notebooks/cross_val_scores.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append('../')\n",
    "import classify"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Running Random Forests...\n",
      "CTL size: 149 ACT size: 132\n",
      "10-fold CV Acc Mean:  0.95713373472\n",
      "CV Scores:  1.0, 0.965517241379, 0.964285714286, 0.964285714286, 0.964285714286, 0.892857142857, 0.928571428571, 0.964285714286, 0.964285714286, 0.962962962963\n",
      "OOB score: 0.953736654804\n",
      "Feature Importances:\n",
      "(\"('AY', 'AZ')_corre\", 0.084043884190262527)\n",
      "(\"('AX', 'AZ')_corre\", 0.063301751465762321)\n",
      "('AX_std', 0.05243036056614292)\n",
      "('AY_std', 0.05164181546140284)\n",
      "('A_mag_max', 0.049407161491838034)\n",
      "(\"('AX', 'AY')_corre\", 0.048872744935373226)\n",
      "('AX_max', 0.042628570197542334)\n",
      "('AZ_std', 0.042482182587204084)\n",
      "('A_mag_mean', 0.038506361534555514)\n",
      "('Fx_mean', 0.035044860186018961)\n",
      "('AZ_max', 0.034326680511527902)\n",
      "('AY_max', 0.028677785453778769)\n",
      "('AZ_min', 0.027542797528111052)\n",
      "('Mx_energy', 0.026916125988044742)\n",
      "('Fx_min', 0.025831415338299184)\n",
      "('A_mag_std', 0.022280530163693579)\n",
      "(\"('Mx', 'Mz')_corre\", 0.018172866067367587)\n",
      "('AY_min', 0.013588863460722037)\n",
      "('A_mag_energy', 0.013337483342551179)\n",
      "('Mz_mean', 0.012952128902698555)\n",
      "('Mx_mean', 0.012881410723728894)\n",
      "('Mx_min', 0.01269097775174588)\n",
      "('My_min', 0.01256438841522111)\n",
      "('AX_min', 0.012170413850049648)\n",
      "('AZ_energy', 0.011804227266923387)\n",
      "('F_mag_min', 0.0083973033372573905)\n",
      "('My_max', 0.00787520434615409)\n",
      "('My_energy', 0.0075740745145130642)\n",
      "('Fy_max', 0.0072492117481464469)\n",
      "('Fy_std', 0.006526663502045219)\n",
      "(\"('Fx', 'Fy')_corre\", 0.0063374072717435849)\n",
      "('Fy_energy', 0.0060476327091941104)\n",
      "('M_mag_entropy', 0.0057064111467481091)\n",
      "('M_mag_mean', 0.0055149139531732718)\n",
      "('F_mag_mean', 0.0052369394672492032)\n",
      "('M_mag_max', 0.0052330514114690774)\n",
      "('Fx_energy', 0.0048467476510516408)\n",
      "('M_mag_energy', 0.0047915028256430084)\n",
      "('Mx_std', 0.0047899289066466802)\n",
      "('Mz_max', 0.0045751319302166932)\n",
      "('AY_energy', 0.0038646037319138762)\n",
      "('Fz_std', 0.0038210592331040325)\n",
      "('My_std', 0.0035193917362036807)\n",
      "('AX_energy', 0.0033381585951717551)\n",
      "('Mz_std', 0.003327054600321126)\n",
      "('Fy_min', 0.0031460089950777858)\n",
      "('Mz_energy', 0.0031292837686783086)\n",
      "('Fx_entropy', 0.0029362202290658191)\n",
      "('F_mag_energy', 0.0028847939600916851)\n",
      "('Mz_kurtosis', 0.0028770988475047225)\n",
      "('A_mag_min', 0.0027957850864507678)\n",
      "('F_mag_max', 0.0027090523206510318)\n",
      "('Fy_skew', 0.0027019043232409988)\n",
      "('M_mag_min', 0.0026276439263293954)\n",
      "('Mz_min', 0.002613570757428324)\n",
      "('Mx_kurtosis', 0.0023237116171709511)\n",
      "('Fz_energy', 0.0022714670638137747)\n",
      "('My_mean', 0.0022348165385732742)\n",
      "('Fy_mean', 0.0022205445799572837)\n",
      "('Mx_entropy', 0.0021909373263747215)\n",
      "('Fy_entropy', 0.0021478153060554023)\n",
      "('M_mag_std', 0.0021074281063220126)\n",
      "('My_entropy', 0.0020766791964278775)\n",
      "('Fz_mean', 0.0020466229808926094)\n",
      "('F_mag_entropy', 0.0020454870658639644)\n",
      "(\"('My', 'Mz')_corre\", 0.0020350590339342891)\n",
      "('Fx_max', 0.0020176998824090308)\n",
      "('AZ_mean', 0.0020117321544436432)\n",
      "('AZ_skew', 0.0019332398159235165)\n",
      "('Fx_skew', 0.0019237014083798431)\n",
      "('AZ_entropy', 0.0019195959315541603)\n",
      "('M_mag_kurtosis', 0.0018868035776318721)\n",
      "('Fx_kurtosis', 0.0017398702374404509)\n",
      "('AY_mean', 0.0015629641314381568)\n",
      "('Fz_entropy', 0.0014983484070083672)\n",
      "('Mx_max', 0.0014476819207363525)\n",
      "('Mz_entropy', 0.0014476423131257635)\n",
      "('AX_skew', 0.0014135089371540301)\n",
      "(\"('Fy', 'Fz')_corre\", 0.0013824393441682786)\n",
      "('AZ_kurtosis', 0.0013294091374475661)\n",
      "(\"('Fx', 'Fz')_corre\", 0.0012328218973346771)\n",
      "('AX_kurtosis', 0.0012189529973740667)\n",
      "('A_mag_entropy', 0.0011475307021449786)\n",
      "('AX_mean', 0.0011196433911963418)\n",
      "('A_mag_kurtosis', 0.0011155245590784972)\n",
      "('Fz_min', 0.0010913935279658891)\n",
      "(\"('Mx', 'My')_corre\", 0.0010641179508443094)\n",
      "('AX_entropy', 0.0010343973774515581)\n",
      "('F_mag_std', 0.0010246246279345967)\n",
      "('AY_entropy', 0.0010032454455463653)\n",
      "('My_kurtosis', 0.00096849115515008747)\n",
      "('A_mag_skew', 0.00093422163873435799)\n",
      "('AY_kurtosis', 0.00093072645751188568)\n",
      "('Fz_kurtosis', 0.00092558571769310583)\n",
      "('Fx_std', 0.00084875575681268107)\n",
      "('Fz_max', 0.00083688230938356169)\n",
      "('Fy_kurtosis', 0.00078761977540527323)\n",
      "('F_mag_kurtosis', 0.00076233486117708801)\n",
      "('M_mag_skew', 0.0006932186620042419)\n",
      "('Fz_skew', 0.00064993678486127476)\n",
      "('F_mag_skew', 0.00062078494549899224)\n",
      "('AY_skew', 0.00055419780857715888)\n",
      "('My_skew', 0.0004154413921030563)\n",
      "('Mz_skew', 0.00040530202719301885)\n",
      "('Mx_skew', 0.00030750000372456129)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "             max_depth=None, max_features='sqrt', max_leaf_nodes=None,\n",
       "             min_samples_leaf=1, min_samples_split=2,\n",
       "             min_weight_fraction_leaf=0.0, n_estimators=200, n_jobs=1,\n",
       "             oob_score=True, random_state=None, verbose=0, warm_start=False),\n",
       " StandardScaler(copy=True, with_mean=True, with_std=True))"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classify.do_random_forests_cross_val()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Running XGB Trees...\n",
      "CTL size: 149 ACT size: 132\n",
      "10-fold CV Acc Mean:  0.964144316731\n",
      "CV Scores:  0.965517241379, 1.0, 1.0, 1.0, 0.964285714286, 0.928571428571, 1.0, 0.964285714286, 0.892857142857, 0.925925925926\n",
      "Feature Importances:\n",
      "('My_min', 0.0452681258447451)\n",
      "('Fy_skew', 0.043667114173486184)\n",
      "(\"('Mx', 'Mz')_corre\", 0.037675865603138917)\n",
      "('My_mean', 0.035337230507206725)\n",
      "('Fx_min', 0.02935683882714131)\n",
      "('M_mag_energy', 0.026198695548547667)\n",
      "('AX_std', 0.023251934466825604)\n",
      "('Fx_mean', 0.02143888571622013)\n",
      "('AZ_std', 0.019969171832504899)\n",
      "('A_mag_max', 0.018386353731475055)\n",
      "('A_mag_std', 0.017167014765036574)\n",
      "(\"('AX', 'AY')_corre\", 0.016949981918934681)\n",
      "('Mx_mean', 0.015420566435391095)\n",
      "('AY_mean', 0.014280324629633235)\n",
      "('AY_std', 0.014120672034549468)\n",
      "('My_energy', 0.01384850418889098)\n",
      "('AY_max', 0.013752911952180515)\n",
      "('M_mag_entropy', 0.013620663369439458)\n",
      "('Mz_entropy', 0.013340075831080126)\n",
      "('F_mag_min', 0.012588476311060104)\n",
      "('AZ_max', 0.012340523070183497)\n",
      "('M_mag_max', 0.011971551293387506)\n",
      "('My_entropy', 0.011666485173951514)\n",
      "('F_mag_energy', 0.01142221237261797)\n",
      "('My_max', 0.011418226256292185)\n",
      "(\"('AX', 'AZ')_corre\", 0.011087077150120091)\n",
      "('Mz_min', 0.010612611652693844)\n",
      "('M_mag_kurtosis', 0.010399615254298011)\n",
      "('Mz_kurtosis', 0.010289736433416866)\n",
      "('F_mag_mean', 0.010144017948216051)\n",
      "('AZ_min', 0.0098970700684035979)\n",
      "(\"('My', 'Mz')_corre\", 0.0096219189527095624)\n",
      "('Mx_std', 0.0094148013158274584)\n",
      "('AY_min', 0.0088724075886252932)\n",
      "('Mx_entropy', 0.0086053846985589856)\n",
      "('Mx_energy', 0.0084515674032027352)\n",
      "(\"('AY', 'AZ')_corre\", 0.0082525302977423439)\n",
      "('A_mag_min', 0.0079357742290349677)\n",
      "('A_mag_mean', 0.0076808923161397957)\n",
      "('Fy_std', 0.0074604520883301629)\n",
      "('AZ_mean', 0.0074509702558369058)\n",
      "('Fz_mean', 0.0073965211703166858)\n",
      "('Mz_mean', 0.0066988747118623083)\n",
      "('Fy_entropy', 0.0065634296946129487)\n",
      "('Fx_energy', 0.0065262818905520922)\n",
      "('Fy_max', 0.0061192486146436379)\n",
      "('AZ_energy', 0.0059445190284766799)\n",
      "(\"('Fx', 'Fy')_corre\", 0.0059385299133004436)\n",
      "('A_mag_energy', 0.0058016476663857609)\n",
      "('My_skew', 0.0057754431286433016)\n",
      "(\"('Mx', 'My')_corre\", 0.0054470450086850442)\n",
      "('AX_mean', 0.0054277451513555378)\n",
      "('AX_max', 0.0052645595633466083)\n",
      "('Fy_energy', 0.0052627929288487883)\n",
      "('F_mag_skew', 0.0051880829864864907)\n",
      "('Mx_max', 0.0050321770919868978)\n",
      "('Mz_max', 0.0049688709865036021)\n",
      "('AX_min', 0.0048716017427292944)\n",
      "('AY_energy', 0.0048517764757937893)\n",
      "('F_mag_max', 0.004754117641395922)\n",
      "('My_std', 0.0046910173135187687)\n",
      "('M_mag_min', 0.0045745182033841538)\n",
      "('Fz_energy', 0.0045148657067760282)\n",
      "('Mx_min', 0.0044374686145603574)\n",
      "('A_mag_kurtosis', 0.0042558409075795335)\n",
      "('Fx_skew', 0.0041399340919038054)\n",
      "('AY_entropy', 0.0040395633410387679)\n",
      "('Fy_min', 0.0040129023401770216)\n",
      "('AX_entropy', 0.0036065120165742516)\n",
      "('My_kurtosis', 0.0034721287384389334)\n",
      "('M_mag_std', 0.003361926432382228)\n",
      "('Mz_std', 0.0032716125084498193)\n",
      "('Fx_max', 0.0032053476702854199)\n",
      "('Fz_min', 0.0031772839800762597)\n",
      "('Fz_max', 0.0030407409183476259)\n",
      "('M_mag_mean', 0.0029829577085809271)\n",
      "('F_mag_entropy', 0.0028158896832537633)\n",
      "('AX_energy', 0.0027965606153249984)\n",
      "(\"('Fy', 'Fz')_corre\", 0.0027908497430944308)\n",
      "('A_mag_entropy', 0.0026758793280089897)\n",
      "('Fy_mean', 0.0026248338707150353)\n",
      "('Fx_std', 0.0026172702976377409)\n",
      "('Fz_std', 0.0025438253695736629)\n",
      "('F_mag_kurtosis', 0.0022036288446038626)\n",
      "('Fx_entropy', 0.002177140875913064)\n",
      "('AX_kurtosis', 0.0016217519800365498)\n",
      "('M_mag_skew', 0.0015564556500569638)\n",
      "('AY_skew', 0.0015119171069694877)\n",
      "('Mz_energy', 0.0014811446724209123)\n",
      "(\"('Fx', 'Fz')_corre\", 0.00093258009170239147)\n",
      "('F_mag_std', 0.00092634892103790026)\n",
      "('AY_kurtosis', 0.00090879263078091227)\n",
      "('Fy_kurtosis', 0.00090844774812469322)\n",
      "('AZ_skew', 0.00059468812856714587)\n",
      "('Fz_entropy', 0.00040631916109789445)\n",
      "('AZ_entropy', 0.00034303039628224343)\n",
      "('AZ_kurtosis', 0.0003295205355453936)\n",
      "('A_mag_skew', 0.0003047788328325023)\n",
      "('Fx_kurtosis', 0.00018298009198353698)\n",
      "('AX_skew', 0.0001769106174082793)\n",
      "('Fz_skew', 0.00017240858591086668)\n",
      "('Fz_kurtosis', 8.460745607590214e-05)\n",
      "('Mz_skew', 3.1010843126399232e-05)\n",
      "('Mx_skew', 2.7308526837612964e-05)\n",
      "('Mx_kurtosis', 0.0)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(GradientBoostingClassifier(init=None, learning_rate=0.1, loss='deviance',\n",
       "               max_depth=3, max_features='sqrt', max_leaf_nodes=None,\n",
       "               min_samples_leaf=1, min_samples_split=2,\n",
       "               min_weight_fraction_leaf=0.0, n_estimators=200,\n",
       "               presort='auto', random_state=None, subsample=1.0, verbose=0,\n",
       "               warm_start=False),\n",
       " StandardScaler(copy=True, with_mean=True, with_std=True))"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classify.do_xgb_trees_cross_val()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Running SVC...\n",
      "CTL size: 149 ACT size: 132\n",
      "10-fold CV Acc Mean:  0.960431490604\n",
      "CV Scores:  0.965517241379, 0.965517241379, 1.0, 0.964285714286, 0.964285714286, 1.0, 0.964285714286, 0.928571428571, 1.0, 0.851851851852\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "   decision_function_shape=None, degree=3, gamma='auto', kernel='rbf',\n",
       "   max_iter=-1, probability=True, random_state=None, shrinking=True,\n",
       "   tol=0.001, verbose=False),\n",
       " StandardScaler(copy=True, with_mean=True, with_std=True))"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classify.do_svc_cross_val()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Running Neural Network...\n",
      "CTL size: 149 ACT size: 132\n",
      "Step #1, avg. loss: 0.98262\n",
      "Step #201, epoch #67, avg. loss: 0.05294\n",
      "Step #401, epoch #133, avg. loss: 0.00411\n",
      "Step #601, epoch #200, avg. loss: 0.00170\n",
      "Step #801, epoch #267, avg. loss: 0.00099\n",
      "Step #1001, epoch #333, avg. loss: 0.00068\n",
      "Step #1201, epoch #400, avg. loss: 0.00051\n",
      "Step #1401, epoch #467, avg. loss: 0.00040\n",
      "Step #1601, epoch #533, avg. loss: 0.00033\n",
      "Step #1801, epoch #600, avg. loss: 0.00028\n",
      "Step #1, avg. loss: 0.97563\n",
      "Step #201, epoch #67, avg. loss: 0.05468\n",
      "Step #401, epoch #133, avg. loss: 0.00487\n",
      "Step #601, epoch #200, avg. loss: 0.00189\n",
      "Step #801, epoch #267, avg. loss: 0.00108\n",
      "Step #1001, epoch #333, avg. loss: 0.00075\n",
      "Step #1201, epoch #400, avg. loss: 0.00056\n",
      "Step #1401, epoch #467, avg. loss: 0.00043\n",
      "Step #1601, epoch #533, avg. loss: 0.00035\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 0.92649\n",
      "Step #201, epoch #67, avg. loss: 0.06095\n",
      "Step #401, epoch #133, avg. loss: 0.00480\n",
      "Step #601, epoch #200, avg. loss: 0.00189\n",
      "Step #801, epoch #267, avg. loss: 0.00108\n",
      "Step #1001, epoch #333, avg. loss: 0.00075\n",
      "Step #1201, epoch #400, avg. loss: 0.00055\n",
      "Step #1401, epoch #467, avg. loss: 0.00044\n",
      "Step #1601, epoch #533, avg. loss: 0.00036\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 1.00535\n",
      "Step #201, epoch #67, avg. loss: 0.05830\n",
      "Step #401, epoch #133, avg. loss: 0.00457\n",
      "Step #601, epoch #200, avg. loss: 0.00182\n",
      "Step #801, epoch #267, avg. loss: 0.00105\n",
      "Step #1001, epoch #333, avg. loss: 0.00072\n",
      "Step #1201, epoch #400, avg. loss: 0.00054\n",
      "Step #1401, epoch #467, avg. loss: 0.00043\n",
      "Step #1601, epoch #533, avg. loss: 0.00035\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 0.99446\n",
      "Step #201, epoch #67, avg. loss: 0.06088\n",
      "Step #401, epoch #133, avg. loss: 0.00498\n",
      "Step #601, epoch #200, avg. loss: 0.00191\n",
      "Step #801, epoch #267, avg. loss: 0.00112\n",
      "Step #1001, epoch #333, avg. loss: 0.00076\n",
      "Step #1201, epoch #400, avg. loss: 0.00057\n",
      "Step #1401, epoch #467, avg. loss: 0.00045\n",
      "Step #1601, epoch #533, avg. loss: 0.00036\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 0.94047\n",
      "Step #201, epoch #67, avg. loss: 0.05992\n",
      "Step #401, epoch #133, avg. loss: 0.00468\n",
      "Step #601, epoch #200, avg. loss: 0.00185\n",
      "Step #801, epoch #267, avg. loss: 0.00106\n",
      "Step #1001, epoch #333, avg. loss: 0.00073\n",
      "Step #1201, epoch #400, avg. loss: 0.00055\n",
      "Step #1401, epoch #467, avg. loss: 0.00043\n",
      "Step #1601, epoch #533, avg. loss: 0.00036\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 0.87530\n",
      "Step #201, epoch #67, avg. loss: 0.06032\n",
      "Step #401, epoch #133, avg. loss: 0.00491\n",
      "Step #601, epoch #200, avg. loss: 0.00192\n",
      "Step #801, epoch #267, avg. loss: 0.00110\n",
      "Step #1001, epoch #333, avg. loss: 0.00076\n",
      "Step #1201, epoch #400, avg. loss: 0.00056\n",
      "Step #1401, epoch #467, avg. loss: 0.00044\n",
      "Step #1601, epoch #533, avg. loss: 0.00036\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 0.94575\n",
      "Step #201, epoch #67, avg. loss: 0.05901\n",
      "Step #401, epoch #133, avg. loss: 0.00510\n",
      "Step #601, epoch #200, avg. loss: 0.00195\n",
      "Step #801, epoch #267, avg. loss: 0.00111\n",
      "Step #1001, epoch #333, avg. loss: 0.00076\n",
      "Step #1201, epoch #400, avg. loss: 0.00057\n",
      "Step #1401, epoch #467, avg. loss: 0.00045\n",
      "Step #1601, epoch #533, avg. loss: 0.00037\n",
      "Step #1801, epoch #600, avg. loss: 0.00031\n",
      "Step #1, avg. loss: 0.99495\n",
      "Step #201, epoch #67, avg. loss: 0.05632\n",
      "Step #401, epoch #133, avg. loss: 0.00455\n",
      "Step #601, epoch #200, avg. loss: 0.00183\n",
      "Step #801, epoch #267, avg. loss: 0.00104\n",
      "Step #1001, epoch #333, avg. loss: 0.00072\n",
      "Step #1201, epoch #400, avg. loss: 0.00054\n",
      "Step #1401, epoch #467, avg. loss: 0.00043\n",
      "Step #1601, epoch #533, avg. loss: 0.00035\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 1.01974\n",
      "Step #201, epoch #67, avg. loss: 0.05959\n",
      "Step #401, epoch #133, avg. loss: 0.00488\n",
      "Step #601, epoch #200, avg. loss: 0.00192\n",
      "Step #801, epoch #267, avg. loss: 0.00111\n",
      "Step #1001, epoch #333, avg. loss: 0.00075\n",
      "Step #1201, epoch #400, avg. loss: 0.00056\n",
      "Step #1401, epoch #467, avg. loss: 0.00044\n",
      "Step #1601, epoch #533, avg. loss: 0.00036\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "10-fold CV Acc Mean:  0.950369458128\n",
      "CV Scores:  0.896551724138, 0.928571428571, 0.964285714286, 0.928571428571, 1.0, 1.0, 0.928571428571, 0.928571428571, 0.928571428571, 1.0\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(TensorFlowEstimator(batch_size=100, class_weight=None,\n",
       "           continue_training=False, early_stopping_rounds=None,\n",
       "           keep_checkpoint_every_n_hours=10000, learning_rate=0.1,\n",
       "           max_to_keep=5, model_fn=<function relu_dnn at 0x110362ed8>,\n",
       "           n_classes=2, num_cores=4, optimizer='SGD', steps=2000,\n",
       "           tf_master='', tf_random_seed=42, verbose=1),\n",
       " StandardScaler(copy=True, with_mean=True, with_std=True))"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classify.do_dnn_cross_val()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Running Ensemble Cross Val...\n",
      "CTL size: 149 ACT size: 132\n",
      "Step #1, avg. loss: 0.87174\n",
      "Step #201, epoch #67, avg. loss: 0.05903\n",
      "Step #401, epoch #133, avg. loss: 0.00475\n",
      "Step #601, epoch #200, avg. loss: 0.00183\n",
      "Step #801, epoch #267, avg. loss: 0.00106\n",
      "Step #1001, epoch #333, avg. loss: 0.00073\n",
      "Step #1201, epoch #400, avg. loss: 0.00055\n",
      "Step #1401, epoch #467, avg. loss: 0.00043\n",
      "Step #1601, epoch #533, avg. loss: 0.00036\n",
      "Step #1801, epoch #600, avg. loss: 0.00029\n",
      "Step #1, avg. loss: 0.84264\n",
      "Step #201, epoch #67, avg. loss: 0.05692\n",
      "Step #401, epoch #133, avg. loss: 0.00423\n",
      "Step #601, epoch #200, avg. loss: 0.00170\n",
      "Step #801, epoch #267, avg. loss: 0.00101\n",
      "Step #1001, epoch #333, avg. loss: 0.00069\n",
      "Step #1201, epoch #400, avg. loss: 0.00051\n",
      "Step #1401, epoch #467, avg. loss: 0.00041\n",
      "Step #1601, epoch #533, avg. loss: 0.00034\n",
      "Step #1801, epoch #600, avg. loss: 0.00028\n",
      "Step #1, avg. loss: 0.95563\n",
      "Step #201, epoch #67, avg. loss: 0.05766\n",
      "Step #401, epoch #133, avg. loss: 0.00456\n",
      "Step #601, epoch #200, avg. loss: 0.00179\n",
      "Step #801, epoch #267, avg. loss: 0.00104\n",
      "Step #1001, epoch #333, avg. loss: 0.00071\n",
      "Step #1201, epoch #400, avg. loss: 0.00053\n",
      "Step #1401, epoch #467, avg. loss: 0.00041\n",
      "Step #1601, epoch #533, avg. loss: 0.00034\n",
      "Step #1801, epoch #600, avg. loss: 0.00028\n",
      "Step #1, avg. loss: 1.02101\n",
      "Step #201, epoch #67, avg. loss: 0.06091\n",
      "Step #401, epoch #133, avg. loss: 0.00510\n",
      "Step #601, epoch #200, avg. loss: 0.00195\n",
      "Step #801, epoch #267, avg. loss: 0.00114\n",
      "Step #1001, epoch #333, avg. loss: 0.00076\n",
      "Step #1201, epoch #400, avg. loss: 0.00057\n",
      "Step #1401, epoch #467, avg. loss: 0.00045\n",
      "Step #1601, epoch #533, avg. loss: 0.00037\n",
      "Step #1801, epoch #600, avg. loss: 0.00031\n",
      "Step #1, avg. loss: 0.93815\n",
      "Step #201, epoch #67, avg. loss: 0.05907\n",
      "Step #401, epoch #133, avg. loss: 0.00498\n",
      "Step #601, epoch #200, avg. loss: 0.00192\n",
      "Step #801, epoch #267, avg. loss: 0.00111\n",
      "Step #1001, epoch #333, avg. loss: 0.00075\n",
      "Step #1201, epoch #400, avg. loss: 0.00056\n",
      "Step #1401, epoch #467, avg. loss: 0.00044\n",
      "Step #1601, epoch #533, avg. loss: 0.00037\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 1.06437\n",
      "Step #201, epoch #67, avg. loss: 0.06087\n",
      "Step #401, epoch #133, avg. loss: 0.00496\n",
      "Step #601, epoch #200, avg. loss: 0.00193\n",
      "Step #801, epoch #267, avg. loss: 0.00111\n",
      "Step #1001, epoch #333, avg. loss: 0.00075\n",
      "Step #1201, epoch #400, avg. loss: 0.00056\n",
      "Step #1401, epoch #467, avg. loss: 0.00044\n",
      "Step #1601, epoch #533, avg. loss: 0.00037\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 0.85960\n",
      "Step #201, epoch #67, avg. loss: 0.05878\n",
      "Step #401, epoch #133, avg. loss: 0.00504\n",
      "Step #601, epoch #200, avg. loss: 0.00193\n",
      "Step #801, epoch #267, avg. loss: 0.00112\n",
      "Step #1001, epoch #333, avg. loss: 0.00076\n",
      "Step #1201, epoch #400, avg. loss: 0.00056\n",
      "Step #1401, epoch #467, avg. loss: 0.00045\n",
      "Step #1601, epoch #533, avg. loss: 0.00037\n",
      "Step #1801, epoch #600, avg. loss: 0.00031\n",
      "Step #1, avg. loss: 0.88344\n",
      "Step #201, epoch #67, avg. loss: 0.05104\n",
      "Step #401, epoch #133, avg. loss: 0.00432\n",
      "Step #601, epoch #200, avg. loss: 0.00176\n",
      "Step #801, epoch #267, avg. loss: 0.00102\n",
      "Step #1001, epoch #333, avg. loss: 0.00070\n",
      "Step #1201, epoch #400, avg. loss: 0.00052\n",
      "Step #1401, epoch #467, avg. loss: 0.00041\n",
      "Step #1601, epoch #533, avg. loss: 0.00034\n",
      "Step #1801, epoch #600, avg. loss: 0.00028\n",
      "Step #1, avg. loss: 0.95364\n",
      "Step #201, epoch #67, avg. loss: 0.06017\n",
      "Step #401, epoch #133, avg. loss: 0.00488\n",
      "Step #601, epoch #200, avg. loss: 0.00190\n",
      "Step #801, epoch #267, avg. loss: 0.00110\n",
      "Step #1001, epoch #333, avg. loss: 0.00074\n",
      "Step #1201, epoch #400, avg. loss: 0.00054\n",
      "Step #1401, epoch #467, avg. loss: 0.00044\n",
      "Step #1601, epoch #533, avg. loss: 0.00036\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "Step #1, avg. loss: 0.86833\n",
      "Step #201, epoch #67, avg. loss: 0.05991\n",
      "Step #401, epoch #133, avg. loss: 0.00476\n",
      "Step #601, epoch #200, avg. loss: 0.00188\n",
      "Step #801, epoch #267, avg. loss: 0.00111\n",
      "Step #1001, epoch #333, avg. loss: 0.00074\n",
      "Step #1201, epoch #400, avg. loss: 0.00055\n",
      "Step #1401, epoch #467, avg. loss: 0.00044\n",
      "Step #1601, epoch #533, avg. loss: 0.00036\n",
      "Step #1801, epoch #600, avg. loss: 0.00030\n",
      "10-fold CV Acc Mean:  0.964408866995\n",
      "CV Scores:  0.965517241379, 1.0, 0.964285714286, 1.0, 0.964285714286, 1.0, 0.964285714286, 0.821428571429, 1.0, 0.964285714286\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(VotingClassifier(estimators=[('dnn', TensorFlowEstimator(batch_size=100, class_weight=None,\n",
       "           continue_training=False, early_stopping_rounds=None,\n",
       "           keep_checkpoint_every_n_hours=10000, learning_rate=0.1,\n",
       "           max_to_keep=5, model_fn=<function relu_dnn at 0x1115b1ed8>,\n",
       "           n_classes=2, num...',\n",
       "   max_iter=-1, probability=True, random_state=None, shrinking=True,\n",
       "   tol=0.001, verbose=False))],\n",
       "          voting='soft', weights=[1, 1, 1, 1]),\n",
       " StandardScaler(copy=True, with_mean=True, with_std=True))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classify.do_ensemble_cross_val()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}