image_import.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n",
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/grid_search.py:43: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import cv2\n",
    "import numpy as np\n",
    "import glob\n",
    "import numpy as np\n",
    "from sklearn.cluster import MiniBatchKMeans, KMeans\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.grid_search import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Getting Path of the Emotion Files(with mail emotion) in the Emotion Folder \\nhe Emotion coded files (Emotion_labels.zip) - ONLY 327 of the 593 sequences have \\nemotion sequences. This is because these are the only ones the fit the prototypic \\ndefinition. Like the FACS files, there is only 1 Emotion file for each sequence \\nwhich is the last frame (the peak frame). There should be only one entry and the \\nnumber will range from 0-7 \\n(i.e. 0=neutral, 1=anger, 2=contempt, 3=disgust, 4=fear, 5=happy, 6=sadness, 7=surprise).\\nN.B there is only 327 files- IF THERE IS NO FILE IT MEANS THAT THERE IS NO EMOTION LABEL \\n(sorry to be explicit but this will avoid confusion)'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''Getting Path of the Emotion Files(with mail emotion) in the Emotion Folder \n",
    "he Emotion coded files (Emotion_labels.zip) - ONLY 327 of the 593 sequences have \n",
    "emotion sequences. This is because these are the only ones the fit the prototypic \n",
    "definition. Like the FACS files, there is only 1 Emotion file for each sequence \n",
    "which is the last frame (the peak frame). There should be only one entry and the \n",
    "number will range from 0-7 \n",
    "(i.e. 0=neutral, 1=anger, 2=contempt, 3=disgust, 4=fear, 5=happy, 6=sadness, 7=surprise).\n",
    "N.B there is only 327 files- IF THERE IS NO FILE IT MEANS THAT THERE IS NO EMOTION LABEL \n",
    "(sorry to be explicit but this will avoid confusion)'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['/Users/avaneeshkumar/Documents/CK+/Emotion/S005/001/S005_001_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S010/002/S010_002_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S010/004/S010_004_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S010/006/S010_006_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S011/001/S011_001_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S011/002/S011_002_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S011/003/S011_003_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S011/004/S011_004_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S011/005/S011_005_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S011/006/S011_006_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S014/001/S014_001_00000029_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S014/002/S014_002_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S014/003/S014_003_00000030_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S014/005/S014_005_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S022/001/S022_001_00000030_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S022/005/S022_005_00000032_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S022/006/S022_006_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S026/001/S026_001_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S026/002/S026_002_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S026/003/S026_003_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S026/006/S026_006_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S028/001/S028_001_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S029/001/S029_001_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S032/001/S032_001_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S032/003/S032_003_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S032/004/S032_004_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S032/005/S032_005_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S032/006/S032_006_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S034/001/S034_001_00000029_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S034/003/S034_003_00000027_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S034/005/S034_005_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S035/001/S035_001_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S035/005/S035_005_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S035/006/S035_006_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S037/001/S037_001_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S037/003/S037_003_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S037/006/S037_006_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S042/001/S042_001_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S042/002/S042_002_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S042/004/S042_004_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S042/006/S042_006_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S044/001/S044_001_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S044/003/S044_003_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S044/006/S044_006_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S045/004/S045_004_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S045/005/S045_005_00000030_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S046/001/S046_001_00000025_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S046/002/S046_002_00000006_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S046/003/S046_003_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S046/004/S046_004_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S050/001/S050_001_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S050/002/S050_002_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S050/004/S050_004_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S050/006/S050_006_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S051/002/S051_002_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S051/003/S051_003_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S052/001/S052_001_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S052/004/S052_004_00000033_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S052/006/S052_006_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S053/001/S053_001_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S053/004/S053_004_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S054/002/S054_002_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S054/003/S054_003_00000007_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S054/004/S054_004_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S055/001/S055_001_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S055/003/S055_003_00000009_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S055/004/S055_004_00000028_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S055/005/S055_005_00000045_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S055/006/S055_006_00000008_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S056/002/S056_002_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S056/003/S056_003_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S056/004/S056_004_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S057/001/S057_001_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S057/003/S057_003_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S057/006/S057_006_00000033_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S058/001/S058_001_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S058/005/S058_005_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S058/006/S058_006_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S059/001/S059_001_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S059/002/S059_002_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S060/002/S060_002_00000026_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S060/003/S060_003_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S060/005/S060_005_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S061/001/S061_001_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S061/002/S061_002_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S061/004/S061_004_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S062/001/S062_001_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S062/002/S062_002_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S062/004/S062_004_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S062/005/S062_005_00000029_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S063/001/S063_001_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S063/002/S063_002_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S064/001/S064_001_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S064/003/S064_003_00000025_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S064/004/S064_004_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S065/002/S065_002_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S065/003/S065_003_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S065/004/S065_004_00000028_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S065/005/S065_005_00000008_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S066/002/S066_002_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S066/003/S066_003_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S066/004/S066_004_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S066/005/S066_005_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S067/002/S067_002_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S067/004/S067_004_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S067/005/S067_005_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S067/006/S067_006_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S068/002/S068_002_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S068/003/S068_003_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S068/004/S068_004_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S068/005/S068_005_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S069/002/S069_002_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S069/003/S069_003_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S069/004/S069_004_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S070/002/S070_002_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S070/003/S070_003_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S070/005/S070_005_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S071/001/S071_001_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S071/002/S071_002_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S071/004/S071_004_00000028_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S071/005/S071_005_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S071/006/S071_006_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S072/005/S072_005_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S072/006/S072_006_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S073/001/S073_001_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S073/006/S073_006_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S074/001/S074_001_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S074/002/S074_002_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S074/004/S074_004_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S074/005/S074_005_00000043_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S075/002/S075_002_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S075/005/S075_005_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S075/006/S075_006_00000025_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S075/008/S075_008_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S076/001/S076_001_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S076/005/S076_005_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S076/006/S076_006_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S077/001/S077_001_00000028_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S077/006/S077_006_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S078/001/S078_001_00000033_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S078/004/S078_004_00000027_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S078/007/S078_007_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S079/001/S079_001_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S079/002/S079_002_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S079/004/S079_004_00000026_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S080/001/S080_001_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S080/005/S080_005_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S080/008/S080_008_00000009_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S081/001/S081_001_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S081/002/S081_002_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S081/008/S081_008_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S082/001/S082_001_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S082/005/S082_005_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S082/007/S082_007_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S083/003/S083_003_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S084/001/S084_001_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S084/002/S084_002_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S085/002/S085_002_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S085/003/S085_003_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S085/004/S085_004_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S086/001/S086_001_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S086/002/S086_002_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S087/001/S087_001_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S087/004/S087_004_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S087/005/S087_005_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S087/007/S087_007_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S088/001/S088_001_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S088/004/S088_004_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S089/001/S089_001_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S089/002/S089_002_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S089/003/S089_003_00000036_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S090/002/S090_002_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S090/006/S090_006_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S090/007/S090_007_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S091/001/S091_001_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S091/003/S091_003_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S092/001/S092_001_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S092/003/S092_003_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S092/004/S092_004_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S093/001/S093_001_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S093/004/S093_004_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S094/001/S094_001_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S094/004/S094_004_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S095/001/S095_001_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S095/006/S095_006_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S095/007/S095_007_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S095/010/S095_010_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S096/001/S096_001_00000007_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S096/003/S096_003_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S096/004/S096_004_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S097/001/S097_001_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S097/004/S097_004_00000030_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S097/006/S097_006_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S098/003/S098_003_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S098/004/S098_004_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S099/001/S099_001_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S099/004/S099_004_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S099/007/S099_007_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S100/002/S100_002_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S100/005/S100_005_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S100/006/S100_006_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S101/002/S101_002_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S102/002/S102_002_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S102/003/S102_003_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S102/009/S102_009_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S105/008/S105_008_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S106/002/S106_002_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S106/004/S106_004_00000008_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S106/006/S106_006_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S107/001/S107_001_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S107/005/S107_005_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S108/005/S108_005_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S108/006/S108_006_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S108/008/S108_008_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S109/003/S109_003_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S109/005/S109_005_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S109/006/S109_006_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S110/001/S110_001_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S111/001/S111_001_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S111/006/S111_006_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S111/007/S111_007_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S112/005/S112_005_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S113/001/S113_001_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S113/003/S113_003_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S113/008/S113_008_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S114/001/S114_001_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S114/006/S114_006_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S115/001/S115_001_00000008_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S115/004/S115_004_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S115/008/S115_008_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S116/001/S116_001_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S116/006/S116_006_00000007_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S116/007/S116_007_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S117/001/S117_001_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S117/003/S117_003_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S117/006/S117_006_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S119/001/S119_001_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S119/003/S119_003_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S119/008/S119_008_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S122/001/S122_001_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S124/001/S124_001_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S124/003/S124_003_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S124/006/S124_006_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S124/007/S124_007_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S125/001/S125_001_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S125/005/S125_005_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S125/006/S125_006_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S125/007/S125_007_00000009_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S125/008/S125_008_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S126/004/S126_004_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S126/008/S126_008_00000029_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S127/001/S127_001_00000017_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S127/004/S127_004_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S127/010/S127_010_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S128/004/S128_004_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S128/011/S128_011_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S129/002/S129_002_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S129/006/S129_006_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S129/011/S129_011_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S129/012/S129_012_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S130/001/S130_001_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S130/007/S130_007_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S130/009/S130_009_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S130/012/S130_012_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S130/013/S130_013_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S131/001/S131_001_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S131/003/S131_003_00000024_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S131/006/S131_006_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S131/010/S131_010_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S132/002/S132_002_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S132/003/S132_003_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S132/005/S132_005_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S132/006/S132_006_00000023_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S132/008/S132_008_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S133/003/S133_003_00000047_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S133/009/S133_009_00000006_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S133/010/S133_010_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S134/003/S134_003_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S134/004/S134_004_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S134/008/S134_008_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S135/001/S135_001_00000039_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S135/012/S135_012_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S136/001/S136_001_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S136/003/S136_003_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S136/005/S136_005_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S136/006/S136_006_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S137/001/S137_001_00000014_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S137/005/S137_005_00000027_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S137/011/S137_011_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S138/001/S138_001_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S138/004/S138_004_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S138/005/S138_005_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S138/007/S138_007_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S138/008/S138_008_00000009_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S139/002/S139_002_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S147/002/S147_002_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S148/002/S148_002_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S149/002/S149_002_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S151/002/S151_002_00000029_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S154/002/S154_002_00000013_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S155/002/S155_002_00000012_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S156/002/S156_002_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S157/002/S157_002_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S158/002/S158_002_00000011_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S160/006/S160_006_00000010_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S501/001/S501_001_00000067_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S501/004/S501_004_00000056_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S501/006/S501_006_00000041_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S502/001/S502_001_00000016_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S502/002/S502_002_00000009_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S502/004/S502_004_00000052_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S503/001/S503_001_00000071_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S503/002/S503_002_00000008_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S503/006/S503_006_00000020_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S504/001/S504_001_00000022_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S504/002/S504_002_00000009_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S504/004/S504_004_00000015_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S504/006/S504_006_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S505/002/S505_002_00000021_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S505/006/S505_006_00000019_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S506/001/S506_001_00000040_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S506/002/S506_002_00000009_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S506/004/S506_004_00000038_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S506/006/S506_006_00000042_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S895/002/S895_002_00000007_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S999/001/S999_001_00000018_emotion.txt',\n",
       " '/Users/avaneeshkumar/Documents/CK+/Emotion/S999/003/S999_003_00000055_emotion.txt']"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import os\n",
    "sub = '.DS_Store'\n",
    "path = \"/Users/avaneeshkumar/Documents/CK+/Emotion\"\n",
    "\n",
    "#Finding Files from the directories\n",
    "def findfiles(path):\n",
    "    File = []\n",
    "    for root, dirs, files in os.walk(path):\n",
    "        #Exclude Directories and dodge \".\" hidden files. \n",
    "        dirs[:] = [d for d in dirs if not d.startswith('.')]\n",
    "        #Appending root \n",
    "        for file in files:\n",
    "            File.append(os.path.join(root, file)) \n",
    "    return File       \n",
    "\n",
    "files = findfiles(path)\n",
    "files\n",
    "\n",
    "# without .DS_Store, Files List\n",
    "def noDS(files):\n",
    "    Files_sequence = []\n",
    "    #not selecting \".DS_Store\"\n",
    "    sub = '.DS_Store'\n",
    "    for text in files:\n",
    "        if sub not in text:\n",
    "            Files_sequence.append(text)\n",
    "    return Files_sequence     \n",
    "\n",
    "FS = noDS(files)\n",
    "FS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Computing labelled data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S005/001/S005_001_00000011.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S010/002/S010_002_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S010/004/S010_004_00000019.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S010/006/S010_006_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S011/001/S011_001_00000016.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S011/002/S011_002_00000022.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S011/003/S011_003_00000014.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S011/004/S011_004_00000021.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S011/005/S011_005_00000020.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S011/006/S011_006_00000013.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S014/001/S014_001_00000029.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S014/002/S014_002_00000016.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S014/003/S014_003_00000030.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S014/005/S014_005_00000017.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S022/001/S022_001_00000030.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S022/005/S022_005_00000032.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S022/006/S022_006_00000017.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S026/001/S026_001_00000015.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S026/002/S026_002_00000016.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S026/003/S026_003_00000015.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S026/006/S026_006_00000013.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S028/001/S028_001_00000024.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S029/001/S029_001_00000019.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S032/001/S032_001_00000022.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S032/003/S032_003_00000017.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S032/004/S032_004_00000014.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S032/005/S032_005_00000016.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S032/006/S032_006_00000016.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S034/001/S034_001_00000029.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S034/003/S034_003_00000027.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S034/005/S034_005_00000010.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S035/001/S035_001_00000015.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S035/005/S035_005_00000019.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S035/006/S035_006_00000018.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S037/001/S037_001_00000020.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S037/003/S037_003_00000022.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S037/006/S037_006_00000021.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S042/001/S042_001_00000019.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S042/002/S042_002_00000016.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S042/004/S042_004_00000020.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S042/006/S042_006_00000017.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S044/001/S044_001_00000024.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S044/003/S044_003_00000014.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S044/006/S044_006_00000019.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S045/004/S045_004_00000015.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S045/005/S045_005_00000030.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S046/001/S046_001_00000025.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S046/002/S046_002_00000006.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S046/003/S046_003_00000016.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S046/004/S046_004_00000017.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S050/001/S050_001_00000017.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S050/002/S050_002_00000018.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S050/004/S050_004_00000021.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S050/006/S050_006_00000023.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S051/002/S051_002_00000019.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S051/003/S051_003_00000018.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S052/001/S052_001_00000015.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S052/004/S052_004_00000033.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S052/006/S052_006_00000013.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S053/001/S053_001_00000023.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S053/004/S053_004_00000024.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S054/002/S054_002_00000015.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S054/003/S054_003_00000007.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S054/004/S054_004_00000024.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S055/001/S055_001_00000012.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S055/003/S055_003_00000009.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S055/004/S055_004_00000028.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S055/005/S055_005_00000045.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S055/006/S055_006_00000008.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S056/002/S056_002_00000010.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S056/003/S056_003_00000010.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S056/004/S056_004_00000020.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S057/001/S057_001_00000019.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S057/003/S057_003_00000015.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S057/006/S057_006_00000033.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S058/001/S058_001_00000020.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S058/005/S058_005_00000010.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S058/006/S058_006_00000018.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S059/001/S059_001_00000018.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S059/002/S059_002_00000017.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S060/002/S060_002_00000026.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S060/003/S060_003_00000018.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S060/005/S060_005_00000021.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S061/001/S061_001_00000012.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S061/002/S061_002_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S061/004/S061_004_00000022.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S062/001/S062_001_00000017.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S062/002/S062_002_00000016.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S062/004/S062_004_00000024.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S062/005/S062_005_00000029.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S063/001/S063_001_00000013.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S063/002/S063_002_00000023.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S064/001/S064_001_00000012.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S064/003/S064_003_00000025.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S064/004/S064_004_00000014.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S065/002/S065_002_00000022.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S065/003/S065_003_00000022.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S065/004/S065_004_00000028.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S065/005/S065_005_00000008.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S066/002/S066_002_00000022.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S066/003/S066_003_00000012.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S066/004/S066_004_00000010.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S066/005/S066_005_00000011.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S067/002/S067_002_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S067/004/S067_004_00000023.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S067/005/S067_005_00000022.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S067/006/S067_006_00000011.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S068/002/S068_002_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S068/003/S068_003_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S068/004/S068_004_00000010.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S068/005/S068_005_00000021.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S069/002/S069_002_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S069/003/S069_003_00000011.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S069/004/S069_004_00000017.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S070/002/S070_002_00000016.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S070/003/S070_003_00000017.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S070/005/S070_005_00000016.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S071/001/S071_001_00000013.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S071/002/S071_002_00000020.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S071/004/S071_004_00000028.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S071/005/S071_005_00000021.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S071/006/S071_006_00000014.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S072/005/S072_005_00000019.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S072/006/S072_006_00000022.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S073/001/S073_001_00000013.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S073/006/S073_006_00000014.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S074/001/S074_001_00000020.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S074/002/S074_002_00000016.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S074/004/S074_004_00000018.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S074/005/S074_005_00000043.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S075/002/S075_002_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S075/005/S075_005_00000012.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S075/006/S075_006_00000025.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S075/008/S075_008_00000012.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S076/001/S076_001_00000017.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S076/005/S076_005_00000012.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S076/006/S076_006_00000019.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S077/001/S077_001_00000028.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S077/006/S077_006_00000014.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S078/001/S078_001_00000033.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S078/004/S078_004_00000027.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S078/007/S078_007_00000013.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S079/001/S079_001_00000012.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S079/002/S079_002_00000012.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S079/004/S079_004_00000026.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S080/001/S080_001_00000018.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S080/005/S080_005_00000013.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S080/008/S080_008_00000009.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S081/001/S081_001_00000019.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S081/002/S081_002_00000024.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S081/008/S081_008_00000011.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S082/001/S082_001_00000015.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S082/005/S082_005_00000017.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S082/007/S082_007_00000010.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S083/003/S083_003_00000019.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S084/001/S084_001_00000010.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S084/002/S084_002_00000023.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S085/002/S085_002_00000014.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S085/003/S085_003_00000013.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S085/004/S085_004_00000017.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S086/001/S086_001_00000019.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S086/002/S086_002_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S087/001/S087_001_00000011.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S087/004/S087_004_00000012.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S087/005/S087_005_00000012.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S087/007/S087_007_00000016.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S088/001/S088_001_00000017.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S088/004/S088_004_00000020.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S089/001/S089_001_00000016.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S089/002/S089_002_00000021.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S089/003/S089_003_00000036.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S090/002/S090_002_00000011.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S090/006/S090_006_00000011.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S090/007/S090_007_00000014.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S091/001/S091_001_00000015.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S091/003/S091_003_00000021.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S092/001/S092_001_00000017.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S092/003/S092_003_00000014.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S092/004/S092_004_00000024.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S093/001/S093_001_00000020.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S093/004/S093_004_00000016.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S094/001/S094_001_00000010.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S094/004/S094_004_00000012.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S095/001/S095_001_00000016.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S095/006/S095_006_00000013.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S095/007/S095_007_00000021.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S095/010/S095_010_00000014.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S096/001/S096_001_00000007.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S096/003/S096_003_00000012.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S096/004/S096_004_00000011.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S097/001/S097_001_00000021.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S097/004/S097_004_00000030.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S097/006/S097_006_00000019.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S098/003/S098_003_00000013.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S098/004/S098_004_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S099/001/S099_001_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S099/004/S099_004_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S099/007/S099_007_00000012.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S100/002/S100_002_00000015.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S100/005/S100_005_00000023.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S100/006/S100_006_00000016.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S101/002/S101_002_00000019.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S102/002/S102_002_00000018.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S102/003/S102_003_00000016.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S102/009/S102_009_00000015.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S105/008/S105_008_00000010.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S106/002/S106_002_00000016.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S106/004/S106_004_00000008.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S106/006/S106_006_00000011.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S107/001/S107_001_00000010.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S107/005/S107_005_00000011.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S108/005/S108_005_00000022.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S108/006/S108_006_00000020.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S108/008/S108_008_00000013.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S109/003/S109_003_00000017.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S109/005/S109_005_00000014.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S109/006/S109_006_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S110/001/S110_001_00000013.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S111/001/S111_001_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S111/006/S111_006_00000010.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S111/007/S111_007_00000014.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S112/005/S112_005_00000017.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S113/001/S113_001_00000012.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S113/003/S113_003_00000015.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S113/008/S113_008_00000023.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S114/001/S114_001_00000018.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S114/006/S114_006_00000023.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S115/001/S115_001_00000008.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S115/004/S115_004_00000017.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S115/008/S115_008_00000017.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S116/001/S116_001_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S116/006/S116_006_00000007.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S116/007/S116_007_00000017.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S117/001/S117_001_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S117/003/S117_003_00000014.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S117/006/S117_006_00000010.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S119/001/S119_001_00000011.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S119/003/S119_003_00000024.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S119/008/S119_008_00000018.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S122/001/S122_001_00000012.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S124/001/S124_001_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S124/003/S124_003_00000011.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S124/006/S124_006_00000011.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S124/007/S124_007_00000024.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S125/001/S125_001_00000014.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S125/005/S125_005_00000013.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S125/006/S125_006_00000022.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S125/007/S125_007_00000009.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S125/008/S125_008_00000010.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S126/004/S126_004_00000012.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S126/008/S126_008_00000029.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S127/001/S127_001_00000017.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S127/004/S127_004_00000016.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S127/010/S127_010_00000018.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S128/004/S128_004_00000013.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S128/011/S128_011_00000016.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S129/002/S129_002_00000011.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S129/006/S129_006_00000010.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S129/011/S129_011_00000018.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S129/012/S129_012_00000011.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S130/001/S130_001_00000018.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S130/007/S130_007_00000020.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S130/009/S130_009_00000019.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S130/012/S130_012_00000011.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S130/013/S130_013_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S131/001/S131_001_00000016.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S131/003/S131_003_00000024.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S131/006/S131_006_00000022.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S131/010/S131_010_00000018.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S132/002/S132_002_00000018.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S132/003/S132_003_00000023.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S132/005/S132_005_00000016.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S132/006/S132_006_00000023.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S132/008/S132_008_00000010.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S133/003/S133_003_00000047.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S133/009/S133_009_00000006.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S133/010/S133_010_00000014.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S134/003/S134_003_00000011.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S134/004/S134_004_00000015.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S134/008/S134_008_00000013.png', 3), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S135/001/S135_001_00000039.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S135/012/S135_012_00000020.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S136/001/S136_001_00000019.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S136/003/S136_003_00000014.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S136/005/S136_005_00000010.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S136/006/S136_006_00000020.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S137/001/S137_001_00000014.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S137/005/S137_005_00000027.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S137/011/S137_011_00000020.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S138/001/S138_001_00000012.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S138/004/S138_004_00000013.png', 7), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S138/005/S138_005_00000016.png', 5), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S138/007/S138_007_00000011.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S138/008/S138_008_00000009.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S139/002/S139_002_00000013.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S147/002/S147_002_00000013.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S148/002/S148_002_00000015.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S149/002/S149_002_00000013.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S151/002/S151_002_00000029.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S154/002/S154_002_00000013.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S155/002/S155_002_00000012.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S156/002/S156_002_00000021.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S157/002/S157_002_00000011.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S158/002/S158_002_00000011.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S160/006/S160_006_00000010.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S501/001/S501_001_00000067.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S501/004/S501_004_00000056.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S501/006/S501_006_00000041.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S502/001/S502_001_00000016.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S502/002/S502_002_00000009.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S502/004/S502_004_00000052.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S503/001/S503_001_00000071.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S503/002/S503_002_00000008.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S503/006/S503_006_00000020.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S504/001/S504_001_00000022.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S504/002/S504_002_00000009.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S504/004/S504_004_00000015.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S504/006/S504_006_00000018.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S505/002/S505_002_00000021.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S505/006/S505_006_00000019.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S506/001/S506_001_00000040.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S506/002/S506_002_00000009.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S506/004/S506_004_00000038.png', 4), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S506/006/S506_006_00000042.png', 6), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S895/002/S895_002_00000007.png', 2), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S999/001/S999_001_00000018.png', 1), ('/Users/avaneeshkumar/Documents/CK+/cohn-kanade-images/S999/003/S999_003_00000055.png', 4)]\n"
     ]
    }
   ],
   "source": [
    "#make emotion list by reading test file\n",
    "def emo_seq(txt_file):    \n",
    "    k = open(txt_file, 'r')\n",
    "    emo_tag=k.readlines()\n",
    "    emo=int(float(emo_tag[0]))\n",
    "    return emo\n",
    "\n",
    "def find_label(img_path,labeled_data):\n",
    "    for rows in labeled_data:\n",
    "        if rows[0]==img_path:\n",
    "            return rows[1]\n",
    "\n",
    "def getimagePath(txt_path):\n",
    "    marker = \"Emotion\"\n",
    "    replacement = \"cohn-kanade-images\"\n",
    "    mn = i.find(marker)\n",
    "    replaced = i[:mn] + replacement + i[mn + len(marker):-12] + \".png\"\n",
    "    return replaced\n",
    "\n",
    "def labeled_data(i):\n",
    "        emo=emo_seq(i)\n",
    "        img_path=getimagePath(i)\n",
    "        y = (img_path,emo)\n",
    "        return y\n",
    "\n",
    "LD = []\n",
    "for i in FS:\n",
    "    x = labeled_data(i)\n",
    "    LD.append(x)\n",
    "    \n",
    "print(LD)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def train_test_val_split_idxs(total_rows, percent_test):\n",
    "    \"\"\"\n",
    "    Get indexes for training, test, and validation rows, given a total number of rows.\n",
    "    Assumes indexes are sequential integers starting at 0: eg [0,1,2,3,...N]\n",
    "\n",
    "    Returns:\n",
    "    --------\n",
    "    training_idxs, test_idxs, val_idxs\n",
    "        Both lists of integers\n",
    "    \"\"\"\n",
    "    if percent_test + percent_test > 1.0:\n",
    "        raise ValueError('percent_test and percent_val must sum to less than 1.0')\n",
    "\n",
    "    row_range = range(total_rows)\n",
    "\n",
    "    no_test_rows = int(total_rows*(percent_test))\n",
    "    test_idxs = np.random.choice(row_range, size=no_test_rows, replace=False)\n",
    "    # remove test indexes\n",
    "    row_range = [idx for idx in row_range if idx not in test_idxs]\n",
    "\n",
    "   # no_val_rows = int(total_rows*(percent_val))\n",
    "   # val_idxs = np.random.choice(row_range, size=no_val_rows, replace=False)\n",
    "    # remove validation indexes\n",
    "    training_idxs = row_range\n",
    "\n",
    "    #print 'Train-test-val split: %i training rows, %i test rows, %i validation rows' % (len(training_idxs), len(test_idxs), len(val_idxs))\n",
    "\n",
    "    return training_idxs, test_idxs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def image_pre_harr(img_path):\n",
    "    face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')\n",
    "    img = cv2.imread(img_path)\n",
    "    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\n",
    "    faces = face_cascade.detectMultiScale(gray, 1.3, 5)\n",
    "    return faces"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def img_desc_sift(img_path):\n",
    "    face_cord=image_pre_harr(img_path)\n",
    "    img = cv2.imread(img_path)\n",
    "    if len(face_cord) == 0 :\n",
    "        return []\n",
    "    (x,y,w,h) = face_cord[0]\n",
    "    img=img[y:y+h, x:x+w]\n",
    "    gray= cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)\n",
    "    sift = cv2.xfeatures2d.SIFT_create()\n",
    "    kp, des = sift.detectAndCompute(gray,None)\n",
    "    #print 'image:',img_path,'# descriptors:',des.shape\n",
    "    return des"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def decs_train(train,labeled_data):\n",
    "    des_train=[]\n",
    "    Y=[]\n",
    "    print 'computing descriptors:', len(train)\n",
    "    for i,img_path in enumerate(train):\n",
    "        print 'image number',i\n",
    "        des=img_desc_sift(img_path)\n",
    "        if len(des)==0:\n",
    "            continue\n",
    "        des_train.append(des)\n",
    "        label=find_label(img_path,labeled_data)\n",
    "        #print label\n",
    "        Y.append(label)\n",
    "    #des_stack_train=[desc for img in des_train for desc in img]\n",
    "    return des_train,Y\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def find_label(img_path,labeled_data):\n",
    "    for rows in labeled_data:\n",
    "        if rows[0]==img_path:\n",
    "            return rows[1]\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def cluster_features(img_descs,cluster_model):\n",
    "    \"\"\"\n",
    "    Cluster the training features using the cluster_model\n",
    "    and convert each set of descriptors in img_descs\n",
    "    to a Visual Bag of Words histogram.\n",
    "\n",
    "    Parameters:\n",
    "    -----------\n",
    "    X : list of lists of SIFT descriptors (img_descs)\n",
    "\n",
    "    training_idxs : array/list of integers\n",
    "        Indicies for the training rows in img_descs\n",
    "\n",
    "    cluster_model : clustering model (eg KMeans from scikit-learn)\n",
    "        The model used to cluster the SIFT features\n",
    "\n",
    "    Returns:\n",
    "    --------\n",
    "    cluster_mode :\n",
    "    cluster_model has been fit to the training set\n",
    "    \"\"\"\n",
    "    n_clusters = cluster_model.n_clusters\n",
    "    des_stack_train=[desc for img in img_descs for desc in img]\n",
    "    all_train_descriptors = np.array(des_stack_train)\n",
    "\n",
    "    if all_train_descriptors.shape[1] != 128:\n",
    "        raise ValueError('Expected SIFT descriptors to have 128 features, got', all_train_descriptors.shape[1])\n",
    "\n",
    "    print '%i descriptors before clustering' % all_train_descriptors.shape[0]\n",
    "\n",
    "    # Cluster descriptors to get codebook\n",
    "    print 'Using clustering model %s...' % repr(cluster_model)\n",
    "    print 'Clustering on training set to get codebook of %i words' % n_clusters\n",
    "\n",
    "    # train kmeans or other cluster model on those descriptors selected above\n",
    "    cluster_model.fit(all_train_descriptors)\n",
    "    print 'done clustering.'\n",
    "\n",
    "    return cluster_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def histogram(cluster_model,des,K):\n",
    "    words=cluster_model.predict(des)\n",
    "    img_bow_hist = np.array([np.bincount(words, minlength=K)])\n",
    "    return img_bow_hist"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#Splitting datainto test, train, and validation , clustreing using KMeans and MINIBatchKMeans\n",
    "#computing descriptors of training data and producing corresponding labels\n",
    "#"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def cluster_split(labeled_data, K, percent_test):\n",
    "    emo1=[]\n",
    "    emo2=[]\n",
    "    emo3=[]\n",
    "    emo4=[]\n",
    "    emo5=[]\n",
    "    emo6=[]\n",
    "    emo7=[]\n",
    "    emotions=[1,2,3,4,5,6,7]\n",
    "    \n",
    "    print '\\n total size of data:',len(labeled_data)\n",
    "    # segregating database images based on emotion\n",
    "\n",
    "    for rows in labeled_data:\n",
    "        emo= str(rows[1])\n",
    "        a='emo'+emo\n",
    "        locals()[a].append(rows[0])\n",
    "    #splitting data into test, train and validation emotion category-wise\n",
    "    train=[]\n",
    "    test=[]\n",
    "    #validation=[]\n",
    "    for emo in emotions:\n",
    "        total_rows=len(locals()['emo'+str(emo)])\n",
    "        training_idxs, test_idxs = train_test_val_split_idxs(total_rows,percent_test)\n",
    "        train_emo=[locals()['emo'+str(emo)][i] for i in training_idxs]\n",
    "        test_emo=[locals()['emo'+str(emo)][i] for i in test_idxs]\n",
    "        #val_emo=[locals()['emo'+str(emo)][i] for i in val_idxs]\n",
    "        for rows in train_emo:\n",
    "            train.append(rows)\n",
    "        for rows in test_emo:\n",
    "            test.append(rows)\n",
    "       # for rows in val_emo:\n",
    "          #  validation.append(rows)\n",
    "    print '\\n size of training data:',len(train),'\\n size of test data:',len(test)\n",
    "    # getting descriptors of training set along with labels\n",
    "    print '\\n computing descriptors of training data and producing corresponding labels'\n",
    "    des_train,Y_train = decs_train(train,labeled_data)\n",
    "    # building cluster model\n",
    "    cluster_model = cluster_features(des_train,cluster_model=MiniBatchKMeans(n_clusters=K))\n",
    "    \n",
    "    X_train=[]   # initialising array of input vectors for training set\n",
    "    \n",
    "    print '\\n calculting histogram vectors of training set'\n",
    "    for des in des_train:\n",
    "        hist=histogram(cluster_model,des,K)\n",
    "        X_train.append(hist)\n",
    "    X_train=np.array(X_train)\n",
    "    \n",
    "    # building validation set in terms of labels and vectors\n",
    "    #X_val=[]\n",
    "   # print '\\n computing descriptors of validation data and producing corresponding labels'\n",
    "    #des_val,Y_val=decs_train(validation,labeled_data)\n",
    "    \n",
    "   # print '\\n calculting histogram vectors of validation set'\n",
    "   # for des in des_val:\n",
    "     #   hist=histogram(cluster_model,des,K)\n",
    "     #   X_val.append(hist)\n",
    "    #X_val=np.array(X_val)\n",
    "    \n",
    "    # building test set in terms of labels and vectors\n",
    "    X_test=[]\n",
    "    print '\\n computing descriptors of test data and producing corresponding labels'\n",
    "    des_test,Y_test=decs_train(test,labeled_data)\n",
    "    \n",
    "    print '\\n calculting histogram vectors of test set'\n",
    "    for des in des_test:\n",
    "        hist=histogram(cluster_model,des,K)\n",
    "        X_test.append(hist)\n",
    "    X_test=np.array(X_test)\n",
    "    return X_train, Y_train, X_test, Y_test, cluster_model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "ld = LD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#X_train, Y_train, X_val, Y_val, X_test, Y_test, cluster_mode = cluster_split(ld, 200, 0.1,0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def run_svm(X_train, X_test, Y_train, Y_test, scoring,\n",
    "    c_vals=[1, 5, 10], gamma_vals=[0.1, 0.01, 0.0001, 0.00001]):\n",
    "\n",
    "    param_grid = [\n",
    "    #   {'C': c_vals, 'kernel': ['linear']},\n",
    "      {'C': c_vals, 'gamma': gamma_vals, 'kernel': ['rbf']},\n",
    "     ]\n",
    "\n",
    "    svc = GridSearchCV(SVC(), param_grid, n_jobs=-1, cv=10, scoring=scoring)\n",
    "    svc.fit(X_train, Y_train)\n",
    "    print 'train score (%s):'%scoring, svc.score(X_train, Y_train)\n",
    "    test_score = svc.score(X_test, Y_test)\n",
    "    print 'test score (%s):'%scoring, test_score\n",
    "\n",
    "    print svc.best_estimator_\n",
    "\n",
    "    return svc, test_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "59787 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=100, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 100 words\n",
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.624161073826\n",
      "test score (recall_micro): 0.607142857143\n",
      "SVC(C=5, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=100 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/grid_search.py:438: ChangedBehaviorWarning: The long-standing behavior to use the estimator's score function in GridSearchCV.score has changed. The scoring parameter is now used.\n",
      "  ChangedBehaviorWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "59786 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=150, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 150 words\n",
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.761744966443\n",
      "test score (recall_micro): 0.5\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=150 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "60636 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=200, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 200 words\n",
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.748322147651\n",
      "test score (recall_micro): 0.607142857143\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=200 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "60172 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=350, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 350 words\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/cluster/k_means_.py:1381: RuntimeWarning: init_size=300 should be larger than k=350. Setting it to 3*k\n",
      "  init_size=init_size)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.781879194631\n",
      "test score (recall_micro): 0.607142857143\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=350 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "59436 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=400, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 400 words\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/cluster/k_means_.py:1381: RuntimeWarning: init_size=300 should be larger than k=400. Setting it to 3*k\n",
      "  init_size=init_size)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.751677852349\n",
      "test score (recall_micro): 0.678571428571\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=400 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "60083 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=450, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 450 words\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/cluster/k_means_.py:1381: RuntimeWarning: init_size=300 should be larger than k=450. Setting it to 3*k\n",
      "  init_size=init_size)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.771812080537\n",
      "test score (recall_micro): 0.678571428571\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=450 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "59453 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=500, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 500 words\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/cluster/k_means_.py:1381: RuntimeWarning: init_size=300 should be larger than k=500. Setting it to 3*k\n",
      "  init_size=init_size)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.744966442953\n",
      "test score (recall_micro): 0.607142857143\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=500 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "60051 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=530, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 530 words\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/cluster/k_means_.py:1381: RuntimeWarning: init_size=300 should be larger than k=530. Setting it to 3*k\n",
      "  init_size=init_size)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.748322147651\n",
      "test score (recall_micro): 0.714285714286\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=530 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "60292 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=550, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 550 words\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/cluster/k_means_.py:1381: RuntimeWarning: init_size=300 should be larger than k=550. Setting it to 3*k\n",
      "  init_size=init_size)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.768456375839\n",
      "test score (recall_micro): 0.642857142857\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=550 DONE ***\n",
      "\n",
      "\n",
      " total size of data: 327\n",
      "\n",
      " size of training data: 299 \n",
      " size of test data: 28\n",
      "\n",
      " computing descriptors of training data and producing corresponding labels\n",
      "computing descriptors: 299\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "image number 28\n",
      "image number 29\n",
      "image number 30\n",
      "image number 31\n",
      "image number 32\n",
      "image number 33\n",
      "image number 34\n",
      "image number 35\n",
      "image number 36\n",
      "image number 37\n",
      "image number 38\n",
      "image number 39\n",
      "image number 40\n",
      "image number 41\n",
      "image number 42\n",
      "image number 43\n",
      "image number 44\n",
      "image number 45\n",
      "image number 46\n",
      "image number 47\n",
      "image number 48\n",
      "image number 49\n",
      "image number 50\n",
      "image number 51\n",
      "image number 52\n",
      "image number 53\n",
      "image number 54\n",
      "image number 55\n",
      "image number 56\n",
      "image number 57\n",
      "image number 58\n",
      "image number 59\n",
      "image number 60\n",
      "image number 61\n",
      "image number 62\n",
      "image number 63\n",
      "image number 64\n",
      "image number 65\n",
      "image number 66\n",
      "image number 67\n",
      "image number 68\n",
      "image number 69\n",
      "image number 70\n",
      "image number 71\n",
      "image number 72\n",
      "image number 73\n",
      "image number 74\n",
      "image number 75\n",
      "image number 76\n",
      "image number 77\n",
      "image number 78\n",
      "image number 79\n",
      "image number 80\n",
      "image number 81\n",
      "image number 82\n",
      "image number 83\n",
      "image number 84\n",
      "image number 85\n",
      "image number 86\n",
      "image number 87\n",
      "image number 88\n",
      "image number 89\n",
      "image number 90\n",
      "image number 91\n",
      "image number 92\n",
      "image number 93\n",
      "image number 94\n",
      "image number 95\n",
      "image number 96\n",
      "image number 97\n",
      "image number 98\n",
      "image number 99\n",
      "image number 100\n",
      "image number 101\n",
      "image number 102\n",
      "image number 103\n",
      "image number 104\n",
      "image number 105\n",
      "image number 106\n",
      "image number 107\n",
      "image number 108\n",
      "image number 109\n",
      "image number 110\n",
      "image number 111\n",
      "image number 112\n",
      "image number 113\n",
      "image number 114\n",
      "image number 115\n",
      "image number 116\n",
      "image number 117\n",
      "image number 118\n",
      "image number 119\n",
      "image number 120\n",
      "image number 121\n",
      "image number 122\n",
      "image number 123\n",
      "image number 124\n",
      "image number 125\n",
      "image number 126\n",
      "image number 127\n",
      "image number 128\n",
      "image number 129\n",
      "image number 130\n",
      "image number 131\n",
      "image number 132\n",
      "image number 133\n",
      "image number 134\n",
      "image number 135\n",
      "image number 136\n",
      "image number 137\n",
      "image number 138\n",
      "image number 139\n",
      "image number 140\n",
      "image number 141\n",
      "image number 142\n",
      "image number 143\n",
      "image number 144\n",
      "image number 145\n",
      "image number 146\n",
      "image number 147\n",
      "image number 148\n",
      "image number 149\n",
      "image number 150\n",
      "image number 151\n",
      "image number 152\n",
      "image number 153\n",
      "image number 154\n",
      "image number 155\n",
      "image number 156\n",
      "image number 157\n",
      "image number 158\n",
      "image number 159\n",
      "image number 160\n",
      "image number 161\n",
      "image number 162\n",
      "image number 163\n",
      "image number 164\n",
      "image number 165\n",
      "image number 166\n",
      "image number 167\n",
      "image number 168\n",
      "image number 169\n",
      "image number 170\n",
      "image number 171\n",
      "image number 172\n",
      "image number 173\n",
      "image number 174\n",
      "image number 175\n",
      "image number 176\n",
      "image number 177\n",
      "image number 178\n",
      "image number 179\n",
      "image number 180\n",
      "image number 181\n",
      "image number 182\n",
      "image number 183\n",
      "image number 184\n",
      "image number 185\n",
      "image number 186\n",
      "image number 187\n",
      "image number 188\n",
      "image number 189\n",
      "image number 190\n",
      "image number 191\n",
      "image number 192\n",
      "image number 193\n",
      "image number 194\n",
      "image number 195\n",
      "image number 196\n",
      "image number 197\n",
      "image number 198\n",
      "image number 199\n",
      "image number 200\n",
      "image number 201\n",
      "image number 202\n",
      "image number 203\n",
      "image number 204\n",
      "image number 205\n",
      "image number 206\n",
      "image number 207\n",
      "image number 208\n",
      "image number 209\n",
      "image number 210\n",
      "image number 211\n",
      "image number 212\n",
      "image number 213\n",
      "image number 214\n",
      "image number 215\n",
      "image number 216\n",
      "image number 217\n",
      "image number 218\n",
      "image number 219\n",
      "image number 220\n",
      "image number 221\n",
      "image number 222\n",
      "image number 223\n",
      "image number 224\n",
      "image number 225\n",
      "image number 226\n",
      "image number 227\n",
      "image number 228\n",
      "image number 229\n",
      "image number 230\n",
      "image number 231\n",
      "image number 232\n",
      "image number 233\n",
      "image number 234\n",
      "image number 235\n",
      "image number 236\n",
      "image number 237\n",
      "image number 238\n",
      "image number 239\n",
      "image number 240\n",
      "image number 241\n",
      "image number 242\n",
      "image number 243\n",
      "image number 244\n",
      "image number 245\n",
      "image number 246\n",
      "image number 247\n",
      "image number 248\n",
      "image number 249\n",
      "image number 250\n",
      "image number 251\n",
      "image number 252\n",
      "image number 253\n",
      "image number 254\n",
      "image number 255\n",
      "image number 256\n",
      "image number 257\n",
      "image number 258\n",
      "image number 259\n",
      "image number 260\n",
      "image number 261\n",
      "image number 262\n",
      "image number 263\n",
      "image number 264\n",
      "image number 265\n",
      "image number 266\n",
      "image number 267\n",
      "image number 268\n",
      "image number 269\n",
      "image number 270\n",
      "image number 271\n",
      "image number 272\n",
      "image number 273\n",
      "image number 274\n",
      "image number 275\n",
      "image number 276\n",
      "image number 277\n",
      "image number 278\n",
      "image number 279\n",
      "image number 280\n",
      "image number 281\n",
      "image number 282\n",
      "image number 283\n",
      "image number 284\n",
      "image number 285\n",
      "image number 286\n",
      "image number 287\n",
      "image number 288\n",
      "image number 289\n",
      "image number 290\n",
      "image number 291\n",
      "image number 292\n",
      "image number 293\n",
      "image number 294\n",
      "image number 295\n",
      "image number 296\n",
      "image number 297\n",
      "image number 298\n",
      "59674 descriptors before clustering\n",
      "Using clustering model MiniBatchKMeans(batch_size=100, compute_labels=True, init='k-means++',\n",
      "        init_size=None, max_iter=100, max_no_improvement=10,\n",
      "        n_clusters=600, n_init=3, random_state=None,\n",
      "        reassignment_ratio=0.01, tol=0.0, verbose=0)...\n",
      "Clustering on training set to get codebook of 600 words\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/avaneeshkumar/.virtualenvs/cv/lib/python2.7/site-packages/sklearn/cluster/k_means_.py:1381: RuntimeWarning: init_size=300 should be larger than k=600. Setting it to 3*k\n",
      "  init_size=init_size)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done clustering.\n",
      "\n",
      " calculting histogram vectors of training set\n",
      "\n",
      " computing descriptors of test data and producing corresponding labels\n",
      "computing descriptors: 28\n",
      "image number 0\n",
      "image number 1\n",
      "image number 2\n",
      "image number 3\n",
      "image number 4\n",
      "image number 5\n",
      "image number 6\n",
      "image number 7\n",
      "image number 8\n",
      "image number 9\n",
      "image number 10\n",
      "image number 11\n",
      "image number 12\n",
      "image number 13\n",
      "image number 14\n",
      "image number 15\n",
      "image number 16\n",
      "image number 17\n",
      "image number 18\n",
      "image number 19\n",
      "image number 20\n",
      "image number 21\n",
      "image number 22\n",
      "image number 23\n",
      "image number 24\n",
      "image number 25\n",
      "image number 26\n",
      "image number 27\n",
      "\n",
      " calculting histogram vectors of test set\n",
      "\n",
      "SVM Scores: \n",
      "train score (recall_micro): 0.741610738255\n",
      "test score (recall_micro): 0.607142857143\n",
      "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
      "  decision_function_shape=None, degree=3, gamma=0.0001, kernel='rbf',\n",
      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
      "  tol=0.001, verbose=False)\n",
      "\n",
      "*** K=600 DONE ***\n",
      "\n",
      "**************************\n",
      "***** FINISHED ALL K *****\n",
      "**************************\n",
      "\n"
     ]
    }
   ],
   "source": [
    "scoring = 'recall_micro'\n",
    "results = {}\n",
    "K_vals = [100,150,200,350,400,450,500,530,550,600]\n",
    "\n",
    "for K in K_vals:\n",
    "    X_train, Y_train, X_test, Y_test, cluster_model = cluster_split(ld, K,0.1)\n",
    "    X_train=X_train.reshape(len(X_train),K)\n",
    "    X_test=X_test.reshape(len(X_test),K)\n",
    "    print '\\nSVM Scores: '\n",
    "    svmGS, svm_score = run_svm(X_train, X_test, Y_train, Y_test, scoring)\n",
    "    \n",
    "    results[K] = dict(svmGS=svmGS,cluster_model=cluster_model,svm_score=svm_score)\n",
    "    print '\\n*** K=%i DONE ***\\n' % K\n",
    "\n",
    "print '**************************'\n",
    "print '***** FINISHED ALL K *****'\n",
    "print '**************************\\n'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "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": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}