2016-12-14(1).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/M2015eliu/.pyenv/versions/anaconda2-4.1.1/lib/python2.7/site-packages/gensim/utils.py:1015: UserWarning: Pattern library is not installed, lemmatization won't be available.\n",
      "  warnings.warn(\"Pattern library is not installed, lemmatization won't be available.\")\n",
      "Using gpu device 0: GeForce GTX TITAN X (CNMeM is enabled with initial size: 95.0% of memory, cuDNN 4007)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting to training the modle..., saving to weights/201612140109_e10_1k1k.p\n",
      "Training\n",
      "the length of the training data is  1998\n",
      "Before trianing, the error is:\n",
      "(4.1240959, 0.095272966, 0.046075466722721184)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [] : 10\n",
      "Epoch 0 ...\n",
      "Epoch  0 Update  40 Cost  0.112757906318\n",
      "(2.0110817, 0.76414847, 0.75776556286056573)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103] : 103\n",
      "epoch took: 60.8310480118\n",
      "Epoch 1 ...\n",
      "Epoch  1 Update  80 Cost  0.154227882624\n",
      "Epoch  1 Update  120 Cost  0.102236732841\n",
      "(1.8783286, 0.77263105, 0.7593900274694152)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118] : 118\n",
      "epoch took: 66.8899929523\n",
      "Epoch 2 ...\n",
      "Epoch  2 Update  160 Cost  0.113972477615\n",
      "(1.5767787, 0.81996793, 0.79867524781989885)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165] : 165\n",
      "epoch took: 65.3131170273\n",
      "Epoch 3 ...\n",
      "Epoch  3 Update  200 Cost  0.0882736444473\n",
      "Epoch  3 Update  240 Cost  0.0730952769518\n",
      "(1.5959966, 0.8154093, 0.79442843665525376)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187] : 187\n",
      "epoch took: 62.2263939381\n",
      "Epoch 4 ...\n",
      "Epoch  4 Update  280 Cost  0.0679541006684\n",
      "(1.3896998, 0.85671198, 0.8229655913356031)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238] : 238\n",
      "epoch took: 62.6313569546\n",
      "Epoch 5 ...\n",
      "Epoch  5 Update  320 Cost  0.0801030173898\n",
      "Epoch  5 Update  360 Cost  0.0551847554743\n",
      "(1.4355819, 0.83700067, 0.80849482000567185)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221] : 221\n",
      "epoch took: 63.1824800968\n",
      "Epoch 6 ...\n",
      "Epoch  6 Update  400 Cost  0.053138025105\n",
      "Epoch  6 Update  440 Cost  0.0752167925239\n",
      "(1.4809409, 0.82804465, 0.80357809047903928)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213] : 213\n",
      "epoch took: 63.8639528751\n",
      "Epoch 7 ...\n",
      "Epoch  7 Update  480 Cost  0.0967376902699\n",
      "(1.2287828, 0.86859202, 0.831792446961378)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360] : 360\n",
      "epoch took: 59.319257021\n",
      "Epoch 8 ...\n",
      "Epoch  8 Update  520 Cost  0.0612242631614\n",
      "Epoch  8 Update  560 Cost  0.13301704824\n",
      "(1.038427, 0.90152979, 0.84768113697772873)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388] : 388\n",
      "epoch took: 68.2661421299\n",
      "Epoch 9 ...\n",
      "Epoch  9 Update  600 Cost  0.0561785772443\n",
      "(0.98308897, 0.90103805, 0.8479171677633347)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414] : 414\n",
      "epoch took: 65.6565818787\n",
      "Epoch 10 ...\n",
      "Epoch  10 Update  640 Cost  0.0669415146112\n",
      "Epoch  10 Update  680 Cost  0.0953700616956\n",
      "(1.029853, 0.90210289, 0.84821567728630709)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403] : 403\n",
      "epoch took: 64.148542881\n",
      "Epoch 11 ...\n",
      "Epoch  11 Update  720 Cost  0.0681083649397\n",
      "(1.0817852, 0.89301783, 0.84405737116290158)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353] : 353\n",
      "epoch took: 64.3637759686\n",
      "Epoch 12 ...\n",
      "Epoch  12 Update  760 Cost  0.0648547783494\n",
      "Epoch  12 Update  800 Cost  0.0944069772959\n",
      "(1.0992512, 0.88560176, 0.83974460181172417)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353, 364] : 364\n",
      "epoch took: 62.6349179745\n",
      "Epoch 13 ...\n",
      "Epoch  13 Update  840 Cost  0.072304174304\n",
      "Epoch  13 Update  880 Cost  0.048543330282\n",
      "(1.0745524, 0.88657248, 0.83998063259733025)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353, 364, 403] : 403\n",
      "epoch took: 66.2509250641\n",
      "Epoch 14 ...\n",
      "Epoch  14 Update  920 Cost  0.0975828021765\n",
      "(0.89580035, 0.91572565, 0.85520635378940479)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353, 364, 403, 474] : 474\n",
      "epoch took: 68.3363528252\n",
      "Epoch 15 ...\n",
      "Epoch  15 Update  960 Cost  0.0546577982605\n",
      "Epoch  15 Update  1000 Cost  0.0570617802441\n",
      "(0.84340042, 0.92573893, 0.85647154822107274)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353, 364, 403, 474, 533] : 533\n",
      "epoch took: 60.0742878914\n",
      "Epoch 16 ...\n",
      "Epoch  16 Update  1040 Cost  0.0486805066466\n",
      "(0.77291435, 0.92735243, 0.85677526495000556)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353, 364, 403, 474, 533, 551] : 551\n",
      "epoch took: 56.327986002\n",
      "Epoch 17 ...\n",
      "Epoch  17 Update  1080 Cost  0.0539721436799\n",
      "Epoch  17 Update  1120 Cost  0.0677587613463\n",
      "(0.74793178, 0.92833596, 0.8567110500476437)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353, 364, 403, 474, 533, 551, 550] : 550\n",
      "epoch took: 58.7750248909\n",
      "Epoch 18 ...\n",
      "Epoch  18 Update  1160 Cost  0.0441583767533\n",
      "(0.7177375, 0.93440861, 0.85851772686981964)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353, 364, 403, 474, 533, 551, 550, 590] : 590\n",
      "epoch took: 64.8745889664\n",
      "Epoch 19 ...\n",
      "Epoch  19 Update  1200 Cost  0.0592876039445\n",
      "Epoch  19 Update  1240 Cost  0.0502151399851\n",
      "(0.70728278, 0.93290466, 0.85802831009378344)\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "top: [103, 118, 165, 187, 238, 221, 213, 360, 388, 414, 403, 353, 364, 403, 474, 533, 551, 550, 590, 576] : 576\n",
      "epoch took: 61.5027010441\n",
      "saving the model...\n",
      "Evaluate the model using fast estimation...\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n",
      "Finish preparing the data!\n",
      "Finished embedding,start projecting...\n"
     ]
    }
   ],
   "source": [
    "run main.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the results of epoch 0 :\n",
      "count    998.000000\n",
      "mean     103.648297\n",
      "std      136.337764\n",
      "min        1.500000\n",
      "25%       14.500000\n",
      "50%       55.500000\n",
      "75%      125.500000\n",
      "max      740.500000\n",
      "dtype: float64\n",
      "the results of epoch 1 :\n",
      "count    998.000000\n",
      "mean     103.785571\n",
      "std      135.316263\n",
      "min        1.500000\n",
      "25%       13.500000\n",
      "50%       43.500000\n",
      "75%      145.000000\n",
      "max      647.500000\n",
      "dtype: float64\n",
      "the results of epoch 2 :\n",
      "count    998.000000\n",
      "mean      76.000000\n",
      "std      106.055716\n",
      "min        1.500000\n",
      "25%        7.500000\n",
      "50%       29.500000\n",
      "75%      100.250000\n",
      "max      795.500000\n",
      "dtype: float64\n",
      "the results of epoch 3 :\n",
      "count    998.000000\n",
      "mean      75.280561\n",
      "std      115.215497\n",
      "min        1.500000\n",
      "25%        6.500000\n",
      "50%       31.500000\n",
      "75%       83.500000\n",
      "max      755.500000\n",
      "dtype: float64\n",
      "the results of epoch 4 :\n",
      "count    998.000000\n",
      "mean      48.267535\n",
      "std       75.114931\n",
      "min        1.500000\n",
      "25%        5.500000\n",
      "50%       17.500000\n",
      "75%       52.500000\n",
      "max      557.500000\n",
      "dtype: float64\n",
      "the results of epoch 5 :\n",
      "count    998.000000\n",
      "mean      56.690381\n",
      "std       89.763312\n",
      "min        1.500000\n",
      "25%        5.500000\n",
      "50%       20.500000\n",
      "75%       61.500000\n",
      "max      553.500000\n",
      "dtype: float64\n",
      "the results of epoch 6 :\n",
      "count    998.000000\n",
      "mean      49.181363\n",
      "std       84.377481\n",
      "min        1.500000\n",
      "25%        5.500000\n",
      "50%       14.500000\n",
      "75%       41.500000\n",
      "max      687.500000\n",
      "dtype: float64\n",
      "the results of epoch 7 :\n",
      "count    998.000000\n",
      "mean      35.654309\n",
      "std       67.453318\n",
      "min        1.500000\n",
      "25%        2.500000\n",
      "50%        9.500000\n",
      "75%       28.500000\n",
      "max      525.500000\n",
      "dtype: float64\n",
      "the results of epoch 8 :\n",
      "count    998.000000\n",
      "mean      24.778557\n",
      "std       44.304438\n",
      "min        1.500000\n",
      "25%        2.500000\n",
      "50%        7.500000\n",
      "75%       21.250000\n",
      "max      323.500000\n",
      "dtype: float64\n",
      "the results of epoch 9 :\n",
      "count    998.000000\n",
      "mean      23.700401\n",
      "std       45.248738\n",
      "min        1.500000\n",
      "25%        2.500000\n",
      "50%        6.500000\n",
      "75%       22.250000\n",
      "max      415.500000\n",
      "dtype: float64\n",
      "the results of epoch 10 :\n",
      "count    998.000000\n",
      "mean      24.431864\n",
      "std       44.456411\n",
      "min        1.500000\n",
      "25%        2.500000\n",
      "50%        6.500000\n",
      "75%       21.500000\n",
      "max      345.500000\n",
      "dtype: float64\n",
      "the results of epoch 11 :\n",
      "count    998.000000\n",
      "mean      27.363727\n",
      "std       47.609995\n",
      "min        1.500000\n",
      "25%        3.500000\n",
      "50%        8.500000\n",
      "75%       27.500000\n",
      "max      587.500000\n",
      "dtype: float64\n",
      "the results of epoch 12 :\n",
      "count    998.000000\n",
      "mean      26.564128\n",
      "std       46.564672\n",
      "min        1.500000\n",
      "25%        3.500000\n",
      "50%        9.500000\n",
      "75%       25.500000\n",
      "max      575.500000\n",
      "dtype: float64\n",
      "the results of epoch 13 :\n",
      "count    998.000000\n",
      "mean      24.756513\n",
      "std       42.688655\n",
      "min        1.500000\n",
      "25%        2.500000\n",
      "50%        6.500000\n",
      "75%       24.500000\n",
      "max      489.500000\n",
      "dtype: float64\n",
      "the results of epoch 14 :\n",
      "count    998.000000\n",
      "mean      16.604208\n",
      "std       27.148042\n",
      "min        1.500000\n",
      "25%        2.500000\n",
      "50%        5.500000\n",
      "75%       16.500000\n",
      "max      313.500000\n",
      "dtype: float64\n",
      "the results of epoch 15 :\n",
      "count    998.000000\n",
      "mean      14.019038\n",
      "std       27.340733\n",
      "min        1.500000\n",
      "25%        1.500000\n",
      "50%        4.500000\n",
      "75%       11.500000\n",
      "max      309.500000\n",
      "dtype: float64\n",
      "the results of epoch 16 :\n",
      "count    998.000000\n",
      "mean      11.874749\n",
      "std       22.331186\n",
      "min        1.500000\n",
      "25%        1.500000\n",
      "50%        4.500000\n",
      "75%       11.500000\n",
      "max      327.500000\n",
      "dtype: float64\n",
      "the results of epoch 17 :\n",
      "count    998.000000\n",
      "mean      11.315631\n",
      "std       21.037295\n",
      "min        1.500000\n",
      "25%        1.500000\n",
      "50%        4.500000\n",
      "75%       10.250000\n",
      "max      289.500000\n",
      "dtype: float64\n",
      "the results of epoch 18 :\n",
      "count    998.000000\n",
      "mean       9.826653\n",
      "std       18.342870\n",
      "min        1.500000\n",
      "25%        1.500000\n",
      "50%        3.500000\n",
      "75%        9.500000\n",
      "max      191.500000\n",
      "dtype: float64\n",
      "the results of epoch 19 :\n",
      "count    998.000000\n",
      "mean       9.359719\n",
      "std       15.731905\n",
      "min        1.500000\n",
      "25%        1.500000\n",
      "50%        3.500000\n",
      "75%        9.500000\n",
      "max      199.500000\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# 这个sls.rank记录的精度是不对的，误差增大了一倍，现在已经被修复\n",
    "for i in range(20):\n",
    "    print \"the results of epoch\", i, \":\"\n",
    "    print pd.Series(sls.rank[i][:998]).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[9.5,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 41.5,\n",
       " 1.5,\n",
       " 14.5,\n",
       " 3.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 68.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 19.5,\n",
       " 33.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 68.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 199.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 33.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 8.5,\n",
       " 33.5,\n",
       " 7.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 8.5,\n",
       " 14.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 13.5,\n",
       " 9.5,\n",
       " 4.5,\n",
       " 6.5,\n",
       " 5.5,\n",
       " 23.5,\n",
       " 1.5,\n",
       " 25.5,\n",
       " 51.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 33.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 13.5,\n",
       " 2.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 75.5,\n",
       " 1.5,\n",
       " 8.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 11.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 97.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 65.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 33.5,\n",
       " 9.5,\n",
       " 8.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 31.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 51.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 85.5,\n",
       " 10.5,\n",
       " 8.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 5.5,\n",
       " 25.5,\n",
       " 4.5,\n",
       " 17.5,\n",
       " 4.5,\n",
       " 5.5,\n",
       " 6.5,\n",
       " 69.5,\n",
       " 1.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 25.5,\n",
       " 19.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 49.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 43.5,\n",
       " 9.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 97.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 23.5,\n",
       " 5.5,\n",
       " 17.5,\n",
       " 9.5,\n",
       " 27.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 5.5,\n",
       " 20.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 25.5,\n",
       " 1.5,\n",
       " 65.5,\n",
       " 4.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 1.5,\n",
       " 19.5,\n",
       " 31.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 179.5,\n",
       " 20.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 21.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 33.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 30.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 21.5,\n",
       " 1.5,\n",
       " 32.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 41.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 17.5,\n",
       " 2.5,\n",
       " 13.5,\n",
       " 11.5,\n",
       " 3.5,\n",
       " 30.5,\n",
       " 2.5,\n",
       " 23.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 8.5,\n",
       " 13.5,\n",
       " 7.5,\n",
       " 32.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 8.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 8.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 33.5,\n",
       " 13.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 6.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 27.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 11.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 13.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 46.5,\n",
       " 31.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 46.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 5.5,\n",
       " 18.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 13.5,\n",
       " 15.5,\n",
       " 13.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 4.5,\n",
       " 4.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 13.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 4.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 57.5,\n",
       " 2.5,\n",
       " 21.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 15.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 66.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 6.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 25.5,\n",
       " 27.5,\n",
       " 37.5,\n",
       " 39.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 41.5,\n",
       " 11.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 25.5,\n",
       " 17.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 51.5,\n",
       " 17.5,\n",
       " 26.5,\n",
       " 71.5,\n",
       " 31.5,\n",
       " 2.5,\n",
       " 14.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 37.5,\n",
       " 7.5,\n",
       " 29.5,\n",
       " 4.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 13.5,\n",
       " 6.5,\n",
       " 4.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 33.5,\n",
       " 4.5,\n",
       " 9.5,\n",
       " 26.5,\n",
       " 1.5,\n",
       " 56.5,\n",
       " 8.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 26.5,\n",
       " 11.5,\n",
       " 11.5,\n",
       " 3.5,\n",
       " 29.5,\n",
       " 1.5,\n",
       " 21.5,\n",
       " 6.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 17.5,\n",
       " 13.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 14.5,\n",
       " 25.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 26.5,\n",
       " 1.5,\n",
       " 8.5,\n",
       " 4.5,\n",
       " 13.5,\n",
       " 57.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 25.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 19.5,\n",
       " 6.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 57.5,\n",
       " 2.5,\n",
       " 41.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 23.5,\n",
       " 18.5,\n",
       " 3.5,\n",
       " 27.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 10.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 66.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 5.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 13.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 3.5,\n",
       " 6.5,\n",
       " 4.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 11.5,\n",
       " 4.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 8.5,\n",
       " 6.5,\n",
       " 31.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 6.5,\n",
       " 4.5,\n",
       " 12.5,\n",
       " 6.5,\n",
       " 5.5,\n",
       " 5.5,\n",
       " 53.5,\n",
       " 39.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 7.5,\n",
       " 69.5,\n",
       " 2.5,\n",
       " 10.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 10.5,\n",
       " 6.5,\n",
       " 7.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 7.5,\n",
       " 71.5,\n",
       " 10.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 17.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 5.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 56.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 4.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 16.5,\n",
       " 3.5,\n",
       " 14.5,\n",
       " 2.5,\n",
       " 13.5,\n",
       " 5.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 25.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 23.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 13.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 47.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 13.5,\n",
       " 7.5,\n",
       " 3.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 5.5,\n",
       " 7.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 14.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 11.5,\n",
       " 49.5,\n",
       " 5.5,\n",
       " 16.5,\n",
       " 5.5,\n",
       " 12.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 17.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 15.5,\n",
       " 5.5,\n",
       " 4.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 49.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 52.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 38.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 21.5,\n",
       " 21.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 35.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 35.5,\n",
       " 3.5,\n",
       " 18.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 26.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 44.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 56.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 44.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 19.5,\n",
       " 11.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 10.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 22.5,\n",
       " 47.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 21.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 99.5,\n",
       " 3.5,\n",
       " 52.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 45.5,\n",
       " 37.5,\n",
       " 35.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 37.5,\n",
       " 44.5,\n",
       " 16.5,\n",
       " 13.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 13.5,\n",
       " 3.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 17.5,\n",
       " 4.5,\n",
       " 45.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 10.5,\n",
       " 7.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 44.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 15.5,\n",
       " 6.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 37.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 25.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 18.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 11.5,\n",
       " 16.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 21.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 31.5,\n",
       " 56.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 39.5,\n",
       " 22.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 38.5,\n",
       " 26.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sls.rank[i][:998]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[5.0,\n",
       " 8.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 21.0,\n",
       " 1.0,\n",
       " 7.5,\n",
       " 2.0,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 5.0,\n",
       " 1.5,\n",
       " 3.0,\n",
       " 5.0,\n",
       " 34.5,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 8.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 10.0,\n",
       " 17.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 5.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 34.5,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 5.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 9.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 9.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 100.0,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 17.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 17.0,\n",
       " 4.0,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 7.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 4.0,\n",
       " 7.0,\n",
       " 5.0,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 3.0,\n",
       " 12.0,\n",
       " 1.0,\n",
       " 13.0,\n",
       " 26.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 17.0,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 5.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 7.0,\n",
       " 1.5,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 38.0,\n",
       " 1.0,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 6.0,\n",
       " 1.5,\n",
       " 4.0,\n",
       " 49.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 5.5,\n",
       " 4.5,\n",
       " 1.0,\n",
       " 33.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 17.0,\n",
       " 5.0,\n",
       " 4.5,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 16.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 26.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 4.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 6.0,\n",
       " 1.0,\n",
       " 43.0,\n",
       " 5.5,\n",
       " 4.5,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 3.0,\n",
       " 13.0,\n",
       " 2.5,\n",
       " 9.0,\n",
       " 2.5,\n",
       " 3.0,\n",
       " 3.5,\n",
       " 35.0,\n",
       " 1.0,\n",
       " 9.0,\n",
       " 3.0,\n",
       " 13.0,\n",
       " 10.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 1.0,\n",
       " 25.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 22.0,\n",
       " 5.0,\n",
       " 5.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 49.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 12.0,\n",
       " 3.0,\n",
       " 9.0,\n",
       " 5.0,\n",
       " 14.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 5.0,\n",
       " 3.0,\n",
       " 10.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 13.0,\n",
       " 1.0,\n",
       " 33.0,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 5.5,\n",
       " 1.0,\n",
       " 10.0,\n",
       " 16.0,\n",
       " 5.0,\n",
       " 1.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 4.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 90.0,\n",
       " 10.5,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 11.0,\n",
       " 1.0,\n",
       " 5.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 5.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 1.5,\n",
       " 17.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 5.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 15.5,\n",
       " 4.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 11.0,\n",
       " 1.0,\n",
       " 16.5,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 21.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 9.0,\n",
       " 1.5,\n",
       " 7.0,\n",
       " 6.0,\n",
       " 2.0,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 12.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 2.5,\n",
       " 5.0,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 8.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 7.0,\n",
       " 4.0,\n",
       " 16.5,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 5.0,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 5.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 6.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 17.0,\n",
       " 7.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 6.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.5,\n",
       " 14.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 6.0,\n",
       " 1.5,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 2.5,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 7.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 23.5,\n",
       " 16.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 23.5,\n",
       " 1.0,\n",
       " 5.5,\n",
       " 3.0,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 4.0,\n",
       " 7.0,\n",
       " 8.0,\n",
       " 7.0,\n",
       " 3.0,\n",
       " 5.0,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 4.0,\n",
       " 1.5,\n",
       " 7.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 4.0,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 5.0,\n",
       " 2.0,\n",
       " 5.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 29.0,\n",
       " 1.5,\n",
       " 11.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 6.0,\n",
       " 1.0,\n",
       " 6.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 8.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.5,\n",
       " 33.5,\n",
       " 1.5,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 3.5,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 13.0,\n",
       " 14.0,\n",
       " 19.0,\n",
       " 20.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 3.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 21.0,\n",
       " 6.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 13.0,\n",
       " 9.0,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 26.0,\n",
       " 9.0,\n",
       " 13.5,\n",
       " 36.0,\n",
       " 16.0,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 19.0,\n",
       " 4.0,\n",
       " 15.0,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 7.0,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 17.0,\n",
       " 2.5,\n",
       " 5.0,\n",
       " 13.5,\n",
       " 1.0,\n",
       " 28.5,\n",
       " 4.5,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 13.5,\n",
       " 6.0,\n",
       " 6.0,\n",
       " 2.0,\n",
       " 15.0,\n",
       " 1.0,\n",
       " 11.0,\n",
       " 3.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 9.0,\n",
       " 7.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 7.5,\n",
       " 13.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 13.5,\n",
       " 1.0,\n",
       " 4.5,\n",
       " 2.5,\n",
       " 7.0,\n",
       " 29.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 13.0,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 10.0,\n",
       " 3.5,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 5.0,\n",
       " 29.0,\n",
       " 1.5,\n",
       " 21.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 12.0,\n",
       " 9.5,\n",
       " 2.0,\n",
       " 14.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 5.5,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 5.0,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 33.5,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 8.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 3.0,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 4.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 7.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 5.5,\n",
       " 2.0,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 6.0,\n",
       " 2.5,\n",
       " 3.0,\n",
       " 5.0,\n",
       " 4.5,\n",
       " 3.5,\n",
       " 16.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 3.0,\n",
       " 3.0,\n",
       " 27.0,\n",
       " 20.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 5.0,\n",
       " 4.0,\n",
       " 35.0,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 1.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 5.0,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 4.0,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 4.0,\n",
       " 36.0,\n",
       " 5.5,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 9.0,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 3.0,\n",
       " 3.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 28.5,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 8.5,\n",
       " 2.0,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 7.0,\n",
       " 3.0,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 5.0,\n",
       " 3.5,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 13.0,\n",
       " 2.0,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 5.0,\n",
       " 1.0,\n",
       " 12.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 7.0,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 24.0,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 5.5,\n",
       " 7.0,\n",
       " 4.0,\n",
       " 2.0,\n",
       " 3.5,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.0,\n",
       " 4.0,\n",
       " 5.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 3.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 5.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 6.0,\n",
       " 25.0,\n",
       " 3.0,\n",
       " 8.5,\n",
       " 3.0,\n",
       " 6.5,\n",
       " 4.5,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 9.0,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 8.0,\n",
       " 3.0,\n",
       " 2.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 9.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 25.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 1.5,\n",
       " 5.0,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 26.5,\n",
       " 3.0,\n",
       " 1.5,\n",
       " 19.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 11.0,\n",
       " 11.0,\n",
       " 5.0,\n",
       " 2.0,\n",
       " 18.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 18.0,\n",
       " 2.0,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 5.5,\n",
       " 13.5,\n",
       " 5.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 22.5,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 28.5,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 5.5,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 3.5,\n",
       " 22.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 10.0,\n",
       " 6.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 11.5,\n",
       " 24.0,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 11.0,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 1.5,\n",
       " 50.0,\n",
       " 2.0,\n",
       " 26.5,\n",
       " 9.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 5.5,\n",
       " 23.0,\n",
       " 19.0,\n",
       " 18.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 19.0,\n",
       " 22.5,\n",
       " 8.5,\n",
       " 7.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 7.0,\n",
       " 2.0,\n",
       " 3.5,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 9.0,\n",
       " 2.5,\n",
       " 23.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 4.0,\n",
       " 2.0,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 22.5,\n",
       " 1.0,\n",
       " 4.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 1.0,\n",
       " 8.0,\n",
       " 3.5,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 3.0,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 19.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 5.5,\n",
       " 13.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 5.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 6.0,\n",
       " 8.5,\n",
       " 4.0,\n",
       " 1.5,\n",
       " 11.0,\n",
       " 2.0,\n",
       " 5.0,\n",
       " 1.0,\n",
       " 16.0,\n",
       " 28.5,\n",
       " 1.5,\n",
       " 2.0,\n",
       " 9.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 2.0,\n",
       " 2.0,\n",
       " 3.0,\n",
       " 2.0,\n",
       " 20.0,\n",
       " 11.5,\n",
       " 2.0,\n",
       " 1.0,\n",
       " 19.5,\n",
       " 13.5,\n",
       " 3.0,\n",
       " 1.0,\n",
       " 1.0,\n",
       " 2.0]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rank_results_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  1.33067640e-02,   1.36164144e-01,   1.80685624e-01,\n",
       "         9.10717905e-01,   8.17174137e-01,   3.12454045e-01,\n",
       "         6.08398067e-03,   7.83817768e-02,   1.12365298e-01,\n",
       "         9.41847190e-02,   3.72573376e-01,   1.36063546e-02,\n",
       "         2.49640364e-02,   1.54106319e-01,   2.70841178e-02,\n",
       "         9.61761847e-02,   1.51333762e-02,   1.53069109e-01,\n",
       "         3.59074116e-01,   1.39547001e-05,   2.36534372e-01,\n",
       "         5.07331491e-01,   6.76578507e-02,   1.71287954e-01,\n",
       "         1.05969235e-01,   4.55066562e-02,   2.98235361e-02,\n",
       "         1.85133025e-01,   8.62451047e-02,   2.83597976e-01,\n",
       "         1.18034862e-01,   8.68205801e-02,   1.12397159e-02,\n",
       "         8.19843650e-01,   4.11878794e-01,   1.53966561e-01,\n",
       "         3.90874296e-02,   6.25597984e-02,   1.91835806e-01,\n",
       "         5.10083027e-02,   1.65459178e-02,   1.30264908e-01,\n",
       "         9.68440901e-03,   1.89534239e-02,   5.07331491e-01,\n",
       "         2.70841178e-02,   2.53850430e-01,   3.66447657e-01,\n",
       "         2.47853007e-02,   3.78116071e-01,   2.35142522e-02,\n",
       "         7.92381689e-02,   4.95255828e-01,   8.17174137e-01,\n",
       "         1.65238366e-01,   2.24147812e-01,   1.44146815e-01,\n",
       "         8.28554481e-02,   4.76554334e-02,   1.35574684e-01,\n",
       "         1.51333762e-02,   7.55946115e-02,   3.16736192e-01,\n",
       "         3.16519678e-01,   4.04103771e-02,   4.74321067e-01,\n",
       "         3.63387436e-01,   9.10717905e-01,   3.38964984e-02,\n",
       "         1.46031966e-02,   8.04409832e-02,   4.11878794e-01,\n",
       "         1.37804106e-01,   3.59074116e-01,   1.42120466e-01,\n",
       "         1.91916645e-01,   3.45162377e-02,   2.51785129e-01,\n",
       "         6.13241315e-01,   1.67565927e-01,   8.20322707e-02,\n",
       "         2.85765022e-01,   3.50707144e-01,   2.06700340e-02,\n",
       "         1.32855028e-01,   2.92599071e-02,   1.67565927e-01,\n",
       "         8.20322707e-02,   3.72573376e-01,   2.43611813e-01,\n",
       "         7.06145763e-02,   1.32855028e-01,   2.45948993e-02,\n",
       "         7.86592737e-02,   1.51659757e-01,   5.08663207e-02,\n",
       "         2.32435316e-01,   1.61938176e-01,   1.04200147e-01,\n",
       "         3.76007445e-02,   2.79119521e-01,   3.71318460e-01,\n",
       "         3.78710106e-02,   8.61786425e-01,   5.64610124e-01,\n",
       "         4.19189073e-02,   3.72325741e-02,   8.88955370e-02,\n",
       "         1.79306194e-02,   5.94938695e-02,   3.14253509e-01,\n",
       "         1.78918511e-01,   5.20591140e-02,   8.34955350e-02,\n",
       "         1.00474618e-02,   2.91258007e-01,   1.93323288e-02,\n",
       "         1.44721091e-01,   3.55714411e-02,   4.84719127e-01,\n",
       "         8.76656100e-02,   2.13387266e-01,   1.29115015e-01,\n",
       "         3.90438028e-02,   2.29820147e-01,   5.14060222e-02,\n",
       "         5.94938695e-02,   3.02108210e-02,   5.46349466e-01,\n",
       "         5.08663207e-02,   3.61936063e-01,   2.99747795e-01,\n",
       "         1.16886422e-01,   5.29239476e-01,   4.86395061e-01,\n",
       "         1.47493454e-02,   2.19151173e-02,   1.63463444e-01,\n",
       "         5.45612909e-02,   1.89673863e-02,   2.30443597e-01,\n",
       "         6.48980856e-01,   2.72996753e-01,   1.53471202e-01,\n",
       "         3.30055118e-01,   1.16519421e-01,   1.29352175e-02,\n",
       "         9.50133279e-02,   5.05799055e-01,   1.43396389e-02,\n",
       "         1.52133610e-02,   1.16886422e-01,   5.08341677e-02,\n",
       "         5.20591140e-02,   4.37059738e-02,   8.61567199e-01,\n",
       "         2.08061323e-01,   2.99747795e-01,   2.13387266e-01,\n",
       "         5.91283143e-02,   1.28390908e-01,   3.32497284e-02,\n",
       "         1.59225598e-01,   1.75108053e-02,   2.98446491e-02,\n",
       "         1.56627715e-01,   5.96383438e-02,   1.22605637e-01,\n",
       "         1.03698611e-01,   6.64604679e-02,   3.22243959e-01,\n",
       "         1.11933284e-01,   3.32087614e-02,   1.53471202e-01,\n",
       "         3.32087614e-02,   1.39547001e-05,   2.91258007e-01,\n",
       "         3.11130077e-01,   4.26968448e-02,   1.53471202e-01,\n",
       "         1.26299083e-01,   2.10272539e-02,   2.21886337e-02,\n",
       "         5.37929349e-02,   1.42899267e-02,   2.70896330e-02,\n",
       "         9.53225270e-02,   4.26761001e-01,   1.29115015e-01,\n",
       "         4.46532890e-02,   2.29820147e-01,   2.32435316e-01,\n",
       "         2.72996753e-01,   1.18660159e-01,   3.16073000e-01,\n",
       "         5.63150942e-01,   3.13751213e-02,   6.00217521e-01,\n",
       "         2.14544423e-02,   2.55011171e-02,   1.52491912e-01,\n",
       "         2.37545252e-01,   2.28589866e-02,   1.01587266e-01,\n",
       "         6.11946344e-01,   2.14544423e-02,   4.33863290e-02,\n",
       "         3.45252186e-01,   4.42508250e-01,   2.80413460e-02,\n",
       "         1.83889046e-01,   2.02939715e-02,   2.18496874e-01,\n",
       "         4.61910486e-01,   1.31570145e-01,   1.50290597e-02,\n",
       "         6.35742366e-01,   1.38982954e-02,   4.37762737e-02,\n",
       "         1.54326305e-01,   9.80024338e-02,   1.39547001e-05,\n",
       "         4.67586666e-01,   1.23020545e-01,   4.60606292e-02,\n",
       "         4.45810169e-01,   2.21932475e-02,   4.84737344e-02,\n",
       "         4.81367782e-02,   8.61436054e-02,   2.53406744e-02,\n",
       "         3.27203721e-02,   4.48820405e-02,   1.99303970e-01,\n",
       "         8.31619278e-03,   4.40305442e-01,   9.48829874e-02,\n",
       "         2.40648314e-02,   2.83663005e-01,   5.55409193e-02,\n",
       "         3.82564962e-01,   1.03748187e-01,   5.58828041e-02,\n",
       "         2.32013896e-01,   1.80504650e-01,   4.33863290e-02,\n",
       "         2.90061273e-02,   1.02398649e-01,   3.64734195e-02,\n",
       "         8.41269270e-03,   2.38991827e-01,   4.22725469e-01,\n",
       "         2.53406744e-02,   2.90194023e-02,   6.11946344e-01,\n",
       "         4.45810169e-01,   5.39679639e-02,   1.70745134e-01,\n",
       "         4.01223540e-01,   1.01574741e-01,   4.47860286e-02,\n",
       "         4.74046409e-01,   6.85488135e-02,   7.55339861e-01,\n",
       "         4.45409685e-01,   5.15851378e-02,   6.62156165e-01,\n",
       "         2.36309431e-02,   4.77415994e-02,   3.45127992e-02,\n",
       "         2.98964411e-01,   2.45999731e-02,   8.34390149e-03,\n",
       "         2.48583406e-01,   2.55040340e-02,   5.58126308e-02,\n",
       "         8.35381672e-02,   5.64348936e-01,   2.46701073e-02,\n",
       "         2.21801288e-02,   6.89513087e-01,   7.04379082e-02,\n",
       "         7.04162121e-02,   6.98930547e-02,   2.14142472e-01,\n",
       "         2.25824296e-01,   6.45798445e-02,   3.45063239e-01,\n",
       "         9.53851733e-03,   2.47154813e-02,   1.50749590e-02,\n",
       "         4.33863290e-02,   9.48829874e-02,   6.08398952e-02,\n",
       "         2.08566617e-02,   8.84151384e-02,   5.25150374e-02,\n",
       "         2.51927912e-01,   1.86803997e-01,   4.42508250e-01,\n",
       "         3.35686713e-01,   2.28358641e-01,   1.72398593e-02,\n",
       "         6.89513087e-01,   2.48583406e-01,   5.46064898e-02,\n",
       "         1.92317873e-01,   3.44989523e-02,   4.33863290e-02,\n",
       "         2.65148096e-02,   1.63958445e-02,   2.04458535e-01,\n",
       "         1.35331471e-02,   1.35331471e-02,   1.80907711e-01,\n",
       "         1.03824949e-02,   5.75189888e-02,   2.14142472e-01,\n",
       "         3.82952988e-01,   3.81068587e-01,   3.09436888e-01,\n",
       "         3.45063239e-01,   1.88747749e-01,   3.43413919e-01,\n",
       "         1.43708214e-01,   2.40648314e-02,   5.91800734e-02,\n",
       "         4.81096841e-03,   6.35742366e-01,   3.40637386e-01,\n",
       "         3.91304046e-01,   4.45409685e-01,   1.27063006e-01,\n",
       "         3.82952988e-01,   1.37360319e-01,   8.09926167e-03,\n",
       "         1.14228331e-01,   1.24455392e-01,   1.14857756e-01,\n",
       "         2.60813832e-02,   3.30939263e-01,   1.53458398e-02,\n",
       "         4.12472561e-02,   7.55339861e-01,   1.37360319e-01,\n",
       "         4.45409685e-01,   7.49541447e-02,   3.86336237e-01,\n",
       "         5.16899586e-01,   8.35275277e-02,   2.04458535e-01,\n",
       "         2.55040340e-02,   5.74805308e-03,   8.41269270e-03,\n",
       "         3.00378017e-02,   9.51326732e-03,   1.43214762e-02,\n",
       "         3.24261636e-02,   4.26761001e-01,   1.43708214e-01,\n",
       "         3.20336148e-02,   6.65882081e-02,   2.75211602e-01,\n",
       "         1.70739330e-02,   8.60310644e-02,   3.99239957e-02,\n",
       "         2.00033616e-02,   3.60740609e-02,   2.98782647e-01,\n",
       "         7.88226947e-02,   9.32765827e-02,   2.24970013e-01,\n",
       "         7.04162121e-02,   4.00608748e-01,   6.65882081e-02,\n",
       "         2.09757686e-01,   6.98930547e-02,   2.52346415e-02,\n",
       "         1.89962108e-02,   9.32765827e-02,   3.18329543e-01,\n",
       "         4.24634755e-01,   2.36008707e-02,   3.46442610e-02,\n",
       "         8.26820266e-03,   7.38286227e-02,   4.38030735e-02,\n",
       "         4.29709442e-02,   7.81519413e-02,   4.81541567e-02,\n",
       "         4.24366295e-01,   6.37350619e-01,   7.38286227e-02,\n",
       "         2.09831342e-01,   3.73091698e-01,   3.01777363e-01,\n",
       "         6.89737380e-01,   8.69060531e-02,   2.74680495e-01,\n",
       "         1.26340255e-01,   2.17564285e-01,   1.14617057e-01,\n",
       "         2.21043438e-01,   3.75092290e-02,   1.80173010e-01,\n",
       "         1.92705229e-01,   5.20511940e-02,   3.45575325e-02,\n",
       "         4.24366295e-01,   1.38701469e-01,   2.19959930e-01,\n",
       "         3.30093168e-02,   1.51094526e-01,   6.29530624e-02,\n",
       "         1.06256045e-01,   5.16319573e-02,   4.19359565e-01,\n",
       "         1.67035982e-02,   4.66737121e-01,   5.28554805e-02,\n",
       "         3.34028631e-01,   2.07557589e-01,   3.81171480e-02,\n",
       "         1.39750421e-01,   5.43124378e-01,   1.56172235e-02,\n",
       "         3.14330123e-02,   2.37817749e-01,   5.24508841e-02,\n",
       "         1.68979809e-01,   5.20511940e-02,   1.41864970e-01,\n",
       "         2.34760102e-02,   3.26613307e-01,   6.49467483e-02,\n",
       "         1.56172235e-02,   2.52470933e-02,   4.63026613e-01,\n",
       "         6.24662563e-02,   1.28468588e-01,   1.94668807e-02,\n",
       "         1.57593966e-01,   5.74055798e-02,   2.28380151e-02,\n",
       "         1.96443126e-02,   1.40411809e-01,   5.67799918e-02,\n",
       "         1.66321665e-01,   1.31319454e-02,   2.22754285e-01,\n",
       "         8.87866467e-02,   5.69252372e-01,   1.21373825e-01,\n",
       "         3.36346656e-01,   7.90603086e-03,   2.14330405e-02,\n",
       "         1.46280490e-02,   4.41526204e-01,   6.29530624e-02,\n",
       "         1.92694049e-02,   6.23336993e-02,   2.91050375e-01,\n",
       "         7.39210248e-02,   4.73873429e-02,   2.85166632e-02,\n",
       "         4.32584971e-01,   4.69336182e-01,   1.52809113e-01,\n",
       "         2.12605655e-01,   6.82980299e-01,   6.23686492e-01,\n",
       "         1.46280490e-02,   1.43092781e-01,   1.92705229e-01,\n",
       "         1.76649183e-01,   2.04448216e-02,   3.26613307e-01,\n",
       "         2.81566590e-01,   1.28468588e-01,   1.12938456e-01,\n",
       "         2.70400755e-02,   1.67469345e-02,   8.69060531e-02,\n",
       "         2.64987051e-01,   5.08955359e-01,   2.36527279e-01,\n",
       "         8.85304511e-02,   2.52836868e-02,   1.23924702e-01,\n",
       "         8.14372767e-03,   4.86891240e-01,   2.81176977e-02,\n",
       "         2.86851615e-01,   3.33606005e-02,   1.14263497e-01,\n",
       "         2.62193531e-01,   2.10091680e-01,   5.21427244e-02,\n",
       "         4.76650447e-01,   5.01338691e-02,   7.11131245e-02,\n",
       "         3.17319840e-01,   3.14421743e-01,   2.37791434e-01,\n",
       "         3.74990135e-01,   5.49132049e-01,   1.93443727e-02,\n",
       "         4.19359565e-01,   1.88574027e-02,   3.00265431e-01,\n",
       "         4.97581847e-02,   1.25150830e-01,   7.38863945e-01,\n",
       "         1.46280490e-02,   1.95861742e-01,   3.33250135e-01,\n",
       "         2.36527279e-01,   2.16741804e-02,   3.06864120e-02,\n",
       "         2.00825855e-02,   1.06310785e-01,   7.01278746e-02,\n",
       "         1.14263497e-01,   3.25931430e-01,   4.02316034e-01,\n",
       "         2.06467763e-01,   3.26408684e-01,   5.54605067e-01,\n",
       "         1.53695688e-01,   1.39547001e-05,   1.53998375e-01,\n",
       "         4.74777281e-01,   1.87118843e-01,   2.50030681e-02,\n",
       "         2.81738609e-01,   1.09481262e-02,   4.39131521e-02,\n",
       "         1.86708681e-02,   3.39364380e-01,   7.80244112e-01,\n",
       "         1.92589890e-02,   1.28220767e-01,   1.21344656e-01,\n",
       "         1.76075980e-01,   1.61237419e-02,   2.00805873e-01,\n",
       "         2.88069956e-02,   4.76650447e-01,   3.48606825e-01,\n",
       "         2.79677540e-01,   3.34936306e-02,   1.35285437e-01,\n",
       "         3.11592698e-01,   1.87118843e-01,   4.21923697e-02,\n",
       "         2.06467763e-01,   1.46280490e-02,   2.69783497e-01,\n",
       "         2.45371282e-01,   6.91154599e-02,   1.21665284e-01,\n",
       "         6.32546246e-01,   6.37350619e-01,   7.64984787e-02,\n",
       "         8.03843439e-01,   2.91174538e-02,   3.10906824e-02,\n",
       "         2.46235449e-02,   1.00323334e-02,   2.52739601e-02,\n",
       "         2.20022216e-01,   4.73873429e-02,   2.69112915e-01,\n",
       "         6.07336033e-03,   2.34655514e-02,   4.69191223e-02,\n",
       "         1.65409390e-02,   1.87221076e-02,   1.43851653e-01,\n",
       "         4.86891240e-01,   6.32268131e-01,   5.21427244e-02,\n",
       "         2.74680495e-01,   2.45046038e-02,   7.17880297e-03,\n",
       "         4.79486398e-02,   5.43124378e-01,   3.04933876e-01,\n",
       "         6.89737380e-01,   1.04716748e-01,   2.16274142e-01,\n",
       "         2.02939715e-02,   5.26177660e-02,   2.47855242e-02,\n",
       "         5.55409193e-02,   3.95119727e-01,   1.40128121e-01,\n",
       "         3.52701247e-01,   8.03843439e-01,   3.38964984e-02,\n",
       "         3.70321795e-02,   6.88214526e-02,   2.39515584e-02,\n",
       "         7.56967545e-01,   1.93443727e-02,   1.38982954e-02,\n",
       "         7.28118271e-02,   9.94839668e-02,   5.24508841e-02,\n",
       "         4.41526204e-01,   3.42311822e-02,   8.19607601e-02,\n",
       "         1.93709880e-02,   9.90658719e-03,   5.97671643e-02,\n",
       "         5.25432006e-02,   4.88414951e-02,   7.29382783e-02,\n",
       "         7.27178827e-02,   5.44884026e-01,   1.51094526e-01,\n",
       "         8.56077895e-02,   2.59132773e-01,   7.38863945e-01,\n",
       "         2.27108859e-02,   1.46173742e-02,   4.69336182e-01,\n",
       "         5.35952002e-02,   1.57796606e-01,   7.98916742e-02,\n",
       "         7.57778957e-02,   2.34924778e-02,   5.61721146e-01,\n",
       "         5.81136309e-02,   1.17938064e-01,   3.73091698e-01,\n",
       "         4.69653532e-02,   2.82926351e-01,   3.74377444e-02,\n",
       "         7.80244112e-01,   1.07040536e-02,   4.22343574e-02,\n",
       "         2.48383641e-01,   6.68739974e-02,   5.79607546e-01,\n",
       "         6.12842083e-01,   2.24699363e-01,   1.17287099e-01,\n",
       "         2.11210489e-01,   1.26500487e-01,   4.42833573e-01,\n",
       "         9.86616313e-02,   7.48984993e-01,   7.17300177e-02,\n",
       "         3.41537148e-01,   2.12066192e-02,   4.58810478e-02,\n",
       "         6.14493825e-02,   3.36158544e-01,   5.35191111e-02,\n",
       "         6.14493825e-02,   3.37975055e-01,   2.48258356e-02,\n",
       "         1.95439026e-01,   2.10863743e-02,   3.45127992e-02,\n",
       "         3.77624482e-01,   4.64020044e-01,   4.81707901e-01,\n",
       "         1.39547001e-05,   6.50388896e-01,   4.02696162e-01,\n",
       "         8.22416693e-03,   5.16319573e-02,   3.48753035e-02,\n",
       "         2.95950770e-01,   8.32730755e-02,   2.60120295e-02,\n",
       "         5.97671643e-02,   4.61252153e-01,   6.30505010e-03,\n",
       "         1.06310785e-01,   2.98366874e-01,   5.61721146e-01,\n",
       "         5.04955724e-02,   7.44576991e-01,   4.45074081e-01,\n",
       "         3.44138294e-01,   5.42857111e-01,   1.54274628e-02,\n",
       "         2.80294549e-02,   1.46196559e-01,   5.01431115e-02,\n",
       "         1.30186724e-02,   1.58574115e-02,   2.78033197e-01,\n",
       "         1.36267813e-02,   3.04012686e-01,   4.29279625e-01,\n",
       "         4.61252153e-01,   3.09900697e-02,   1.87167838e-01,\n",
       "         3.18929590e-02,   7.64984787e-02,   1.54501855e-01,\n",
       "         4.15992230e-01,   9.14104879e-02,   8.87110010e-02,\n",
       "         4.25476059e-02,   2.38339990e-01,   4.64020044e-01,\n",
       "         1.97868332e-01,   4.02316034e-01,   5.52808702e-01,\n",
       "         2.19283670e-01,   3.36158544e-01,   7.35383511e-01,\n",
       "         6.19484521e-02,   2.34742701e-01,   1.18169963e-01,\n",
       "         3.90584134e-02,   2.26270799e-02,   5.46638444e-02,\n",
       "         4.58099246e-01,   1.21344656e-01,   4.90670949e-02,\n",
       "         3.00317913e-01,   2.96702255e-02,   4.01023813e-02,\n",
       "         5.46610713e-01,   3.49584259e-02,   1.15435608e-02,\n",
       "         1.01916566e-01,   3.06994226e-02,   4.81541567e-02,\n",
       "         2.56203059e-02,   1.38116807e-01,   2.90646851e-01,\n",
       "         1.61525741e-01,   2.17885241e-01,   2.38445215e-02,\n",
       "         3.74377444e-02,   1.04421787e-02,   1.79890078e-02,\n",
       "         2.14627206e-01,   2.75171511e-02,   3.66925210e-01,\n",
       "         5.81136309e-02,   8.66166800e-02,   2.46146590e-01,\n",
       "         2.82926351e-01,   1.28220767e-01,   6.65816247e-01,\n",
       "         6.26297653e-01,   2.13723511e-01,   4.29843932e-01,\n",
       "         8.37058574e-02,   7.44576991e-01,   2.10342649e-02,\n",
       "         7.67944977e-02,   8.41390863e-02,   7.48984993e-01,\n",
       "         2.55992524e-02,   3.80988754e-02,   5.63641591e-03,\n",
       "         1.07420221e-01,   4.45074081e-01,   1.37681410e-01,\n",
       "         1.94361992e-02,   2.22563408e-02,   6.52900279e-01,\n",
       "         1.72929779e-01,   1.53460711e-01,   3.63146290e-02,\n",
       "         1.53460711e-01,   2.13723511e-01,   1.32793933e-02,\n",
       "         6.91154599e-02,   4.58099246e-01,   1.61525741e-01,\n",
       "         4.38481450e-01,   1.26662299e-01,   2.72810966e-01,\n",
       "         6.14493825e-02,   7.35383511e-01,   3.10346991e-01,\n",
       "         3.85159664e-02,   3.51537883e-01,   2.01694414e-01,\n",
       "         5.47058523e-01,   2.26270799e-02,   3.39002013e-02,\n",
       "         5.35191111e-02,   9.86616313e-02,   1.85574383e-01,\n",
       "         5.21303773e-01,   4.18155752e-02,   6.14493825e-02,\n",
       "         1.07420221e-01,   3.74457315e-02,   4.03312862e-01,\n",
       "         9.24118310e-02,   2.11210489e-01,   3.03336699e-02,\n",
       "         6.58556893e-02,   4.69633400e-01,   3.54543656e-01,\n",
       "         4.42796759e-02,   1.34777680e-01,   4.57760453e-01,\n",
       "         7.54193723e-01,   6.05005361e-02,   6.44062739e-03,\n",
       "         1.96432918e-01,   6.44062739e-03,   1.48852840e-02,\n",
       "         1.96437016e-01,   3.79365534e-01,   9.18027312e-02,\n",
       "         6.75362110e-01,   8.51453990e-02,   2.89419770e-01,\n",
       "         7.01488256e-02,   2.39074603e-01,   4.42115627e-02,\n",
       "         1.48621321e-01,   4.88330144e-03,   5.71256764e-02,\n",
       "         6.93117678e-02,   1.94905289e-02,   4.09509659e-01,\n",
       "         4.88330144e-03,   3.76139492e-01,   8.31025019e-02,\n",
       "         2.80105025e-02,   1.38886832e-02,   2.43484393e-01,\n",
       "         4.06922027e-02,   1.53550133e-01,   1.69814050e-01,\n",
       "         7.58789703e-02,   2.26687133e-01,   1.13889068e-01,\n",
       "         3.38641465e-01,   4.85991806e-01,   5.25802821e-02,\n",
       "         7.93269929e-03,   2.85583492e-02,   2.84822315e-01,\n",
       "         1.25731543e-01,   4.57568057e-02,   1.79471280e-02,\n",
       "         8.12529922e-02,   2.47343570e-01,   3.19625065e-02,\n",
       "         2.29310274e-01,   3.67629379e-02,   1.07830437e-02,\n",
       "         1.64742276e-01,   2.55871397e-02,   7.93269929e-03,\n",
       "         6.08784929e-02,   4.72857982e-01,   1.49150537e-02,\n",
       "         4.66332864e-03,   6.92562684e-02,   1.82399467e-01,\n",
       "         2.26411708e-02,   5.88819012e-02,   1.34836202e-02,\n",
       "         2.21691225e-02,   5.15873395e-02,   4.36566100e-02,\n",
       "         1.47917956e-01,   7.75274858e-02,   9.96802654e-03,\n",
       "         6.04487881e-02,   9.56529751e-02,   2.12116353e-02,\n",
       "         1.29352175e-02,   3.08048785e-01,   4.95008752e-02,\n",
       "         4.80929464e-01,   2.31812358e-01,   7.85021856e-02,\n",
       "         4.57651541e-02,   8.26800168e-02,   6.33903369e-02,\n",
       "         1.48852840e-02,   1.27479047e-01,   9.94726047e-02,\n",
       "         1.96437016e-01,   1.35652125e-01,   8.22845846e-03,\n",
       "         5.36409020e-02,   2.12116353e-02,   3.58536214e-01,\n",
       "         1.51468173e-01,   1.13940939e-01,   5.12129590e-02,\n",
       "         6.52626827e-02,   5.36409020e-02,   2.07965598e-02,\n",
       "         4.87578176e-02,   2.26687133e-01,   2.02395409e-01,\n",
       "         2.87369013e-01,   5.36409020e-02,   5.17398715e-01,\n",
       "         1.61767099e-02,   3.16821247e-01,   2.85118092e-02,\n",
       "         1.07214712e-01,   1.51333762e-02,   6.88211769e-02,\n",
       "         3.39554250e-01,   8.31025019e-02,   1.90173462e-01,\n",
       "         9.96802654e-03,   3.36164013e-02,   2.69759987e-02,\n",
       "         3.47459726e-02,   3.13223302e-02,   1.03547238e-02,\n",
       "         2.55871397e-02,   2.15519685e-02,   4.14633267e-02,\n",
       "         1.66640297e-01,   4.41029035e-02,   2.43484393e-01,\n",
       "         4.02841926e-01,   8.51453990e-02,   9.18027312e-02,\n",
       "         6.92562684e-02,   1.82399467e-01,   1.66930407e-02,\n",
       "         2.92131782e-01,   3.64172272e-02,   1.52828172e-01,\n",
       "         1.94905289e-02,   1.67512074e-02,   5.87521970e-01,\n",
       "         1.07214712e-01,   5.94177619e-02,   5.41316345e-02,\n",
       "         1.59996971e-01,   4.41029035e-02,   6.12157345e-01,\n",
       "         1.72911733e-01,   2.97773760e-02,   2.69759987e-02,\n",
       "         1.20841088e-02,   1.75840676e-01,   1.64742276e-01,\n",
       "         3.19069833e-01,   1.15767391e-02,   9.26958304e-03,\n",
       "         1.05873216e-02,   4.48393047e-01,   2.07965598e-02,\n",
       "         8.51453990e-02,   1.10927567e-01,   3.86442468e-02,\n",
       "         3.76139492e-01,   5.70046827e-02,   7.19265938e-02,\n",
       "         1.34836202e-02,   1.32472785e-02,   3.06702908e-02,\n",
       "         4.63338792e-01,   2.02395409e-01,   2.10731402e-02,\n",
       "         1.46031966e-02,   2.78479606e-02,   2.69759987e-02,\n",
       "         2.47711577e-02,   3.73352435e-03,   8.60471874e-02,\n",
       "         2.29310274e-01,   3.86442468e-02,   5.15079610e-02,\n",
       "         1.16099283e-01,   1.06176706e-02,   9.94726047e-02,\n",
       "         2.26139054e-02,   4.88330144e-03,   1.34836202e-02,\n",
       "         2.07965598e-02,   2.55871397e-02,   4.29463863e-01,\n",
       "         4.95008752e-02,   3.95649225e-01,   2.87510276e-01,\n",
       "         6.75362110e-01,   4.06922027e-02,   1.34381056e-01,\n",
       "         8.19308460e-02,   1.94055419e-02,   9.94726047e-02], dtype=float32)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sim_results_train[3]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.910718\n",
      "[  1.33067640e-02   1.36164144e-01   1.80685624e-01   9.10717905e-01\n",
      "   8.17174137e-01   3.12454045e-01   6.08398067e-03   7.83817768e-02\n",
      "   1.12365298e-01   9.41847190e-02   3.72573376e-01   1.36063546e-02\n",
      "   2.49640364e-02   1.54106319e-01   2.70841178e-02   9.61761847e-02\n",
      "   1.51333762e-02   1.53069109e-01   3.59074116e-01   1.39547001e-05\n",
      "   2.36534372e-01   5.07331491e-01   6.76578507e-02   1.71287954e-01\n",
      "   1.05969235e-01   4.55066562e-02   2.98235361e-02   1.85133025e-01\n",
      "   8.62451047e-02   2.83597976e-01   1.18034862e-01   8.68205801e-02\n",
      "   1.12397159e-02   8.19843650e-01   4.11878794e-01   1.53966561e-01\n",
      "   3.90874296e-02   6.25597984e-02   1.91835806e-01   5.10083027e-02\n",
      "   1.65459178e-02   1.30264908e-01   9.68440901e-03   1.89534239e-02\n",
      "   5.07331491e-01   2.70841178e-02   2.53850430e-01   3.66447657e-01\n",
      "   2.47853007e-02   3.78116071e-01   2.35142522e-02   7.92381689e-02\n",
      "   4.95255828e-01   8.17174137e-01   1.65238366e-01   2.24147812e-01\n",
      "   1.44146815e-01   8.28554481e-02   4.76554334e-02   1.35574684e-01\n",
      "   1.51333762e-02   7.55946115e-02   3.16736192e-01   3.16519678e-01\n",
      "   4.04103771e-02   4.74321067e-01   3.63387436e-01   9.10717905e-01\n",
      "   3.38964984e-02   1.46031966e-02   8.04409832e-02   4.11878794e-01\n",
      "   1.37804106e-01   3.59074116e-01   1.42120466e-01   1.91916645e-01\n",
      "   3.45162377e-02   2.51785129e-01   6.13241315e-01   1.67565927e-01\n",
      "   8.20322707e-02   2.85765022e-01   3.50707144e-01   2.06700340e-02\n",
      "   1.32855028e-01   2.92599071e-02   1.67565927e-01   8.20322707e-02\n",
      "   3.72573376e-01   2.43611813e-01   7.06145763e-02   1.32855028e-01\n",
      "   2.45948993e-02   7.86592737e-02   1.51659757e-01   5.08663207e-02\n",
      "   2.32435316e-01   1.61938176e-01   1.04200147e-01   3.76007445e-02\n",
      "   2.79119521e-01   3.71318460e-01   3.78710106e-02   8.61786425e-01\n",
      "   5.64610124e-01   4.19189073e-02   3.72325741e-02   8.88955370e-02\n",
      "   1.79306194e-02   5.94938695e-02   3.14253509e-01   1.78918511e-01\n",
      "   5.20591140e-02   8.34955350e-02   1.00474618e-02   2.91258007e-01\n",
      "   1.93323288e-02   1.44721091e-01   3.55714411e-02   4.84719127e-01\n",
      "   8.76656100e-02   2.13387266e-01   1.29115015e-01   3.90438028e-02\n",
      "   2.29820147e-01   5.14060222e-02   5.94938695e-02   3.02108210e-02\n",
      "   5.46349466e-01   5.08663207e-02   3.61936063e-01   2.99747795e-01\n",
      "   1.16886422e-01   5.29239476e-01   4.86395061e-01   1.47493454e-02\n",
      "   2.19151173e-02   1.63463444e-01   5.45612909e-02   1.89673863e-02\n",
      "   2.30443597e-01   6.48980856e-01   2.72996753e-01   1.53471202e-01\n",
      "   3.30055118e-01   1.16519421e-01   1.29352175e-02   9.50133279e-02\n",
      "   5.05799055e-01   1.43396389e-02   1.52133610e-02   1.16886422e-01\n",
      "   5.08341677e-02   5.20591140e-02   4.37059738e-02   8.61567199e-01\n",
      "   2.08061323e-01   2.99747795e-01   2.13387266e-01   5.91283143e-02\n",
      "   1.28390908e-01   3.32497284e-02   1.59225598e-01   1.75108053e-02\n",
      "   2.98446491e-02   1.56627715e-01   5.96383438e-02   1.22605637e-01\n",
      "   1.03698611e-01   6.64604679e-02   3.22243959e-01   1.11933284e-01\n",
      "   3.32087614e-02   1.53471202e-01   3.32087614e-02   1.39547001e-05\n",
      "   2.91258007e-01   3.11130077e-01   4.26968448e-02   1.53471202e-01\n",
      "   1.26299083e-01   2.10272539e-02   2.21886337e-02   5.37929349e-02\n",
      "   1.42899267e-02   2.70896330e-02   9.53225270e-02   4.26761001e-01\n",
      "   1.29115015e-01   4.46532890e-02   2.29820147e-01   2.32435316e-01\n",
      "   2.72996753e-01   1.18660159e-01   3.16073000e-01   5.63150942e-01\n",
      "   3.13751213e-02   6.00217521e-01   2.14544423e-02   2.55011171e-02\n",
      "   1.52491912e-01   2.37545252e-01   2.28589866e-02   1.01587266e-01\n",
      "   6.11946344e-01   2.14544423e-02   4.33863290e-02   3.45252186e-01\n",
      "   4.42508250e-01   2.80413460e-02   1.83889046e-01   2.02939715e-02\n",
      "   2.18496874e-01   4.61910486e-01   1.31570145e-01   1.50290597e-02\n",
      "   6.35742366e-01   1.38982954e-02   4.37762737e-02   1.54326305e-01\n",
      "   9.80024338e-02   1.39547001e-05   4.67586666e-01   1.23020545e-01\n",
      "   4.60606292e-02   4.45810169e-01   2.21932475e-02   4.84737344e-02\n",
      "   4.81367782e-02   8.61436054e-02   2.53406744e-02   3.27203721e-02\n",
      "   4.48820405e-02   1.99303970e-01   8.31619278e-03   4.40305442e-01\n",
      "   9.48829874e-02   2.40648314e-02   2.83663005e-01   5.55409193e-02\n",
      "   3.82564962e-01   1.03748187e-01   5.58828041e-02   2.32013896e-01\n",
      "   1.80504650e-01   4.33863290e-02   2.90061273e-02   1.02398649e-01\n",
      "   3.64734195e-02   8.41269270e-03   2.38991827e-01   4.22725469e-01\n",
      "   2.53406744e-02   2.90194023e-02   6.11946344e-01   4.45810169e-01\n",
      "   5.39679639e-02   1.70745134e-01   4.01223540e-01   1.01574741e-01\n",
      "   4.47860286e-02   4.74046409e-01   6.85488135e-02   7.55339861e-01\n",
      "   4.45409685e-01   5.15851378e-02   6.62156165e-01   2.36309431e-02\n",
      "   4.77415994e-02   3.45127992e-02   2.98964411e-01   2.45999731e-02\n",
      "   8.34390149e-03   2.48583406e-01   2.55040340e-02   5.58126308e-02\n",
      "   8.35381672e-02   5.64348936e-01   2.46701073e-02   2.21801288e-02\n",
      "   6.89513087e-01   7.04379082e-02   7.04162121e-02   6.98930547e-02\n",
      "   2.14142472e-01   2.25824296e-01   6.45798445e-02   3.45063239e-01\n",
      "   9.53851733e-03   2.47154813e-02   1.50749590e-02   4.33863290e-02\n",
      "   9.48829874e-02   6.08398952e-02   2.08566617e-02   8.84151384e-02\n",
      "   5.25150374e-02   2.51927912e-01   1.86803997e-01   4.42508250e-01\n",
      "   3.35686713e-01   2.28358641e-01   1.72398593e-02   6.89513087e-01\n",
      "   2.48583406e-01   5.46064898e-02   1.92317873e-01   3.44989523e-02\n",
      "   4.33863290e-02   2.65148096e-02   1.63958445e-02   2.04458535e-01\n",
      "   1.35331471e-02   1.35331471e-02   1.80907711e-01   1.03824949e-02\n",
      "   5.75189888e-02   2.14142472e-01   3.82952988e-01   3.81068587e-01\n",
      "   3.09436888e-01   3.45063239e-01   1.88747749e-01   3.43413919e-01\n",
      "   1.43708214e-01   2.40648314e-02   5.91800734e-02   4.81096841e-03\n",
      "   6.35742366e-01   3.40637386e-01   3.91304046e-01   4.45409685e-01\n",
      "   1.27063006e-01   3.82952988e-01   1.37360319e-01   8.09926167e-03\n",
      "   1.14228331e-01   1.24455392e-01   1.14857756e-01   2.60813832e-02\n",
      "   3.30939263e-01   1.53458398e-02   4.12472561e-02   7.55339861e-01\n",
      "   1.37360319e-01   4.45409685e-01   7.49541447e-02   3.86336237e-01\n",
      "   5.16899586e-01   8.35275277e-02   2.04458535e-01   2.55040340e-02\n",
      "   5.74805308e-03   8.41269270e-03   3.00378017e-02   9.51326732e-03\n",
      "   1.43214762e-02   3.24261636e-02   4.26761001e-01   1.43708214e-01\n",
      "   3.20336148e-02   6.65882081e-02   2.75211602e-01   1.70739330e-02\n",
      "   8.60310644e-02   3.99239957e-02   2.00033616e-02   3.60740609e-02\n",
      "   2.98782647e-01   7.88226947e-02   9.32765827e-02   2.24970013e-01\n",
      "   7.04162121e-02   4.00608748e-01   6.65882081e-02   2.09757686e-01\n",
      "   6.98930547e-02   2.52346415e-02   1.89962108e-02   9.32765827e-02\n",
      "   3.18329543e-01   4.24634755e-01   2.36008707e-02   3.46442610e-02\n",
      "   8.26820266e-03   7.38286227e-02   4.38030735e-02   4.29709442e-02\n",
      "   7.81519413e-02   4.81541567e-02   4.24366295e-01   6.37350619e-01\n",
      "   7.38286227e-02   2.09831342e-01   3.73091698e-01   3.01777363e-01\n",
      "   6.89737380e-01   8.69060531e-02   2.74680495e-01   1.26340255e-01\n",
      "   2.17564285e-01   1.14617057e-01   2.21043438e-01   3.75092290e-02\n",
      "   1.80173010e-01   1.92705229e-01   5.20511940e-02   3.45575325e-02\n",
      "   4.24366295e-01   1.38701469e-01   2.19959930e-01   3.30093168e-02\n",
      "   1.51094526e-01   6.29530624e-02   1.06256045e-01   5.16319573e-02\n",
      "   4.19359565e-01   1.67035982e-02   4.66737121e-01   5.28554805e-02\n",
      "   3.34028631e-01   2.07557589e-01   3.81171480e-02   1.39750421e-01\n",
      "   5.43124378e-01   1.56172235e-02   3.14330123e-02   2.37817749e-01\n",
      "   5.24508841e-02   1.68979809e-01   5.20511940e-02   1.41864970e-01\n",
      "   2.34760102e-02   3.26613307e-01   6.49467483e-02   1.56172235e-02\n",
      "   2.52470933e-02   4.63026613e-01   6.24662563e-02   1.28468588e-01\n",
      "   1.94668807e-02   1.57593966e-01   5.74055798e-02   2.28380151e-02\n",
      "   1.96443126e-02   1.40411809e-01   5.67799918e-02   1.66321665e-01\n",
      "   1.31319454e-02   2.22754285e-01   8.87866467e-02   5.69252372e-01\n",
      "   1.21373825e-01   3.36346656e-01   7.90603086e-03   2.14330405e-02\n",
      "   1.46280490e-02   4.41526204e-01   6.29530624e-02   1.92694049e-02\n",
      "   6.23336993e-02   2.91050375e-01   7.39210248e-02   4.73873429e-02\n",
      "   2.85166632e-02   4.32584971e-01   4.69336182e-01   1.52809113e-01\n",
      "   2.12605655e-01   6.82980299e-01   6.23686492e-01   1.46280490e-02\n",
      "   1.43092781e-01   1.92705229e-01   1.76649183e-01   2.04448216e-02\n",
      "   3.26613307e-01   2.81566590e-01   1.28468588e-01   1.12938456e-01\n",
      "   2.70400755e-02   1.67469345e-02   8.69060531e-02   2.64987051e-01\n",
      "   5.08955359e-01   2.36527279e-01   8.85304511e-02   2.52836868e-02\n",
      "   1.23924702e-01   8.14372767e-03   4.86891240e-01   2.81176977e-02\n",
      "   2.86851615e-01   3.33606005e-02   1.14263497e-01   2.62193531e-01\n",
      "   2.10091680e-01   5.21427244e-02   4.76650447e-01   5.01338691e-02\n",
      "   7.11131245e-02   3.17319840e-01   3.14421743e-01   2.37791434e-01\n",
      "   3.74990135e-01   5.49132049e-01   1.93443727e-02   4.19359565e-01\n",
      "   1.88574027e-02   3.00265431e-01   4.97581847e-02   1.25150830e-01\n",
      "   7.38863945e-01   1.46280490e-02   1.95861742e-01   3.33250135e-01\n",
      "   2.36527279e-01   2.16741804e-02   3.06864120e-02   2.00825855e-02\n",
      "   1.06310785e-01   7.01278746e-02   1.14263497e-01   3.25931430e-01\n",
      "   4.02316034e-01   2.06467763e-01   3.26408684e-01   5.54605067e-01\n",
      "   1.53695688e-01   1.39547001e-05   1.53998375e-01   4.74777281e-01\n",
      "   1.87118843e-01   2.50030681e-02   2.81738609e-01   1.09481262e-02\n",
      "   4.39131521e-02   1.86708681e-02   3.39364380e-01   7.80244112e-01\n",
      "   1.92589890e-02   1.28220767e-01   1.21344656e-01   1.76075980e-01\n",
      "   1.61237419e-02   2.00805873e-01   2.88069956e-02   4.76650447e-01\n",
      "   3.48606825e-01   2.79677540e-01   3.34936306e-02   1.35285437e-01\n",
      "   3.11592698e-01   1.87118843e-01   4.21923697e-02   2.06467763e-01\n",
      "   1.46280490e-02   2.69783497e-01   2.45371282e-01   6.91154599e-02\n",
      "   1.21665284e-01   6.32546246e-01   6.37350619e-01   7.64984787e-02\n",
      "   8.03843439e-01   2.91174538e-02   3.10906824e-02   2.46235449e-02\n",
      "   1.00323334e-02   2.52739601e-02   2.20022216e-01   4.73873429e-02\n",
      "   2.69112915e-01   6.07336033e-03   2.34655514e-02   4.69191223e-02\n",
      "   1.65409390e-02   1.87221076e-02   1.43851653e-01   4.86891240e-01\n",
      "   6.32268131e-01   5.21427244e-02   2.74680495e-01   2.45046038e-02\n",
      "   7.17880297e-03   4.79486398e-02   5.43124378e-01   3.04933876e-01\n",
      "   6.89737380e-01   1.04716748e-01   2.16274142e-01   2.02939715e-02\n",
      "   5.26177660e-02   2.47855242e-02   5.55409193e-02   3.95119727e-01\n",
      "   1.40128121e-01   3.52701247e-01   8.03843439e-01   3.38964984e-02\n",
      "   3.70321795e-02   6.88214526e-02   2.39515584e-02   7.56967545e-01\n",
      "   1.93443727e-02   1.38982954e-02   7.28118271e-02   9.94839668e-02\n",
      "   5.24508841e-02   4.41526204e-01   3.42311822e-02   8.19607601e-02\n",
      "   1.93709880e-02   9.90658719e-03   5.97671643e-02   5.25432006e-02\n",
      "   4.88414951e-02   7.29382783e-02   7.27178827e-02   5.44884026e-01\n",
      "   1.51094526e-01   8.56077895e-02   2.59132773e-01   7.38863945e-01\n",
      "   2.27108859e-02   1.46173742e-02   4.69336182e-01   5.35952002e-02\n",
      "   1.57796606e-01   7.98916742e-02   7.57778957e-02   2.34924778e-02\n",
      "   5.61721146e-01   5.81136309e-02   1.17938064e-01   3.73091698e-01\n",
      "   4.69653532e-02   2.82926351e-01   3.74377444e-02   7.80244112e-01\n",
      "   1.07040536e-02   4.22343574e-02   2.48383641e-01   6.68739974e-02\n",
      "   5.79607546e-01   6.12842083e-01   2.24699363e-01   1.17287099e-01\n",
      "   2.11210489e-01   1.26500487e-01   4.42833573e-01   9.86616313e-02\n",
      "   7.48984993e-01   7.17300177e-02   3.41537148e-01   2.12066192e-02\n",
      "   4.58810478e-02   6.14493825e-02   3.36158544e-01   5.35191111e-02\n",
      "   6.14493825e-02   3.37975055e-01   2.48258356e-02   1.95439026e-01\n",
      "   2.10863743e-02   3.45127992e-02   3.77624482e-01   4.64020044e-01\n",
      "   4.81707901e-01   1.39547001e-05   6.50388896e-01   4.02696162e-01\n",
      "   8.22416693e-03   5.16319573e-02   3.48753035e-02   2.95950770e-01\n",
      "   8.32730755e-02   2.60120295e-02   5.97671643e-02   4.61252153e-01\n",
      "   6.30505010e-03   1.06310785e-01   2.98366874e-01   5.61721146e-01\n",
      "   5.04955724e-02   7.44576991e-01   4.45074081e-01   3.44138294e-01\n",
      "   5.42857111e-01   1.54274628e-02   2.80294549e-02   1.46196559e-01\n",
      "   5.01431115e-02   1.30186724e-02   1.58574115e-02   2.78033197e-01\n",
      "   1.36267813e-02   3.04012686e-01   4.29279625e-01   4.61252153e-01\n",
      "   3.09900697e-02   1.87167838e-01   3.18929590e-02   7.64984787e-02\n",
      "   1.54501855e-01   4.15992230e-01   9.14104879e-02   8.87110010e-02\n",
      "   4.25476059e-02   2.38339990e-01   4.64020044e-01   1.97868332e-01\n",
      "   4.02316034e-01   5.52808702e-01   2.19283670e-01   3.36158544e-01\n",
      "   7.35383511e-01   6.19484521e-02   2.34742701e-01   1.18169963e-01\n",
      "   3.90584134e-02   2.26270799e-02   5.46638444e-02   4.58099246e-01\n",
      "   1.21344656e-01   4.90670949e-02   3.00317913e-01   2.96702255e-02\n",
      "   4.01023813e-02   5.46610713e-01   3.49584259e-02   1.15435608e-02\n",
      "   1.01916566e-01   3.06994226e-02   4.81541567e-02   2.56203059e-02\n",
      "   1.38116807e-01   2.90646851e-01   1.61525741e-01   2.17885241e-01\n",
      "   2.38445215e-02   3.74377444e-02   1.04421787e-02   1.79890078e-02\n",
      "   2.14627206e-01   2.75171511e-02   3.66925210e-01   5.81136309e-02\n",
      "   8.66166800e-02   2.46146590e-01   2.82926351e-01   1.28220767e-01\n",
      "   6.65816247e-01   6.26297653e-01   2.13723511e-01   4.29843932e-01\n",
      "   8.37058574e-02   7.44576991e-01   2.10342649e-02   7.67944977e-02\n",
      "   8.41390863e-02   7.48984993e-01   2.55992524e-02   3.80988754e-02\n",
      "   5.63641591e-03   1.07420221e-01   4.45074081e-01   1.37681410e-01\n",
      "   1.94361992e-02   2.22563408e-02   6.52900279e-01   1.72929779e-01\n",
      "   1.53460711e-01   3.63146290e-02   1.53460711e-01   2.13723511e-01\n",
      "   1.32793933e-02   6.91154599e-02   4.58099246e-01   1.61525741e-01\n",
      "   4.38481450e-01   1.26662299e-01   2.72810966e-01   6.14493825e-02\n",
      "   7.35383511e-01   3.10346991e-01   3.85159664e-02   3.51537883e-01\n",
      "   2.01694414e-01   5.47058523e-01   2.26270799e-02   3.39002013e-02\n",
      "   5.35191111e-02   9.86616313e-02   1.85574383e-01   5.21303773e-01\n",
      "   4.18155752e-02   6.14493825e-02   1.07420221e-01   3.74457315e-02\n",
      "   4.03312862e-01   9.24118310e-02   2.11210489e-01   3.03336699e-02\n",
      "   6.58556893e-02   4.69633400e-01   3.54543656e-01   4.42796759e-02\n",
      "   1.34777680e-01   4.57760453e-01   7.54193723e-01   6.05005361e-02\n",
      "   6.44062739e-03   1.96432918e-01   6.44062739e-03   1.48852840e-02\n",
      "   1.96437016e-01   3.79365534e-01   9.18027312e-02   6.75362110e-01\n",
      "   8.51453990e-02   2.89419770e-01   7.01488256e-02   2.39074603e-01\n",
      "   4.42115627e-02   1.48621321e-01   4.88330144e-03   5.71256764e-02\n",
      "   6.93117678e-02   1.94905289e-02   4.09509659e-01   4.88330144e-03\n",
      "   3.76139492e-01   8.31025019e-02   2.80105025e-02   1.38886832e-02\n",
      "   2.43484393e-01   4.06922027e-02   1.53550133e-01   1.69814050e-01\n",
      "   7.58789703e-02   2.26687133e-01   1.13889068e-01   3.38641465e-01\n",
      "   4.85991806e-01   5.25802821e-02   7.93269929e-03   2.85583492e-02\n",
      "   2.84822315e-01   1.25731543e-01   4.57568057e-02   1.79471280e-02\n",
      "   8.12529922e-02   2.47343570e-01   3.19625065e-02   2.29310274e-01\n",
      "   3.67629379e-02   1.07830437e-02   1.64742276e-01   2.55871397e-02\n",
      "   7.93269929e-03   6.08784929e-02   4.72857982e-01   1.49150537e-02\n",
      "   4.66332864e-03   6.92562684e-02   1.82399467e-01   2.26411708e-02\n",
      "   5.88819012e-02   1.34836202e-02   2.21691225e-02   5.15873395e-02\n",
      "   4.36566100e-02   1.47917956e-01   7.75274858e-02   9.96802654e-03\n",
      "   6.04487881e-02   9.56529751e-02   2.12116353e-02   1.29352175e-02\n",
      "   3.08048785e-01   4.95008752e-02   4.80929464e-01   2.31812358e-01\n",
      "   7.85021856e-02   4.57651541e-02   8.26800168e-02   6.33903369e-02\n",
      "   1.48852840e-02   1.27479047e-01   9.94726047e-02   1.96437016e-01\n",
      "   1.35652125e-01   8.22845846e-03   5.36409020e-02   2.12116353e-02\n",
      "   3.58536214e-01   1.51468173e-01   1.13940939e-01   5.12129590e-02\n",
      "   6.52626827e-02   5.36409020e-02   2.07965598e-02   4.87578176e-02\n",
      "   2.26687133e-01   2.02395409e-01   2.87369013e-01   5.36409020e-02\n",
      "   5.17398715e-01   1.61767099e-02   3.16821247e-01   2.85118092e-02\n",
      "   1.07214712e-01   1.51333762e-02   6.88211769e-02   3.39554250e-01\n",
      "   8.31025019e-02   1.90173462e-01   9.96802654e-03   3.36164013e-02\n",
      "   2.69759987e-02   3.47459726e-02   3.13223302e-02   1.03547238e-02\n",
      "   2.55871397e-02   2.15519685e-02   4.14633267e-02   1.66640297e-01\n",
      "   4.41029035e-02   2.43484393e-01   4.02841926e-01   8.51453990e-02\n",
      "   9.18027312e-02   6.92562684e-02   1.82399467e-01   1.66930407e-02\n",
      "   2.92131782e-01   3.64172272e-02   1.52828172e-01   1.94905289e-02\n",
      "   1.67512074e-02   5.87521970e-01   1.07214712e-01   5.94177619e-02\n",
      "   5.41316345e-02   1.59996971e-01   4.41029035e-02   6.12157345e-01\n",
      "   1.72911733e-01   2.97773760e-02   2.69759987e-02   1.20841088e-02\n",
      "   1.75840676e-01   1.64742276e-01   3.19069833e-01   1.15767391e-02\n",
      "   9.26958304e-03   1.05873216e-02   4.48393047e-01   2.07965598e-02\n",
      "   8.51453990e-02   1.10927567e-01   3.86442468e-02   3.76139492e-01\n",
      "   5.70046827e-02   7.19265938e-02   1.34836202e-02   1.32472785e-02\n",
      "   3.06702908e-02   4.63338792e-01   2.02395409e-01   2.10731402e-02\n",
      "   1.46031966e-02   2.78479606e-02   2.69759987e-02   2.47711577e-02\n",
      "   3.73352435e-03   8.60471874e-02   2.29310274e-01   3.86442468e-02\n",
      "   5.15079610e-02   1.16099283e-01   1.06176706e-02   9.94726047e-02\n",
      "   2.26139054e-02   4.88330144e-03   1.34836202e-02   2.07965598e-02\n",
      "   2.55871397e-02   4.29463863e-01   4.95008752e-02   3.95649225e-01\n",
      "   2.87510276e-01   6.75362110e-01   4.06922027e-02   1.34381056e-01\n",
      "   8.19308460e-02   1.94055419e-02   9.94726047e-02]\n",
      "0      941.0\n",
      "1      418.0\n",
      "2      349.0\n",
      "3        1.5\n",
      "4        6.5\n",
      "5      207.0\n",
      "6      985.0\n",
      "7      546.0\n",
      "8      468.0\n",
      "9      500.0\n",
      "10     159.5\n",
      "11     935.0\n",
      "12     817.0\n",
      "13     380.0\n",
      "14     793.5\n",
      "15     494.0\n",
      "16     913.0\n",
      "17     390.0\n",
      "18     166.5\n",
      "19     997.0\n",
      "20     273.0\n",
      "21      73.5\n",
      "22     581.0\n",
      "23     358.0\n",
      "24     478.0\n",
      "25     685.0\n",
      "26     775.0\n",
      "27     344.0\n",
      "28     517.0\n",
      "29     236.0\n",
      "       ...  \n",
      "969     97.0\n",
      "970    321.5\n",
      "971    860.0\n",
      "972    926.5\n",
      "973    790.0\n",
      "974    797.0\n",
      "975    821.0\n",
      "976    994.0\n",
      "977    519.0\n",
      "978    284.5\n",
      "979    719.5\n",
      "980    654.0\n",
      "981    459.0\n",
      "982    955.0\n",
      "983    489.0\n",
      "984    844.0\n",
      "985    990.0\n",
      "986    939.0\n",
      "987    865.0\n",
      "988    805.0\n",
      "989    122.0\n",
      "990    665.5\n",
      "991    143.0\n",
      "992    230.0\n",
      "993     29.5\n",
      "994    711.5\n",
      "995    423.0\n",
      "996    538.0\n",
      "997    878.0\n",
      "998    489.0\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "rank = pd.Series(sim_results_train[3]).rank(ascending = False)\n",
    "print sim_results_train[3].max()\n",
    "print sim_results_train[3]\n",
    "print rank"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[9.5,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 41.5,\n",
       " 1.5,\n",
       " 14.5,\n",
       " 3.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 68.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 19.5,\n",
       " 33.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 68.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 199.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 33.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 8.5,\n",
       " 33.5,\n",
       " 7.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 8.5,\n",
       " 14.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 13.5,\n",
       " 9.5,\n",
       " 4.5,\n",
       " 6.5,\n",
       " 5.5,\n",
       " 23.5,\n",
       " 1.5,\n",
       " 25.5,\n",
       " 51.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 33.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 13.5,\n",
       " 2.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 75.5,\n",
       " 1.5,\n",
       " 8.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 11.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 97.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 65.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 33.5,\n",
       " 9.5,\n",
       " 8.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 31.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 51.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 85.5,\n",
       " 10.5,\n",
       " 8.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 5.5,\n",
       " 25.5,\n",
       " 4.5,\n",
       " 17.5,\n",
       " 4.5,\n",
       " 5.5,\n",
       " 6.5,\n",
       " 69.5,\n",
       " 1.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 25.5,\n",
       " 19.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 49.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 43.5,\n",
       " 9.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 97.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 23.5,\n",
       " 5.5,\n",
       " 17.5,\n",
       " 9.5,\n",
       " 27.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 5.5,\n",
       " 20.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 25.5,\n",
       " 1.5,\n",
       " 65.5,\n",
       " 4.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 1.5,\n",
       " 19.5,\n",
       " 31.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 179.5,\n",
       " 20.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 21.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 33.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 30.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 21.5,\n",
       " 1.5,\n",
       " 32.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 41.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 17.5,\n",
       " 2.5,\n",
       " 13.5,\n",
       " 11.5,\n",
       " 3.5,\n",
       " 30.5,\n",
       " 2.5,\n",
       " 23.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 8.5,\n",
       " 13.5,\n",
       " 7.5,\n",
       " 32.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 8.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 8.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 33.5,\n",
       " 13.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 6.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 27.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 11.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 13.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 46.5,\n",
       " 31.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 46.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 5.5,\n",
       " 18.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 13.5,\n",
       " 15.5,\n",
       " 13.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 4.5,\n",
       " 4.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 13.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 4.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 57.5,\n",
       " 2.5,\n",
       " 21.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 11.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 15.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 66.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 6.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 25.5,\n",
       " 27.5,\n",
       " 37.5,\n",
       " 39.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 41.5,\n",
       " 11.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 25.5,\n",
       " 17.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 51.5,\n",
       " 17.5,\n",
       " 26.5,\n",
       " 71.5,\n",
       " 31.5,\n",
       " 2.5,\n",
       " 14.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 37.5,\n",
       " 7.5,\n",
       " 29.5,\n",
       " 4.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 13.5,\n",
       " 6.5,\n",
       " 4.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 33.5,\n",
       " 4.5,\n",
       " 9.5,\n",
       " 26.5,\n",
       " 1.5,\n",
       " 56.5,\n",
       " 8.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 26.5,\n",
       " 11.5,\n",
       " 11.5,\n",
       " 3.5,\n",
       " 29.5,\n",
       " 1.5,\n",
       " 21.5,\n",
       " 6.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 17.5,\n",
       " 13.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 14.5,\n",
       " 25.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 26.5,\n",
       " 1.5,\n",
       " 8.5,\n",
       " 4.5,\n",
       " 13.5,\n",
       " 57.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 25.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 19.5,\n",
       " 6.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 57.5,\n",
       " 2.5,\n",
       " 41.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 23.5,\n",
       " 18.5,\n",
       " 3.5,\n",
       " 27.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 10.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 66.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 15.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 5.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 7.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 13.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 3.5,\n",
       " 6.5,\n",
       " 4.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 11.5,\n",
       " 4.5,\n",
       " 5.5,\n",
       " 9.5,\n",
       " 8.5,\n",
       " 6.5,\n",
       " 31.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 6.5,\n",
       " 4.5,\n",
       " 12.5,\n",
       " 6.5,\n",
       " 5.5,\n",
       " 5.5,\n",
       " 53.5,\n",
       " 39.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 7.5,\n",
       " 69.5,\n",
       " 2.5,\n",
       " 10.5,\n",
       " 1.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 10.5,\n",
       " 6.5,\n",
       " 7.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 7.5,\n",
       " 71.5,\n",
       " 10.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 17.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 5.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 56.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 4.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 16.5,\n",
       " 3.5,\n",
       " 14.5,\n",
       " 2.5,\n",
       " 13.5,\n",
       " 5.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 25.5,\n",
       " 3.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 23.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 13.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 47.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 13.5,\n",
       " 7.5,\n",
       " 3.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 5.5,\n",
       " 7.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 14.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 11.5,\n",
       " 49.5,\n",
       " 5.5,\n",
       " 16.5,\n",
       " 5.5,\n",
       " 12.5,\n",
       " 8.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 17.5,\n",
       " 6.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 15.5,\n",
       " 5.5,\n",
       " 4.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 49.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 9.5,\n",
       " 2.5,\n",
       " 2.5,\n",
       " 52.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 38.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 21.5,\n",
       " 21.5,\n",
       " 9.5,\n",
       " 3.5,\n",
       " 35.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 35.5,\n",
       " 3.5,\n",
       " 18.5,\n",
       " 4.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 26.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 44.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 56.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 6.5,\n",
       " 44.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 19.5,\n",
       " 11.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 10.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 22.5,\n",
       " 47.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 21.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 99.5,\n",
       " 3.5,\n",
       " 52.5,\n",
       " 17.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 10.5,\n",
       " 45.5,\n",
       " 37.5,\n",
       " 35.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 37.5,\n",
       " 44.5,\n",
       " 16.5,\n",
       " 13.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 13.5,\n",
       " 3.5,\n",
       " 6.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 17.5,\n",
       " 4.5,\n",
       " 45.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 10.5,\n",
       " 7.5,\n",
       " 3.5,\n",
       " 2.5,\n",
       " 6.5,\n",
       " 44.5,\n",
       " 1.5,\n",
       " 7.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 1.5,\n",
       " 15.5,\n",
       " 6.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 5.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 37.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 10.5,\n",
       " 25.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 2.5,\n",
       " 18.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 11.5,\n",
       " 16.5,\n",
       " 7.5,\n",
       " 2.5,\n",
       " 21.5,\n",
       " 3.5,\n",
       " 9.5,\n",
       " 1.5,\n",
       " 31.5,\n",
       " 56.5,\n",
       " 2.5,\n",
       " 3.5,\n",
       " 17.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 3.5,\n",
       " 5.5,\n",
       " 3.5,\n",
       " 39.5,\n",
       " 22.5,\n",
       " 3.5,\n",
       " 1.5,\n",
       " 38.5,\n",
       " 26.5,\n",
       " 5.5,\n",
       " 1.5,\n",
       " 1.5,\n",
       " 3.5,\n",
       " 1867.5,\n",
       " ...]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sls.rank[i]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    999.000000\n",
       "mean       4.926927\n",
       "std        7.862557\n",
       "min        1.000000\n",
       "25%        1.000000\n",
       "50%        2.000000\n",
       "75%        5.000000\n",
       "max      100.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 真正的误差在这里：\n",
    "pd.Series(rank_results_train).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    1000.000000\n",
       "mean        6.672000\n",
       "std         9.635105\n",
       "min         1.000000\n",
       "25%         1.500000\n",
       "50%         3.000000\n",
       "75%         7.000000\n",
       "max       109.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(rank_results_test).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4YAAAHSCAYAAACw89XFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fce711e24d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "matplotlib.rcParams['figure.figsize'] = (15.0, 7.5)\n",
    "plt.plot(rank_results_train,'ro')\n",
    "plt.xlabel('Document Number')\n",
    "plt.ylabel('The ranking of the real pair')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4YAAAHSCAYAAACw89XFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fce7131c990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib\n",
    "matplotlib.rcParams['figure.figsize'] = (15.0, 7.5)\n",
    "plt.plot(rank_results_test,'bo')\n",
    "plt.xlabel('Document Number')\n",
    "plt.ylabel('The ranking of the real pair')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [Root]",
   "language": "python",
   "name": "Python [Root]"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}