",
    "                                  (n, loss[n][it], np.round(100*acc[n][it]))\n",
    "                                  for n, _ in solvers)\n",
    "            print '%3d) %s' % (it, loss_disp)     \n",
    "    # Save the learned weights from both nets.\n",
    "    weight_dir = tempfile.mkdtemp()\n",
    "    weights = {}\n",
    "    for name, s in solvers:\n",
    "        filename = 'weights.%s.caffemodel' % name\n",
    "        weights[name] = os.path.join(weight_dir, filename)\n",
    "        s.net.save(weights[name])\n",
    "    return loss, acc, weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<caffe.net_spec.Layers at 0x7fe74b9c1550>"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "caffe.layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<function round>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "round"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "103"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "int(round(102.5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}