notebooks/3a_hyperopt.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook I use hyperoptimisation to find the best data filters and the best decision tree parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-24T11:42:36.209833Z",
     "start_time": "2017-03-24T19:42:34.670131+08:00"
    }
   },
   "outputs": [],
   "source": [
    "from path import Path\n",
    "import arrow\n",
    "import json\n",
    "import pytz\n",
    "from pprint import pprint\n",
    "from tqdm import tqdm_notebook as tqdm\n",
    "import re, os, collections, itertools, uuid, logging\n",
    "import tempfile\n",
    "# import tables\n",
    "\n",
    "import zipfile\n",
    "import urllib\n",
    "\n",
    "import ee\n",
    "import pyproj\n",
    "import numpy as np\n",
    "import scipy as sp\n",
    "import pandas as pd\n",
    "import geopandas as gpd\n",
    "from matplotlib import pyplot as plt\n",
    "import seaborn as sns\n",
    "import shapely\n",
    "\n",
    "\n",
    "%matplotlib inline\n",
    "# %precision 4\n",
    "plt.style.use('fivethirtyeight')\n",
    "plt.rcParams['figure.figsize'] = (12, 6) # bigger plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 301,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:18:02.082328Z",
     "start_time": "2017-03-26T13:18:02.079838+08:00"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from collections import OrderedDict\n",
    "import tabulate\n",
    "from IPython.display import Markdown, display"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-24T11:42:36.720162Z",
     "start_time": "2017-03-24T19:42:36.717812+08:00"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "os.environ['CUDA_VISIBLE_DEVICES']=\"\" # don't need this for skleanr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-24T11:42:38.402360Z",
     "start_time": "2017-03-24T19:42:37.166038+08:00"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import keras\n",
    "import tensorflow\n",
    "import sklearn\n",
    "from keras.utils.io_utils import HDF5Matrix\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import Reshape, InputLayer, Permute, RepeatVector, Dropout, LocallyConnected1D\n",
    "from keras.layers.core import Activation\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "from keras.layers.convolutional import UpSampling1D\n",
    "from keras.layers.convolutional import Convolution1D, MaxPooling1D, MaxPooling2D\n",
    "from keras.layers.core import Flatten\n",
    "from keras.optimizers import SGD\n",
    "from keras.datasets import mnist\n",
    "from keras.layers import (Input, LocallyConnected1D, ZeroPadding1D, Cropping1D, Embedding, Merge, merge,\n",
    "    Cropping2D, Convolution1D, Convolution2D, Deconvolution2D, BatchNormalization, UpSampling1D, RepeatVector)\n",
    "from keras.layers import embeddings, convolutional, activations, normalization, advanced_activations, ZeroPadding2D, UpSampling2D\n",
    "from keras.models import Model\n",
    "from keras_tqdm import TQDMCallback, TQDMNotebookCallback"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-24T11:42:38.406410Z",
     "start_time": "2017-03-24T19:42:38.404062+08:00"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# %load_ext autoreload\n",
    "# %autoreload 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-24T11:42:43.013567Z",
     "start_time": "2017-03-24T19:42:39.350448+08:00"
    }
   },
   "outputs": [],
   "source": [
    "# %load_ext autoreload\n",
    "# %autoreload 2\n",
    "\n",
    "helper_dir = str(Path('.').abspath())\n",
    "if helper_dir not in os.sys.path:\n",
    "    os.sys.path.append(helper_dir)\n",
    "    \n",
    "from leak_helpers.earth_engine import display_ee, get_boundary, tifs2np, bands_s2, download_image, bands_s2\n",
    "from leak_helpers.geometry import diffxy, resample_polygon\n",
    "from leak_helpers.modelling import ImageDataGenerator, dice_coef_loss\n",
    "from leak_helpers.visualization import imshow_bands\n",
    "from leak_helpers.analysis import parse_classification_report, find_best_dummy_classification, calculate_result_class\n",
    "from leak_helpers.modelling.filters import is_not_cloudy, is_not_center_cloudy, is_image_within, is_leak, filter_split_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-24T11:42:43.018865Z",
     "start_time": "2017-03-24T19:42:43.015056+08:00"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('1.2.2', '1.0.0')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "keras.__version__, tensorflow.__version__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:19:29.430946Z",
     "start_time": "2017-03-26T13:19:29.418263+08:00"
    }
   },
   "outputs": [],
   "source": [
    "# params\n",
    "\n",
    "data_path = [\n",
    "#         Path('../data/scraped_satellite_images/s2-AUTX_v6_COPERNICUS-S2'),\n",
    "    Path('../data/scraped_satellite_images/l7-AUTX_v2_LANDSAT-LE7_L1T'),\n",
    "#     Path('../data/scraped_satellite_images/l8-AUTX_v2_LANDSAT-LC8_L1T'),\n",
    "#     Path('../data/scraped_satellite_images/s1-all_COPERNICUS-S1_GRD'),   # small\n",
    "][0]\n",
    "\n",
    "notebook_name = 'leak_detection_CNN'\n",
    "batch_size=128\n",
    "random_seed = 1337\n",
    "test_fraction = 0.4\n",
    "timespan_before = 60*60*24*2 # we will only take image that where X seconds before the leak (e.g. 1 day)\n",
    "max_cloud_cover = 1.1 # cloud cover must less than this\n",
    "thresh = 0.5 # balance precision vs recall\n",
    "balanced_classes = False \n",
    "normalized = True\n",
    "\n",
    "# derived\n",
    "seconds_in_a_day = 60*60*24\n",
    "ts=arrow.utcnow().format('YYYYMMDD-HH-mm-ss')\n",
    "model_name = '{notebook_name:}__{ts:}'.format(ts=ts,notebook_name=notebook_name)\n",
    "outdir = Path('../output').joinpath(model_name)\n",
    "outdir.makedirs_p()\n",
    "outdir\n",
    "\n",
    "target_names = ['no leak','leak']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:19:33.440105Z",
     "start_time": "2017-03-26T13:19:30.139255+08:00"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((3564, 14, 25, 25),\n",
       " array([False, False, False, ..., False, False,  True], dtype=bool),\n",
       " dict_keys(['distance', 'name', 'scale', 'leak', 'crs', 'image']))"
      ]
     },
     "execution_count": 303,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "script_metadata = json.load(open(data_path.joinpath('script_metadata.json')))\n",
    "# test load\n",
    "import h5py\n",
    "metadatas = json.load(open(data_path.joinpath('data_metadata.json')))\n",
    "with h5py.File(data_path.joinpath('data.h5'),'r') as h5f:\n",
    "    X_raw = h5f['X'][:]\n",
    "    y_raw = h5f['y'][:]\n",
    "X_raw.shape, y_raw, metadatas[0].keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T09:47:54.194535Z",
     "start_time": "2017-03-13T17:47:54.187303+08:00"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:19:34.133300Z",
     "start_time": "2017-03-26T13:19:33.441815+08:00"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(926, 14, 25, 25) (926,) 926\n",
      "(883, 14, 25, 25) (883,) 883\n",
      "(398, 14, 25, 25) (398,) 398\n"
     ]
    }
   ],
   "source": [
    "X_train, y_train, metadata_train, X_val, y_val, metadata_val, X_test, y_test, metadata_test = filter_split_data(\n",
    "    X_raw,\n",
    "    y_raw,\n",
    "    metadatas,\n",
    "    max_cloud_cover=1.1, \n",
    "    timespan_before=timespan_before,\n",
    "    test_fraction=test_fraction, \n",
    "    random_seed=random_seed,\n",
    "    balanced_classes=balanced_classes,\n",
    "    normalized=normalized,\n",
    "    filter_center_cloudy=True,\n",
    ")   \n",
    "print(X_train.shape,y_train.shape, len(metadata_train))\n",
    "print(X_test.shape,y_test.shape, len(metadata_test))\n",
    "print(X_val.shape,y_val.shape, len(metadata_val))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-12T04:05:05.554070Z",
     "start_time": "2017-03-12T12:05:05.550725+08:00"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# model\n",
    "\n",
    "- batch norm helped\n",
    "- normalising X helped\n",
    "\n",
    "Refs:\n",
    "- conv only net uses strides instead of MaxPooling\n",
    "- example of heatmap generationg https://github.com/heuritech/convnets-keras/blob/master/convnetskeras/convnets.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:19:34.155906Z",
     "start_time": "2017-03-26T13:19:34.134973+08:00"
    },
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "# conv model\n",
    "\n",
    "def get_model(input_shape, batch_norm=True, activation='relu', border_mode='same', channels = 14*2, pooling = 2):    \n",
    "    \n",
    "    model = Sequential()\n",
    "    # batch_input_shape=(None, X_train.shape[1], pixel_length, pixel_length)\n",
    "    model.add(InputLayer(input_shape=input_shape, name='input'))\n",
    "    model.add(Permute((2,3,1)))\n",
    "\n",
    "    model.add(Convolution2D(channels,3,3,border_mode=border_mode,subsample=(pooling, pooling)))\n",
    "    if batch_norm: model.add(BatchNormalization())\n",
    "    model.add(Activation(activation))\n",
    "\n",
    "    model.add(Convolution2D(channels*3,3,3,border_mode=border_mode,subsample=(3, 3)))\n",
    "    if batch_norm: model.add(BatchNormalization())\n",
    "    model.add(Activation(activation))\n",
    "\n",
    "    model.add(Convolution2D(channels*2*3,3,3,border_mode=border_mode,subsample=(pooling, pooling)))\n",
    "    if batch_norm: model.add(BatchNormalization())\n",
    "    model.add(Activation(activation))\n",
    "\n",
    "    model.add(Convolution2D(channels*2*3*2,1,1,border_mode=border_mode,subsample=(pooling, pooling)))\n",
    "    if batch_norm: model.add(BatchNormalization())\n",
    "    model.add(Activation(activation))\n",
    "    \n",
    "    model.add(Convolution2D(1,1,1,border_mode=border_mode))\n",
    "    model.add(Flatten())\n",
    "    model.add(Activation('sigmoid'))\n",
    "    \n",
    "    model.compile(loss='binary_crossentropy',optimizer='nadam', metrics=['accuracy'])\n",
    "    \n",
    "    return model\n",
    "\n",
    "input_shape = (X_train.shape[1], X_train.shape[2], X_train.shape[3])\n",
    "# model = get_model(input_shape, border_mode='same' ,batch_norm=True)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-07T07:32:33.915919Z",
     "start_time": "2017-03-07T15:32:33.893877+08:00"
    }
   },
   "source": [
    "# Hyperopt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-12T02:24:36.259912Z",
     "start_time": "2017-03-12T10:24:36.250787+08:00"
    }
   },
   "source": [
    "## hyperopt nn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:19:34.261396Z",
     "start_time": "2017-03-26T13:19:34.157906+08:00"
    },
    "code_folding": [
     0
    ]
   },
   "outputs": [],
   "source": [
    "# hypropt nn\n",
    "from hyperopt import fmin, tpe, hp, STATUS_OK, Trials\n",
    "from sklearn.metrics import roc_auc_score\n",
    "import sys\n",
    "from sklearn.dummy import DummyClassifier\n",
    "\n",
    "space = {\n",
    "        \n",
    "            # model\n",
    "            'batch_norm': hp.choice('batch_norm', [False, True]),\n",
    "            'activation':hp.choice('batch_norm', ['relu','tanh','hard_sigmoid','sigmoid']),\n",
    "            'channels': 14*2,\n",
    "            'pooling':2,\n",
    "            'optimizer':'nadam',\n",
    "    \n",
    "            # data\n",
    "            'max_cloud_cover':hp.uniform('max_cloud_cover', 0.01, 0.3),\n",
    "            'timespan_before': seconds_in_a_day*1, # hp.uniform('timespan_before', seconds_in_a_day*1, seconds_in_a_day*7),\n",
    "            'balanced_classes':hp.choice('balanced_classes', [False, True]),\n",
    "            'normalized': hp.choice('normalized', [False, True]),\n",
    "            'filter_center_cloudy': hp.choice('filter_center_cloudy', [False, True]),\n",
    "    \n",
    "            # std is 5% so max should prob be ~5%\n",
    "            'channel_shift_range': 0.007, #hp.uniform('channel_shift_range', 0.0, 0.07),\n",
    "            'rotation_range': 25, #hp.uniform('rotation_range', 0.0, 45),\n",
    "    \n",
    "            'batch_size' : batch_size,\n",
    "            'samples_per_epoch': len(X_raw)*4,\n",
    "            'nb_epochs' :  40,\n",
    "        }\n",
    "\n",
    "# from keras.models import Sequential\n",
    "# from keras.layers import InputLayer\n",
    "# from keras.layers.core import Dense, Dropout, Activation\n",
    "# from keras.optimizers import Adadelta, Adam, rmsprop\n",
    "\n",
    "# TODO make everything run of the params\n",
    "# TODO encode params into hyperopt\n",
    "# TODO run some trials to try differen't data types, could use decision tree, or even null model?\n",
    "def f_nn(params):   \n",
    "\n",
    "    # Get data\n",
    "    X_train, y_train, metadata_train, X_val, y_val, metadata_val, X_test, y_test, metadata_test = filter_split_data(\n",
    "        X_raw,\n",
    "        y_raw,\n",
    "        metadatas,\n",
    "        max_cloud_cover=params['max_cloud_cover'], \n",
    "        timespan_before=params['timespan_before'],\n",
    "        test_fraction=test_fraction, \n",
    "        random_seed=random_seed,\n",
    "        balanced_classes=params['balanced_classes'],\n",
    "        normalized=params['normalized'],\n",
    "        filter_center_cloudy=params['filter_center_cloudy']\n",
    "    )    \n",
    "\n",
    "    datagen = ImageDataGenerator(\n",
    "    #     featurewise_center=True,\n",
    "    #     featurewise_std_normalization=True,\n",
    "        channel_shift_range=params['channel_shift_range'],\n",
    "        rotation_range=params['rotation_range'],\n",
    "    #     width_shift_range=0.05,\n",
    "    #     height_shift_range=0.05,\n",
    "        horizontal_flip=True,\n",
    "        vertical_flip=True,\n",
    "       dim_ordering='th',\n",
    "    #     rescale=0.02,\n",
    "    #     zoom_range=0.05,\n",
    "    #     shear_range=0.05,\n",
    "    )\n",
    "\n",
    "    # datagen.fit(X_train)\n",
    "\n",
    "    # Get model\n",
    "    input_shape = (X_train.shape[1], X_train.shape[2], X_train.shape[3])\n",
    "    model = get_model(\n",
    "        input_shape, \n",
    "        border_mode=params['border_mode'],\n",
    "        batch_norm=params['batch_norm'],\n",
    "        channels=params['channels'],\n",
    "        pooling=params['pooling']\n",
    "        \n",
    "    )\n",
    "\n",
    "    if balanced_classes:\n",
    "        model.compile(loss='binary_crossentropy',optimizer=params['optimizer'], metrics=['accuracy','matthews_correlation'])\n",
    "    else:\n",
    "        model.compile(loss=dice_coef_loss,optimizer=params['optimizer'], metrics=['accuracy','matthews_correlation'])\n",
    "\n",
    "    # pretrain/test\n",
    "    history = model.fit_generator(\n",
    "        datagen.flow(X_train, y_train, batch_size=params['batch_size']),\n",
    "        samples_per_epoch=params['samples_per_epoch'],\n",
    "        verbose=0, \n",
    "        nb_epoch=params['nb_epoch'], \n",
    "        validation_data=[X_val,y_val],\n",
    "        callbacks=[\n",
    "              TQDMNotebookCallback(),\n",
    "                keras.callbacks.EarlyStopping(patience=6, monitor='loss'),\n",
    "                keras.callbacks.ReduceLROnPlateau(monitor='loss', patience=3, verbose=1)\n",
    "        ]\n",
    "    )\n",
    "\n",
    "    # results\n",
    "    last_metrics = [x[-1] for x in history.history.values()]\n",
    "    last_metrics = dict(zip(history.history.keys(),last_metrics))\n",
    "\n",
    "    # accuracy X_train\n",
    "    model.compile(loss='binary_crossentropy',optimizer='nadam', metrics=['accuracy','mean_squared_error','precision','recall','fbeta_score','fmeasure','matthews_correlation'])\n",
    "    score = model.evaluate(X_val,y_val, batch_size=batch_size, verbose=False)\n",
    "    metrics = dict(zip(model.metrics_names,score))\n",
    "\n",
    "    # score on X_val\n",
    "    y_pred = model.predict(X_val, verbose=False).T[0]\n",
    "    report = parse_classification_report(sklearn.metrics.classification_report(y_val>thresh,y_pred>thresh,target_names=target_names))\n",
    "    loss=-metrics['matthews_correlation']\n",
    "    \n",
    "    X_train2 = X_train.reshape((-1,14*24*24))\n",
    "    X_val2 = X_val.reshape((-1,14*24*24))\n",
    "    \n",
    "    clf = DummyClassifier(strategy='uniform')\n",
    "    clf.fit(X_train2,y_train)\n",
    "    y_pred = clf.predict(X_val2)\n",
    "    score_dummy = clf.score(X_val2, y_val)\n",
    "    matthews_corrcoef_dummy = sklearn.metrics.matthews_corrcoef(y_val>thresh, y_pred>thresh)\n",
    "    report_dummy = parse_classification_report(sklearn.metrics.classification_report(y_val>thresh,y_pred>thresh,target_names=target_names))\n",
    "    \n",
    "    \n",
    "    \n",
    "    sys.stdout.flush() \n",
    "    return dict(\n",
    "                # core info\n",
    "                loss=-loss,\n",
    "                status=STATUS_OK,\n",
    "        \n",
    "                # info\n",
    "                history=history.history, \n",
    "                metrics=dict(\n",
    "                    last_metrics=last_metrics,\n",
    "                    report=report.to_dict(),\n",
    "                    metrics=metrics,\n",
    "                ),\n",
    "                dummy_metrics=dict(\n",
    "                    report_dummy=report_dummy.to_dict(),\n",
    "                    score_dummy=score_dummy,\n",
    "                    matthews_corrcoef_dummy=matthews_corrcoef_dummy,\n",
    "                ),\n",
    "                \n",
    "                # data dumps as json\n",
    "                attachments=dict(\n",
    "                    model=model.to_json()\n",
    "                )\n",
    "               )\n",
    "            \n",
    "\n",
    "\n",
    "# trials = Trials()\n",
    "# best = fmin(f_nn, space, algo=tpe.suggest, max_evals=2, trials=trials, verbose=2)\n",
    "# print ('best: ')\n",
    "# print (best)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-10T06:30:44.245328Z",
     "start_time": "2017-03-10T14:30:44.242000+08:00"
    }
   },
   "source": [
    "## hyperopt filter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:19:34.343184Z",
     "start_time": "2017-03-26T13:19:34.263426+08:00"
    }
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "This experiment will look at differen't way of filtering data\n",
    "and will use decision trees as quick ways of evaluating data.\n",
    "Mathews cooeffecient is used as loss since it performs well for unbalanced datasets.\n",
    "\"\"\"\n",
    "\n",
    "from hyperopt import fmin, tpe, hp, STATUS_OK, Trials\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from hyperopt.mongoexp import MongoTrials\n",
    "import sys\n",
    "from sklearn import tree\n",
    "from sklearn.dummy import DummyClassifier\n",
    "import time\n",
    "seconds_in_a_day = 60*60*24\n",
    "\n",
    "trials = Trials(exp_key='data_filters_%s'%data_path.basename())\n",
    "# trials = MongoTrials('mongo://localhost:27017/hyperopt_leaks2/jobs', exp_key='data_filters_v2')\n",
    "\n",
    "\n",
    "\n",
    "space = {\n",
    "\n",
    "            'max_cloud_cover': hp.uniform('max_cloud_cover', 0.01, 1.0),\n",
    "            'timespan_before': hp.uniform('timespan_before', seconds_in_a_day*0.5, seconds_in_a_day*7),\n",
    "            'balanced_classes':hp.choice('balanced_classes', [False, True]),\n",
    "            'normalized': hp.choice('normalized', [False, True]),\n",
    "            'filter_center_cloudy': False, #hp.choice('filter_center_cloudy', [False, True]),\n",
    "    \n",
    "            # std is 5% so max should prob be ~5%\n",
    "            'channel_shift_range': hp.uniform('channel_shift_range', 0.0, 0.25),\n",
    "            'rotation_range': hp.uniform('rotation_range', 0.0, 45),\n",
    "            'width_shift_range': hp.uniform('width_shift_range', 0.0, 0.5),\n",
    "            'height_shift_range': hp.uniform('height_shift_range', 0.0, 0.5),\n",
    "            'horizontal_flip': True, #hp.choice('horizontal_flip', [False, True]),\n",
    "            'vertical_flip': True, #hp.choice('vertical_flip', [False, True]),\n",
    "            'rescale': hp.uniform('rescale', 0.0, 0.5),\n",
    "            'zoom_range': 0 ,# hp.uniform('zoom_range', 0.0, 0.5),\n",
    "            'shear_range': 0, # hp.uniform('shear_range', 0.0, 0.5),\n",
    "    \n",
    "            'max_depth': 50,\n",
    "            'batch_size' : 2000*10,\n",
    "            'nb_epochs' :  1,\n",
    "        }\n",
    "\n",
    "\n",
    "def f_dt(params):\n",
    "    t0 = time.time()\n",
    "\n",
    "    # Get data\n",
    "    X_train, y_train, metadata_train, X_val, y_val, metadata_val, X_test, y_test, metadata_test = filter_split_data(\n",
    "        X_raw,\n",
    "        y_raw,\n",
    "        metadatas,\n",
    "        max_cloud_cover=params['max_cloud_cover'], \n",
    "        timespan_before=params['timespan_before'],\n",
    "        test_fraction=test_fraction,\n",
    "        val_fraction=test_fraction,\n",
    "        random_seed=random_seed,\n",
    "        balanced_classes=params['balanced_classes'],\n",
    "        normalized=params['normalized'],\n",
    "        filter_center_cloudy=params['filter_center_cloudy']\n",
    "    )  \n",
    "\n",
    "    datagen = ImageDataGenerator(\n",
    "    #     featurewise_center=True,\n",
    "    #     featurewise_std_normalization=True,\n",
    "        channel_shift_range=params['channel_shift_range'],\n",
    "        rotation_range=params['rotation_range'],\n",
    "        width_shift_range=params['width_shift_range'],\n",
    "        height_shift_range=params['height_shift_range'],\n",
    "        horizontal_flip=params['horizontal_flip'],\n",
    "        vertical_flip=params['vertical_flip'],\n",
    "        rescale=params['rescale'],\n",
    "        zoom_range=params['zoom_range'],\n",
    "        shear_range=params['shear_range'],\n",
    "        dim_ordering='th',\n",
    "    )\n",
    "\n",
    "    # datagen.fit(X_train)\n",
    "    \n",
    "    gen = datagen.flow(X_train, y_train, batch_size=params['batch_size']*params['nb_epochs'])\n",
    "    X_trainb, y_trainb = next(gen)\n",
    "    X_train2b = X_trainb.reshape((len(X_trainb),-1))\n",
    "    X_val2 = X_val.reshape((len(X_val),-1))\n",
    "    X_test2 = X_test.reshape((len(X_test),-1))\n",
    "    \n",
    "    t1=time.time()-t0\n",
    "\n",
    "    # Get model    \n",
    "    model = tree.DecisionTreeRegressor(max_depth=params['max_depth'])\n",
    "    model.fit(X_train2b, y_trainb) \n",
    "    y_pred = model.predict(X_val2)\n",
    "    score = model.score(X_val2, y_val)\n",
    "\n",
    "\n",
    "    # score on X_val\n",
    "    report = parse_classification_report(sklearn.metrics.classification_report(y_val>thresh,y_pred>thresh,target_names=target_names))\n",
    "    matthews_corrcoef = sklearn.metrics.matthews_corrcoef(y_val>thresh, y_pred>thresh)\n",
    "    loss=-matthews_corrcoef\n",
    "    \n",
    "    # do a dummy score too\n",
    "    clf = DummyClassifier(strategy='uniform')\n",
    "    clf.fit(X_train2b,y_trainb)\n",
    "    y_pred = clf.predict(X_val2)\n",
    "    score_dummy = clf.score(X_val2, y_val)\n",
    "    matthews_corrcoef_dummy = sklearn.metrics.matthews_corrcoef(y_val>thresh, y_pred>thresh)\n",
    "    report_dummy = parse_classification_report(sklearn.metrics.classification_report(y_val>thresh,y_pred>thresh,target_names=target_names))\n",
    "    \n",
    "    \n",
    "#     sys.stdout.flush()\n",
    "    print(t1,'loss',loss)\n",
    "    return dict(loss=-loss,\n",
    "                status=STATUS_OK,\n",
    "                run_time=t1,\n",
    "                metrics=dict(\n",
    "                    report=report.to_dict(),\n",
    "                    matthews_corrcoef=matthews_corrcoef,\n",
    "                ),             \n",
    "                dummy_metrics=dict(\n",
    "                    report_dummy=report_dummy.to_dict(),\n",
    "                    score_dummy=score_dummy,\n",
    "                    matthews_corrcoef_dummy=matthews_corrcoef_dummy,\n",
    "                ),\n",
    "                attachments=dict(                \n",
    "#                     model=model.__dict__\n",
    "                )\n",
    "               )\n",
    "            \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2017-03-26T05:19:36.220Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.595038652420044 loss -0.0386642846727\n",
      "0.7030308246612549 loss 0.0932923927765\n",
      "1.8368890285491943 loss -0.00368705710051\n",
      "1.4079639911651611 loss -0.0166153571975\n",
      "1.2733204364776611 loss 0.0494893096183\n",
      "1.181145191192627 loss -0.0272881007883\n",
      "1.0876176357269287 loss -0.0873704056661\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/isisilon/.virtualenvs/py3syspck/lib/python3.4/site-packages/sklearn/metrics/classification.py:1113: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/home/isisilon/.virtualenvs/py3syspck/lib/python3.4/site-packages/sklearn/metrics/classification.py:516: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  mcc = cov_ytyp / np.sqrt(var_yt * var_yp)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.743326187133789 loss -0.0\n",
      "1.926445722579956 loss 0.0193785923943\n",
      "1.1760334968566895 loss 0.0256270290006\n",
      "1.0056226253509521 loss 0.0151752911359\n",
      "0.6968693733215332 loss -0.0\n",
      "1.6277260780334473 loss 0.0313566404476\n",
      "1.0868237018585205 loss 0.0327399851106\n",
      "1.5668323040008545 loss -0.0130328376203\n",
      "1.2413296699523926 loss 0.0866871379433\n",
      "1.6522138118743896 loss -0.0115787928458\n",
      "1.666116714477539 loss 0.032069708152\n",
      "1.478454351425171 loss 0.0554018715248\n",
      "1.4123950004577637 loss 0.0432574237166\n",
      "0.8946154117584229 loss -0.0573102276588\n",
      "1.3224105834960938 loss -0.0328301167799\n",
      "0.896176815032959 loss 0.0690553296203\n",
      "1.2526631355285645 loss -0.0116495714929\n",
      "0.7282388210296631 loss -0.0490225639152\n",
      "1.40065336227417 loss -0.0580845778331\n",
      "1.062377691268921 loss -0.0385831356798\n",
      "1.9305071830749512 loss -0.00965632680232\n",
      "1.2048382759094238 loss -0.00383326827693\n",
      "1.3617136478424072 loss 0.0413901492425\n",
      "0.6661429405212402 loss 0.0660413176414\n",
      "1.8069255352020264 loss -0.00166692218296\n",
      "1.3109245300292969 loss 0.0129863050497\n",
      "1.1031620502471924 loss 0.0430726496665\n",
      "1.2229349613189697 loss 0.0108699471106\n",
      "1.1721446514129639 loss -0.077683279551\n",
      "1.4071061611175537 loss -0.0438634463307\n",
      "0.7453796863555908 loss 0.0217796037852\n",
      "1.3287794589996338 loss 0.00682451169305\n",
      "1.309830904006958 loss -0.052583048205\n",
      "0.6199619770050049 loss 0.0999807747763\n",
      "0.9225940704345703 loss 0.0700152751617\n",
      "0.6101760864257812 loss 0.0503144528418\n",
      "1.5531032085418701 loss 0.0435480061659\n",
      "1.0505895614624023 loss 0.023719131803\n",
      "0.8544209003448486 loss -0.0194424132609\n",
      "1.5737848281860352 loss 0.0127326155371\n",
      "0.9572079181671143 loss 0.0444937556926\n",
      "1.6960012912750244 loss 0.0044726058412\n",
      "0.8747713565826416 loss 0.0530595449973\n",
      "1.0718188285827637 loss 0.0212958855\n",
      "1.4130747318267822 loss -0.0348723689193\n",
      "1.6548922061920166 loss 0.0353705673104\n",
      "0.8295555114746094 loss -0.019279313829\n",
      "1.3360414505004883 loss -0.00261073502835\n",
      "1.7084426879882812 loss 0.00219235985541\n",
      "0.9738762378692627 loss 0.0568051873845\n",
      "1.614677906036377 loss -0.0107594694354\n",
      "1.5387177467346191 loss 0.0176764509818\n",
      "0.6341819763183594 loss 0.00328230714366\n",
      "0.9057796001434326 loss 0.0618931372168\n",
      "0.6017038822174072 loss -0.116887940771\n",
      "1.061206579208374 loss -0.0378005077239\n",
      "1.5743980407714844 loss 0.0382214483929\n",
      "1.1713993549346924 loss 0.0690436611898\n",
      "1.332535743713379 loss -0.0238934099208\n",
      "0.9691309928894043 loss -0.0186232596099\n",
      "0.9107766151428223 loss 0.0127178228831\n",
      "1.2141294479370117 loss 0.00759240268123\n",
      "0.8369030952453613 loss -0.116529696664\n",
      "0.992255687713623 loss 0.0122812687697\n",
      "1.63787841796875 loss 0.0134579230323\n",
      "0.6554107666015625 loss 0.00119038762326\n",
      "1.7501065731048584 loss 0.0614732962055\n",
      "1.2191259860992432 loss -0.0501327966455\n",
      "0.7595231533050537 loss 0.0114108774431\n",
      "0.6717150211334229 loss -0.0109562134162\n",
      "1.1173820495605469 loss -0.0618928733173\n",
      "1.3781368732452393 loss -0.0112584600853\n",
      "0.873100996017456 loss 0.0903319761035\n",
      "0.8814060688018799 loss -0.00137026019844\n",
      "1.2248499393463135 loss 0.0772080107003\n",
      "1.048046350479126 loss 0.0369481080435\n",
      "0.6875677108764648 loss 0.0455947313531\n",
      "1.265657901763916 loss 0.0334124726781\n",
      "0.7941527366638184 loss 0.000484192752986\n",
      "0.6798803806304932 loss -0.0\n",
      "1.7228796482086182 loss -0.0355747649438\n",
      "1.1057746410369873 loss 0.0711045157968\n",
      "1.7806882858276367 loss -8.94488927178e-05\n",
      "0.9567673206329346 loss -0.0417146318964\n",
      "1.4447898864746094 loss -0.0357176371229\n",
      "1.6102213859558105 loss -0.0482595423299\n",
      "1.7290699481964111 loss -0.0169082056544\n",
      "0.852022647857666 loss -0.0242807227571\n",
      "1.2406377792358398 loss 0.0338834593377\n",
      "0.9319663047790527 loss 0.146705722762\n",
      "1.2706272602081299 loss -0.0273949329549\n",
      "0.7375099658966064 loss -0.0222425591861\n",
      "0.7998201847076416 loss 0.038686711521\n",
      "1.1476457118988037 loss -0.00120406221701\n",
      "1.3158257007598877 loss -0.00933390173225\n",
      "1.1581008434295654 loss 0.0062051897926\n",
      "0.5744457244873047 loss 0.102184964982\n",
      "0.6817755699157715 loss -0.0\n",
      "0.5449984073638916 loss -0.0\n",
      "0.5889229774475098 loss -0.0\n",
      "0.6153061389923096 loss -0.0\n",
      "0.8628952503204346 loss -0.0\n",
      "1.0558323860168457 loss 0.0322737021256\n",
      "0.6251144409179688 loss -0.044509847741\n",
      "1.5085465908050537 loss 0.0447944331866\n",
      "0.8061904907226562 loss 0.0572947955087\n",
      "1.4444513320922852 loss -0.00361967060397\n",
      "1.4304285049438477 loss -0.0465433515149\n",
      "0.8511807918548584 loss -0.068126616735\n",
      "0.5867478847503662 loss -0.0\n",
      "1.125075101852417 loss -0.064826993097\n",
      "1.0349152088165283 loss 0.0449242270766\n",
      "0.8231236934661865 loss 0.119058832269\n",
      "0.833223819732666 loss -0.0407356201158\n",
      "1.1510579586029053 loss 0.0420082461241\n",
      "0.934943675994873 loss -0.0439431325156\n",
      "1.0907318592071533 loss -0.0407359086189\n",
      "0.8918247222900391 loss -0.0\n",
      "1.0513832569122314 loss 0.0356479391417\n",
      "1.1104772090911865 loss 0.00731015994035\n",
      "1.408228874206543 loss -0.133003640706\n",
      "0.973811149597168 loss 0.0451101496672\n",
      "1.3069138526916504 loss -0.0\n",
      "1.2364673614501953 loss 0.0377409796137\n",
      "0.851762056350708 loss 0.0598666606811\n",
      "1.345979928970337 loss 0.00653099537468\n",
      "1.1836156845092773 loss 0.0387386753699\n",
      "1.0684335231781006 loss 0.0742648052688\n",
      "1.2261989116668701 loss -0.0137099976738\n",
      "1.1661624908447266 loss -0.0329718915356\n",
      "1.3868944644927979 loss -0.0\n",
      "1.1146290302276611 loss 0.0635641726164\n",
      "1.8587803840637207 loss -0.00458912545953\n",
      "1.4971342086791992 loss -0.045020356031\n",
      "1.2714331150054932 loss -0.0428223813581\n",
      "0.7388341426849365 loss 0.224575080296\n",
      "0.9647023677825928 loss -0.0355700501987\n",
      "1.2559854984283447 loss -0.112022144028\n",
      "0.7002313137054443 loss 0.053266897675\n",
      "0.7441058158874512 loss -0.0\n",
      "0.7224020957946777 loss -0.0114653193415\n",
      "0.7976553440093994 loss 0.0086322446841\n",
      "0.6011712551116943 loss -0.0\n",
      "0.9683444499969482 loss -0.0520434570825\n",
      "0.6496193408966064 loss -0.142822452774\n",
      "0.7613770961761475 loss -0.0\n",
      "1.7269132137298584 loss 0.0128480534762\n",
      "0.8062286376953125 loss 0.113423068656\n",
      "0.9843013286590576 loss -0.0207524220712\n",
      "0.8339107036590576 loss -0.137102124277\n",
      "1.1088101863861084 loss -0.0741536011437\n",
      "0.8982605934143066 loss 0.0687672280911\n",
      "1.1768219470977783 loss 0.0623487803619\n",
      "0.8841915130615234 loss -0.0616938884593\n",
      "0.6374380588531494 loss -0.0\n",
      "0.9129416942596436 loss 0.13050777928\n",
      "1.4777328968048096 loss 0.0441109256145\n",
      "1.1985671520233154 loss 0.0651169352801\n",
      "1.0956332683563232 loss -0.00792740084713\n",
      "1.3011178970336914 loss -0.0050815734162\n",
      "1.220155954360962 loss 0.0577113170327\n",
      "1.0504302978515625 loss -0.0841245865555\n",
      "1.272538423538208 loss -0.00238716668334\n",
      "1.2349607944488525 loss -0.00642535782539\n",
      "1.2985484600067139 loss -0.00787953227911\n",
      "1.6275668144226074 loss 0.00126928085048\n",
      "1.275913953781128 loss -0.012422599875\n",
      "1.6009936332702637 loss -0.0544613236995\n",
      "1.346703290939331 loss -0.0106132370652\n",
      "0.9042720794677734 loss 0.0160205853577\n",
      "0.8895664215087891 loss 0.0231146547074\n",
      "1.2004714012145996 loss -0.0878327250163\n",
      "1.2653095722198486 loss 0.00513448186877\n",
      "0.8661878108978271 loss -0.0\n",
      "1.056656837463379 loss 0.0128588627472\n",
      "0.7781689167022705 loss -0.0846895567655\n",
      "1.06817626953125 loss 0.0147581293181\n",
      "1.1168842315673828 loss 0.0165267291918\n",
      "0.9096884727478027 loss 0.0730450476665\n",
      "1.8387911319732666 loss 0.0243226147172\n",
      "1.3280961513519287 loss 0.00252870706323\n",
      "1.1553597450256348 loss -0.00263481972505\n",
      "0.7319984436035156 loss -0.0682525031202\n",
      "1.4143457412719727 loss -0.0460114848826\n",
      "1.0314042568206787 loss -0.0\n",
      "0.8726749420166016 loss -0.105613726734\n",
      "0.9312679767608643 loss -0.0113155812799\n",
      "0.6728718280792236 loss -0.108179912582\n",
      "1.58372163772583 loss 0.0883216207531\n",
      "1.328761100769043 loss -0.0905272707416\n",
      "1.0624229907989502 loss -0.0205571175969\n",
      "0.718254566192627 loss 0.00510017737124\n",
      "1.239013433456421 loss -0.00527981373667\n",
      "0.5938320159912109 loss -0.0\n",
      "0.9526882171630859 loss 0.00729927007299\n",
      "1.1819355487823486 loss -0.0189743095742\n",
      "0.8125112056732178 loss 0.0776202482096\n",
      "0.8347675800323486 loss -0.0131490801626\n",
      "1.4646801948547363 loss 0.0322193898535\n",
      "1.2850096225738525 loss -0.00145023939453\n",
      "0.5780060291290283 loss -0.0452981254343\n",
      "1.2683050632476807 loss 0.0204236884723\n",
      "0.8873462677001953 loss 0.0649352824413\n",
      "1.4087839126586914 loss 0.0470264292919\n",
      "1.0920274257659912 loss -1.60246890532e-17\n",
      "0.7435271739959717 loss 0.0467686465901\n",
      "0.6104557514190674 loss -0.103083578431\n",
      "0.8999857902526855 loss 0.0335190282443\n",
      "0.5675115585327148 loss -0.11996255412\n",
      "0.6062302589416504 loss 0.0854981960071\n",
      "1.385622262954712 loss -0.00708545677544\n",
      "1.0545649528503418 loss -0.088311387964\n",
      "0.8331112861633301 loss -0.0742442184566\n",
      "1.4473271369934082 loss -0.0495677735251\n",
      "1.6564116477966309 loss -0.0118873146362\n",
      "0.9508056640625 loss -0.0158855295116\n",
      "1.214658498764038 loss 0.032832505387\n",
      "0.6311907768249512 loss 0.125164957183\n",
      "1.8857269287109375 loss -0.00665124682644\n",
      "1.6624643802642822 loss -0.00593407477667\n",
      "0.6244237422943115 loss 0.0564715652604\n",
      "0.7941701412200928 loss -0.00362484831885\n",
      "0.5914764404296875 loss 0.0245741136319\n",
      "1.185908317565918 loss 0.0625306069675\n",
      "0.9425904750823975 loss -0.0344190237371\n",
      "0.739734411239624 loss -0.0396124549096\n",
      "1.7436907291412354 loss 0.0207798838346\n",
      "0.6109311580657959 loss 0.0624080874038\n",
      "0.887667179107666 loss -0.0299882685992\n",
      "0.6273677349090576 loss -0.00716114874039\n",
      "1.8764030933380127 loss -0.0446068803531\n",
      "1.4935698509216309 loss -0.0329964727906\n",
      "1.646672010421753 loss 0.0282081093629\n",
      "1.359626293182373 loss 0.0827951975822\n",
      "0.9599535465240479 loss -0.0242216334248\n",
      "0.8965458869934082 loss -0.0334149074796\n",
      "0.8898842334747314 loss 0.0544915826349\n",
      "0.97625732421875 loss -0.0674166794269\n",
      "1.3371529579162598 loss -0.0190759211223\n",
      "1.0880393981933594 loss -0.0355079941504\n",
      "0.6052131652832031 loss 0.179039970747\n",
      "0.7585365772247314 loss 0.0814378114426\n",
      "1.2565340995788574 loss 0.0688105690204\n",
      "1.700885534286499 loss 0.0355831380676\n",
      "0.836794376373291 loss -0.0277279198035\n",
      "1.3266935348510742 loss 0.000616297395268\n",
      "1.0807526111602783 loss 0.025265782303\n",
      "0.6690149307250977 loss -0.0\n",
      "0.7723128795623779 loss 0.0740133816681\n",
      "0.6162757873535156 loss -0.180735758041\n",
      "0.6254394054412842 loss -0.053593307857\n",
      "0.6167583465576172 loss 0.0351643377343\n",
      "0.6260216236114502 loss 0.0114707866935\n",
      "0.6687088012695312 loss 0.0650521662018\n",
      "0.8499691486358643 loss -0.103863873375\n",
      "0.7313566207885742 loss 0.0218912342853\n",
      "0.800161600112915 loss -0.0465820628584\n",
      "0.8931105136871338 loss 0.0681799324189\n",
      "0.7244887351989746 loss -0.0474570115536\n",
      "0.9013831615447998 loss 0.0748384480339\n",
      "0.6224260330200195 loss -0.0\n",
      "0.7572798728942871 loss -0.0444908798733\n",
      "0.6181795597076416 loss -0.113423623453\n",
      "0.6235337257385254 loss 0.0742315527404\n",
      "0.5809857845306396 loss 0.130871839614\n",
      "0.7870333194732666 loss -0.161476406487\n",
      "0.8197917938232422 loss -0.113394663875\n",
      "1.0151422023773193 loss -0.0117070017336\n",
      "0.8511686325073242 loss -0.0666688374126\n",
      "0.6355428695678711 loss 0.130732203725\n",
      "0.6608762741088867 loss -0.0\n",
      "0.7700026035308838 loss 0.104364570124\n",
      "0.6694729328155518 loss 0.0685919395489\n",
      "0.7360856533050537 loss -0.11104392983\n",
      "0.6703271865844727 loss -0.128346672014\n",
      "0.8340926170349121 loss -0.00809201691481\n",
      "0.6265885829925537 loss 0.0304431282774\n",
      "0.6198718547821045 loss -0.0\n",
      "0.7517120838165283 loss -0.00963265995672\n",
      "0.6595849990844727 loss 0.141428329615\n",
      "0.9543981552124023 loss -0.0403551018643\n",
      "1.5319733619689941 loss -0.0367552427787\n",
      "0.7052953243255615 loss -0.168549965616\n",
      "1.629629135131836 loss 0.00590417602837\n",
      "1.5468635559082031 loss 0.00731889649218\n",
      "0.6960842609405518 loss 0.00971328156231\n",
      "0.708319902420044 loss 0.0579711900497\n",
      "0.884023904800415 loss -0.0804342274119\n",
      "0.7745778560638428 loss -0.017954546351\n",
      "0.9118368625640869 loss -0.0787690062096\n",
      "0.646277904510498 loss -0.187596198911\n",
      "1.584261178970337 loss -0.0167338709049\n",
      "0.8912017345428467 loss 0.110444259899\n",
      "0.6290709972381592 loss 0.0701652950995\n",
      "1.5293290615081787 loss -0.0226577075318\n",
      "1.5704498291015625 loss -0.0574706316299\n",
      "0.8637986183166504 loss 0.123965321754\n",
      "0.6768815517425537 loss 0.136438498065\n",
      "0.7258210182189941 loss 0.0681598176591\n",
      "0.5948197841644287 loss -0.0991189255567\n",
      "0.9021413326263428 loss -0.0786844364608\n",
      "0.9102156162261963 loss -0.0442993911872\n",
      "0.8312222957611084 loss 0.0451298701299\n",
      "0.7466747760772705 loss 0.0957689343611\n",
      "0.6502716541290283 loss 0.123531549007\n",
      "0.717228889465332 loss -0.169220771297\n",
      "0.5793960094451904 loss 0.0481078986736\n",
      "1.1109960079193115 loss 0.000148486093877\n",
      "0.6365237236022949 loss -0.233118821357\n",
      "0.8560211658477783 loss -0.0186989398002\n",
      "0.9205100536346436 loss -0.13897279002\n",
      "0.8603379726409912 loss -0.0404320087475\n",
      "0.6256051063537598 loss -0.115162585126\n",
      "0.6190330982208252 loss 0.0430506910359\n",
      "1.8858165740966797 loss -0.0148456234053\n",
      "0.9813737869262695 loss -0.0656532164299\n",
      "0.7410244941711426 loss 0.0819680012537\n",
      "1.7096445560455322 loss -0.0309851428862\n",
      "1.7374324798583984 loss 0.0332334752144\n",
      "0.656376838684082 loss 0.0792765188848\n",
      "0.8672158718109131 loss -0.104548750744\n",
      "0.7355263233184814 loss -0.199718596002\n",
      "1.0816049575805664 loss 0.0431473791475\n",
      "0.8540570735931396 loss -0.0195965450417\n",
      "0.6247105598449707 loss 0.0413701456934\n",
      "0.8444550037384033 loss -0.0391767729012\n",
      "1.5278067588806152 loss -0.0304681398375\n",
      "0.6412947177886963 loss 0.048113657423\n",
      "1.1438729763031006 loss -0.0883533011295\n",
      "1.5017602443695068 loss 0.0146599261893\n",
      "0.9255697727203369 loss 0.0408347947443\n",
      "0.7621235847473145 loss 0.0941497622381\n",
      "1.1891093254089355 loss -0.0229785259821\n",
      "0.5510604381561279 loss -0.0887992379444\n",
      "0.7080883979797363 loss -0.0559297435205\n",
      "1.1038293838500977 loss -0.0729324957489\n",
      "0.56815505027771 loss 0.0678844233302\n",
      "0.8690965175628662 loss -0.0614308457127\n",
      "0.8792247772216797 loss -0.0753499751802\n",
      "0.8266885280609131 loss -0.0128113911453\n",
      "0.6253194808959961 loss 0.0283325867246\n",
      "1.7609972953796387 loss 0.00983533330076\n",
      "0.6522014141082764 loss 0.00593241642211\n",
      "0.800828218460083 loss -0.039276280561\n",
      "0.9572291374206543 loss -0.0547891591642\n",
      "1.378739356994629 loss 0.000668362000551\n",
      "0.6280074119567871 loss -0.0529442046223\n",
      "0.7708072662353516 loss 0.0793127961669\n",
      "0.8982110023498535 loss -0.0862875267559\n",
      "1.5705759525299072 loss -0.021992514471\n",
      "0.819199800491333 loss 0.0568108782543\n",
      "0.6245849132537842 loss -0.0931641541445\n",
      "1.3084261417388916 loss 0.0531191896859\n",
      "0.6763179302215576 loss -0.0222979633307\n",
      "0.8855314254760742 loss 0.0118035705299\n",
      "0.7748112678527832 loss -0.0857415087317\n",
      "0.6131620407104492 loss 0.0718491834491\n",
      "0.6947085857391357 loss 0.142986205436\n",
      "1.4035253524780273 loss 0.0278423838478\n",
      "1.3024461269378662 loss -0.0232436337532\n",
      "2.032801628112793 loss -0.0184197540386\n",
      "1.4866788387298584 loss -0.0168346544832\n",
      "1.1261968612670898 loss 0.0177719582828\n",
      "0.9024298191070557 loss 0.108811246623\n",
      "0.6977088451385498 loss 0.0766716538893\n",
      "1.400850534439087 loss 0.073984726055\n",
      "1.0872952938079834 loss 0.0604425780514\n",
      "1.2858600616455078 loss -0.0452753068883\n",
      "0.8975028991699219 loss 0.0123638030995\n",
      "0.9381847381591797 loss -0.0111462678634\n",
      "0.9077610969543457 loss -0.0225721650111\n",
      "1.6980910301208496 loss 0.0236090386501\n",
      "1.530531644821167 loss -0.0258748978206\n",
      "1.7171704769134521 loss -0.0633460366171\n",
      "1.268232822418213 loss -0.0470247966956\n",
      "0.7107362747192383 loss -0.00555020492776\n",
      "0.8781700134277344 loss 0.097395621521\n",
      "0.805680513381958 loss -0.0517612358466\n",
      "1.4493155479431152 loss -0.0844747324687\n",
      "0.9556694030761719 loss 0.0147981899731\n",
      "1.7127974033355713 loss -0.0356169997161\n",
      "1.9584522247314453 loss -0.0352629781421\n",
      "0.9015700817108154 loss -0.0924539446912\n",
      "1.9149584770202637 loss -0.00794953353552\n",
      "0.9381206035614014 loss 0.0366596089566\n",
      "1.1401758193969727 loss 0.0990619745995\n",
      "1.008387565612793 loss -0.00973608309437\n",
      "1.2541706562042236 loss 0.0164519085609\n",
      "0.9614682197570801 loss -0.0607014425394\n",
      "0.7869200706481934 loss -0.108392274195\n",
      "0.6666150093078613 loss 0.0658401021775\n",
      "0.9282586574554443 loss -0.0\n",
      "0.6385796070098877 loss 0.0905976508334\n",
      "1.022139072418213 loss -0.0631913732539\n",
      "0.7731952667236328 loss -0.095428402398\n",
      "0.7420549392700195 loss 0.0644743149451\n",
      "0.8499624729156494 loss -0.00863662202512\n",
      "0.945167064666748 loss -0.0915883050779\n",
      "0.7280013561248779 loss 0.173384657924\n",
      "0.6662602424621582 loss -0.0838220874549\n",
      "0.8279883861541748 loss -0.110806669363\n",
      "0.6429991722106934 loss 0.172613366514\n",
      "0.62784743309021 loss 0.111369209278\n",
      "0.6359422206878662 loss 0.0292838933271\n",
      "0.8831648826599121 loss -0.0743029690614\n",
      "0.9263384342193604 loss 0.0465728665172\n",
      "0.7317500114440918 loss 0.206999386669\n",
      "0.6702392101287842 loss 0.100503781526\n",
      "0.6894106864929199 loss -0.0469659346822\n",
      "0.7816047668457031 loss 0.0786055796407\n",
      "0.8380076885223389 loss 0.0479557022512\n",
      "0.9627280235290527 loss 0.0512890975952\n",
      "0.7734911441802979 loss 0.0064619350018\n",
      "0.642258882522583 loss -0.0774984950799\n",
      "0.6575782299041748 loss -0.0874386600619\n",
      "0.6508216857910156 loss 0.122304469149\n",
      "0.8262345790863037 loss -0.00514048262303\n",
      "0.8728644847869873 loss 0.0455647283263\n",
      "0.880244255065918 loss 0.0538965197898\n",
      "1.0289990901947021 loss -0.0188870935345\n",
      "0.8954801559448242 loss 0.0105851224805\n",
      "0.858149528503418 loss 0.0350851617911\n",
      "0.6639113426208496 loss 0.0835343790618\n",
      "1.2273271083831787 loss 0.0845653963489\n",
      "0.8824150562286377 loss -0.0665190105238\n",
      "0.731224775314331 loss -0.0\n",
      "0.6813962459564209 loss 0.0559540889527\n",
      "0.7975661754608154 loss -0.00405993046135\n",
      "0.701866865158081 loss -0.0276435920325\n",
      "1.8514676094055176 loss 0.036090434614\n",
      "0.6405460834503174 loss -0.0543358164785\n",
      "0.9079174995422363 loss -0.0017160048889\n",
      "1.80204176902771 loss -0.016489285924\n",
      "0.9552202224731445 loss 0.00979535101076\n",
      "0.6773080825805664 loss -0.102069430827\n",
      "0.9342315196990967 loss 0.0646406088921\n",
      "0.9697771072387695 loss 0.016618726161\n",
      "1.611084222793579 loss 0.00517213090188\n",
      "1.6213562488555908 loss -0.013585704183\n",
      "1.1365206241607666 loss -0.0100462342483\n",
      "0.8340551853179932 loss 0.0289923588024\n",
      "0.6929271221160889 loss -0.0\n",
      "1.0260512828826904 loss -0.101804226339\n",
      "1.024073839187622 loss 0.00557444691534\n",
      "0.7377381324768066 loss -0.0530877773387\n",
      "0.653372049331665 loss -0.0657081503209\n",
      "0.8541033267974854 loss 0.0580624255522\n",
      "1.1920719146728516 loss -0.0431580672192\n",
      "0.651944637298584 loss 0.170251306152\n",
      "0.6400363445281982 loss 0.0105809738923\n",
      "0.6383473873138428 loss -0.0299876326518\n",
      "0.6522893905639648 loss 0.302127659574\n",
      "0.6481778621673584 loss 0.0590052943236\n",
      "0.9067423343658447 loss 0.0027822750468\n",
      "0.775482177734375 loss -0.0475876221239\n",
      "0.6312267780303955 loss -0.107009647187\n",
      "1.321681022644043 loss 0.00639598931788\n",
      "0.6289224624633789 loss 0.0592852361809\n",
      "0.5913906097412109 loss 0.180402691064\n",
      "0.7777912616729736 loss -0.0730753607159\n",
      "0.752708911895752 loss 0.0627761130112\n",
      "0.5788469314575195 loss -1.60246890532e-17\n",
      "1.5205533504486084 loss -0.0500376807359\n",
      "0.578894853591919 loss 0.0188578571601\n",
      "0.9674246311187744 loss -0.0200392399453\n",
      "0.6230981349945068 loss 0.303613707719\n",
      "0.59116530418396 loss 0.0437521893592\n",
      "0.582390308380127 loss 0.0457229956859\n",
      "0.6309168338775635 loss -0.0499501712931\n",
      "0.6317458152770996 loss -0.0944911182523\n",
      "0.7944545745849609 loss -0.0274941168884\n",
      "0.775275468826294 loss 0.0101144347485\n",
      "1.0592544078826904 loss 0.0158669597815\n",
      "0.6439845561981201 loss -0.0658506533157\n",
      "0.562361478805542 loss 0.217327693928\n",
      "1.253655195236206 loss -0.038904929107\n",
      "0.7457976341247559 loss 0.0155824498335\n",
      "0.57781982421875 loss -0.144620305212\n",
      "0.5554769039154053 loss -0.0361550763031\n",
      "0.5573074817657471 loss 0.0332844215926\n",
      "0.588477373123169 loss 0.0421370524277\n",
      "1.2520477771759033 loss -0.0116694603048\n",
      "0.7681684494018555 loss 0.002720052225\n",
      "0.6435463428497314 loss 0.0434756752368\n",
      "1.4845929145812988 loss -0.0070042843007\n",
      "0.5805013179779053 loss 0.0744331447696\n",
      "0.5346319675445557 loss -0.0857492925713\n",
      "0.6894063949584961 loss 0.00318002335489\n",
      "0.6249356269836426 loss -0.0575136373397\n",
      "0.9834654331207275 loss -0.036767126545\n",
      "0.7947084903717041 loss 0.00618589574132\n",
      "0.760753870010376 loss 0.122801485828\n",
      "1.0514411926269531 loss 0.0780425688565\n",
      "1.5640554428100586 loss -0.0445155860716\n",
      "0.5823814868927002 loss -0.0498668594716\n",
      "0.6332778930664062 loss 0.311988770318\n",
      "0.7470483779907227 loss -0.0760181321946\n",
      "1.1045465469360352 loss -0.0283446911891\n",
      "0.9331920146942139 loss 0.0551250478344\n",
      "0.6408722400665283 loss 0.106425587458\n",
      "0.5545661449432373 loss 0.108229153499\n",
      "0.7879197597503662 loss 0.0568956642938\n",
      "1.0623464584350586 loss 0.0348424650786\n",
      "0.7667465209960938 loss 0.129287900336\n",
      "0.7702038288116455 loss -0.0911872197824\n",
      "0.9906303882598877 loss 0.00214785550583\n",
      "0.8700385093688965 loss 0.0393050016684\n",
      "0.7068824768066406 loss -0.0496370285719\n",
      "0.5875177383422852 loss -0.142137001445\n",
      "1.2461488246917725 loss -0.0678187475658\n",
      "0.7462866306304932 loss -0.0279727119432\n",
      "0.7872388362884521 loss -0.0462938663889\n",
      "0.6521944999694824 loss -0.0770669989118\n",
      "1.2595634460449219 loss -0.0112229366084\n",
      "0.5874342918395996 loss 0.0682230446225\n",
      "0.6292843818664551 loss 0.0583657677719\n",
      "1.277355670928955 loss -0.0631339165885\n",
      "1.148529052734375 loss -0.00296146856885\n",
      "0.5969994068145752 loss 0.060909634902\n",
      "0.6432943344116211 loss -0.079549853583\n",
      "0.7745583057403564 loss 0.185245829028\n",
      "1.4837536811828613 loss 0.0529171469799\n",
      "0.8413481712341309 loss -0.0521753038836\n",
      "1.059098243713379 loss -0.0318456152845\n",
      "0.6281008720397949 loss -0.106963490396\n",
      "0.996645450592041 loss 0.0575346284415\n",
      "1.0245304107666016 loss -0.0496116294367\n",
      "1.118166446685791 loss -0.023855406116\n",
      "0.5864231586456299 loss -0.0605227532669\n",
      "1.0653481483459473 loss -0.0855986149253\n",
      "0.9454448223114014 loss -0.00674453174082\n",
      "0.6170730590820312 loss -0.0706381021135\n",
      "0.6102433204650879 loss 0.132630194016\n",
      "0.9108474254608154 loss 0.0552245651609\n",
      "0.6746346950531006 loss -0.210666210154\n",
      "1.4615600109100342 loss -0.0200369837852\n",
      "1.0239942073822021 loss -0.00313225787277\n",
      "1.0438008308410645 loss -0.0421238812825\n",
      "0.5410184860229492 loss -0.181585677749\n",
      "0.761336088180542 loss 0.0013059599783\n",
      "0.8836100101470947 loss -0.0364364430081\n",
      "0.6854104995727539 loss 0.0504450361794\n",
      "0.5763771533966064 loss -0.0813073664672\n",
      "0.6647803783416748 loss -0.00151367539367\n",
      "0.7726585865020752 loss -0.0357937953105\n",
      "1.3726677894592285 loss 0.0423002592506\n",
      "1.5556321144104004 loss 0.0300687482107\n",
      "0.6134412288665771 loss -0.0353113495418\n",
      "1.2117252349853516 loss 0.054474312977\n",
      "0.607586145401001 loss -0.204228737128\n",
      "0.7913341522216797 loss -0.0460906862107\n",
      "0.9321341514587402 loss -0.0796652438496\n",
      "1.2447319030761719 loss 0.0333386164246\n",
      "1.2216675281524658 loss -0.0306117788978\n",
      "1.4633495807647705 loss -0.00508977980495\n",
      "0.7652065753936768 loss 0.00519470685211\n",
      "0.5909926891326904 loss 0.146457475187\n",
      "0.557349681854248 loss -0.112151516164\n",
      "0.6229147911071777 loss 0.136232386642\n",
      "0.6290717124938965 loss 0.0315834311811\n",
      "0.7997381687164307 loss 0.058867596215\n",
      "1.1745498180389404 loss -0.00709220757377\n",
      "0.6940145492553711 loss -0.0373384068826\n",
      "0.5802969932556152 loss 0.166692321507\n",
      "0.6537356376647949 loss -0.101929873451\n",
      "0.574887752532959 loss -0.0519498260999\n",
      "0.8989531993865967 loss 0.0548703261169\n",
      "1.5388939380645752 loss -0.0273620159321\n",
      "1.0285229682922363 loss 0.0685455926381\n",
      "1.3521907329559326 loss 0.0243822350642\n",
      "0.7559089660644531 loss 0.0573214459878\n",
      "0.8020765781402588 loss 0.0907666273747\n",
      "0.6422033309936523 loss 0.0152480841033\n",
      "0.6526174545288086 loss -0.058333513109\n",
      "0.6239335536956787 loss -0.0111120988971\n",
      "0.8268945217132568 loss 0.0813569895212\n",
      "0.6901936531066895 loss 0.128523713932\n",
      "0.754847526550293 loss 0.0196321768579\n",
      "0.6329681873321533 loss -0.0386772872244\n",
      "0.6663472652435303 loss -0.0669418969014\n",
      "0.8991613388061523 loss 0.0452833412852\n",
      "0.6060805320739746 loss 0.0832313360027\n",
      "0.9528865814208984 loss 0.0461894996228\n",
      "0.8258070945739746 loss 0.0828765163423\n",
      "0.661505937576294 loss -0.0185927893482\n",
      "0.7051324844360352 loss -0.0440270997284\n",
      "0.6298959255218506 loss 0.117490927899\n",
      "0.6098484992980957 loss -0.118425484673\n",
      "1.023961067199707 loss 0.0333135506827\n",
      "0.8071699142456055 loss -0.121333508155\n",
      "0.8822462558746338 loss -0.164758616987\n",
      "0.6529858112335205 loss -0.0334423867969\n",
      "0.8422563076019287 loss -0.126045677661\n",
      "0.9272317886352539 loss -0.032686022523\n",
      "0.8018016815185547 loss -0.055335580164\n",
      "0.6795642375946045 loss -0.0277350098113\n",
      "0.7768363952636719 loss -0.0210072907951\n",
      "0.9000840187072754 loss -0.0258652545379\n",
      "0.6454212665557861 loss -0.0100627524433\n",
      "0.6493780612945557 loss -0.0777082995277\n",
      "0.673459529876709 loss 0.0273684210526\n",
      "1.043868064880371 loss 0.0630231445263\n",
      "0.7238588333129883 loss -0.121620847596\n",
      "0.8839395046234131 loss 0.0443472683634\n",
      "0.6500120162963867 loss -0.0251497727414\n",
      "0.5864906311035156 loss -0.0354663451066\n",
      "0.6185610294342041 loss 0.0810892854428\n",
      "0.9187667369842529 loss 0.0882547343485\n",
      "0.7178311347961426 loss -0.100572929358\n",
      "0.9058132171630859 loss -0.0475033192803\n",
      "0.763094425201416 loss 0.155806081034\n",
      "0.7611026763916016 loss 0.085688827673\n",
      "0.6463911533355713 loss -0.13744256927\n",
      "0.6156811714172363 loss -0.212471698265\n",
      "1.16823410987854 loss -0.0835472916842\n",
      "0.7528781890869141 loss 0.0643625417468\n",
      "0.8762490749359131 loss -0.0427002096023\n",
      "0.7986154556274414 loss -0.0165597186854\n",
      "0.9645633697509766 loss -0.086197853396\n",
      "0.6325094699859619 loss 0.0543245581884\n",
      "0.6325311660766602 loss -0.00379868588199\n",
      "0.5586802959442139 loss -0.0482674745967\n",
      "0.990868091583252 loss -0.0324454108225\n",
      "0.9625711441040039 loss 0.0324506014469\n",
      "0.7972569465637207 loss 0.00397957259196\n",
      "0.8801615238189697 loss 0.00416694577636\n",
      "0.6399970054626465 loss 0.12770236106\n",
      "0.749011754989624 loss 0.077330889334\n",
      "0.6381497383117676 loss 0.129911175653\n",
      "0.7351274490356445 loss 0.0446656139974\n",
      "0.5912165641784668 loss -0.0871842563633\n",
      "1.5234887599945068 loss -0.0449539044901\n",
      "1.0675709247589111 loss -0.0409221129574\n",
      "1.0363988876342773 loss 0.0107227796253\n",
      "0.9509241580963135 loss -0.051573896202\n",
      "0.746286153793335 loss -0.10721425275\n",
      "0.980252742767334 loss 0.0196651409706\n",
      "0.9359958171844482 loss 0.0889831629145\n",
      "0.6601150035858154 loss 0.0360466564804\n",
      "0.8689699172973633 loss 0.0103813749558\n",
      "0.6736986637115479 loss -0.0807734416128\n",
      "1.1655662059783936 loss 0.0239875689713\n",
      "0.7353265285491943 loss -0.0506033763699\n",
      "1.1179301738739014 loss 0.0600905207343\n",
      "0.6362252235412598 loss -0.11266715931\n",
      "0.6147654056549072 loss -0.0987313844457\n",
      "0.6179461479187012 loss -0.0563368285019\n",
      "0.7706248760223389 loss 0.0940030770434\n",
      "0.6012580394744873 loss 0.0872093253557\n",
      "0.937528133392334 loss -0.0521916750479\n",
      "1.03812837600708 loss -0.0320539164393\n",
      "0.7304239273071289 loss 0.142857142857\n",
      "0.7828834056854248 loss -0.151206300238\n",
      "1.6873955726623535 loss 0.0447644577695\n",
      "0.7529640197753906 loss 0.097477393325\n",
      "0.7708570957183838 loss -0.190362444208\n",
      "1.2396845817565918 loss 0.0417720727562\n",
      "0.6172645092010498 loss -0.104012823225\n",
      "1.5226421356201172 loss -0.0270103916739\n",
      "1.7090611457824707 loss 0.0526959905428\n",
      "0.5879125595092773 loss -0.155485768403\n",
      "0.9088435173034668 loss 0.0457210403795\n",
      "0.64276123046875 loss 0.101993569975\n",
      "0.9412682056427002 loss 0.0378679996675\n",
      "0.8087723255157471 loss -0.0231292517007\n",
      "0.719346284866333 loss 0.0462552766256\n",
      "1.0338966846466064 loss 0.00975827319355\n",
      "0.7078244686126709 loss -0.186200499939\n",
      "1.062319040298462 loss -0.0206078657013\n",
      "0.7408356666564941 loss -0.0437521893592\n",
      "1.5988059043884277 loss 0.0117743977157\n",
      "0.675961971282959 loss 0.0912328360967\n",
      "0.6257247924804688 loss -0.224679258568\n",
      "0.8592367172241211 loss -0.0373035978037\n",
      "0.6293036937713623 loss -0.0579711900497\n",
      "0.84751296043396 loss -0.00480375223519\n",
      "0.8877859115600586 loss -0.0464027870135\n",
      "1.5752992630004883 loss -0.0486208525992\n",
      "0.6851105690002441 loss 0.0906702160439\n",
      "0.7404325008392334 loss -0.0123973581096\n",
      "1.216672658920288 loss -0.00280678134254\n",
      "0.6571757793426514 loss 0.0227711918929\n",
      "1.6786916255950928 loss 0.0146783541025\n",
      "1.0877082347869873 loss -0.0133235690924\n",
      "0.670750617980957 loss -0.038244005432\n",
      "0.6494946479797363 loss -0.060888137589\n",
      "1.068873643875122 loss 0.0771762246221\n",
      "0.9481017589569092 loss -0.0284364437304\n",
      "0.8589894771575928 loss -0.00719924569201\n",
      "1.3119707107543945 loss -0.079728097053\n",
      "0.8035919666290283 loss -0.0859404840964\n",
      "0.7692561149597168 loss -0.0\n",
      "0.7164194583892822 loss -0.0293464486921\n",
      "1.0704429149627686 loss 0.0272057589027\n",
      "1.0659725666046143 loss 0.00212219062016\n",
      "1.0925607681274414 loss -0.0575777640408\n",
      "0.633458137512207 loss 0.3649002246\n",
      "0.5341339111328125 loss -0.185312329165\n",
      "0.6607904434204102 loss 0.374802530691\n",
      "0.585374116897583 loss -0.246414303882\n",
      "1.6730096340179443 loss -0.0110726882232\n",
      "0.6627650260925293 loss 0.306344368452\n",
      "0.9368484020233154 loss 0.104032008888\n",
      "1.1048731803894043 loss -0.0525290538176\n",
      "0.7365634441375732 loss -0.108490319178\n",
      "1.1061677932739258 loss 0.118395663198\n",
      "0.6588184833526611 loss 0.0347284611304\n",
      "1.0130512714385986 loss -0.0374253074585\n",
      "0.6443796157836914 loss 0.0172589633712\n",
      "0.7446939945220947 loss -0.107992126233\n",
      "0.6458868980407715 loss 0.0121640460025\n",
      "0.6421661376953125 loss -0.077151674981\n",
      "0.6493241786956787 loss 0.0213691071826\n",
      "0.6081006526947021 loss -0.031714434561\n",
      "1.0347607135772705 loss -0.0364068966322\n",
      "0.919513463973999 loss 0.0451706817648\n",
      "1.2426159381866455 loss -0.050075116027\n",
      "0.5953328609466553 loss 0.0220153123503\n",
      "0.7839562892913818 loss 0.0443690892084\n",
      "0.7439670562744141 loss -0.0194779638102\n",
      "1.160048246383667 loss 0.0779752772916\n",
      "0.7952132225036621 loss -0.0975048816307\n",
      "0.6398415565490723 loss 0.166428022045\n",
      "1.0095622539520264 loss 0.0286181903516\n",
      "0.8327603340148926 loss -0.124013537505\n",
      "0.5919032096862793 loss -0.151716521227\n",
      "0.7065796852111816 loss 0.083266012835\n",
      "0.7780132293701172 loss 0.105616112055\n",
      "0.6554219722747803 loss 0.10334831938\n",
      "1.1217820644378662 loss -0.0164419061461\n",
      "0.7280950546264648 loss 0.0798577553264\n",
      "0.592090368270874 loss -0.027397312735\n",
      "0.8999931812286377 loss -0.0442106274075\n",
      "1.4660577774047852 loss -0.0156702739163\n",
      "0.6114017963409424 loss -0.133194661082\n",
      "0.7792131900787354 loss -0.0588887711115\n",
      "0.6487267017364502 loss -0.0\n",
      "0.8974058628082275 loss -0.0478347090515\n",
      "0.986119270324707 loss 0.068199433947\n",
      "1.400038719177246 loss -0.0249283122311\n",
      "0.6275384426116943 loss -0.0\n",
      "1.2837417125701904 loss -0.0267713760679\n",
      "0.6274905204772949 loss -0.166465773667\n",
      "0.6300022602081299 loss -0.115470053838\n",
      "0.8254199028015137 loss -0.0817009225206\n",
      "1.4049155712127686 loss 0.0506966935195\n",
      "0.7748146057128906 loss 0.0576964019587\n",
      "0.6700904369354248 loss 0.0170374810924\n",
      "1.3036470413208008 loss -0.0274112755504\n",
      "0.7242529392242432 loss -0.0292177138609\n",
      "1.0829555988311768 loss 0.0794212798307\n",
      "0.745380163192749 loss 0.0517087689995\n",
      "1.0005993843078613 loss 0.161840641137\n",
      "0.9408903121948242 loss -0.112264028803\n",
      "1.0498299598693848 loss 0.05859899439\n",
      "0.7961673736572266 loss -0.109534313572\n",
      "0.6803662776947021 loss -0.057728176915\n",
      "0.8751351833343506 loss -0.0630447649664\n",
      "0.6034634113311768 loss -0.0\n",
      "1.011850357055664 loss 0.025994340627\n",
      "1.7557387351989746 loss 0.023221639858\n",
      "1.338268518447876 loss 0.0990475458312\n",
      "0.7080097198486328 loss 0.00858927775688\n",
      "0.6837539672851562 loss -0.0443888015143\n",
      "0.9141378402709961 loss -0.00246837671067\n",
      "1.066162347793579 loss -0.132513661202\n",
      "1.394887924194336 loss 0.0589160760744\n",
      "1.0494790077209473 loss -0.0857402867116\n",
      "1.22873854637146 loss 0.0702057417425\n",
      "1.1752047538757324 loss -0.0269280678504\n",
      "0.7011222839355469 loss -0.113376632523\n",
      "1.6502516269683838 loss -0.078303540841\n",
      "0.6553905010223389 loss -0.0\n",
      "1.0274648666381836 loss -0.0546268530538\n",
      "1.136338710784912 loss 0.050202030123\n",
      "0.8146355152130127 loss -0.100516976244\n",
      "0.9554874897003174 loss 0.0332608922434\n",
      "0.7001667022705078 loss -0.093250480824\n",
      "0.6544435024261475 loss -0.0422637165317\n",
      "0.7040610313415527 loss 0.064626266662\n",
      "0.6833798885345459 loss -0.145521375022\n",
      "0.7051928043365479 loss -0.148522131447\n",
      "0.7492256164550781 loss 0.0994512269648\n",
      "0.6805000305175781 loss 0.00312280504909\n",
      "0.7440974712371826 loss 0.0439740632729\n",
      "0.6553351879119873 loss 0.0534471969013\n",
      "0.635404109954834 loss -0.209275398864\n",
      "1.2562882900238037 loss 0.0768984459845\n",
      "0.717921257019043 loss -0.118425484673\n",
      "0.7718939781188965 loss 0.0823356983752\n",
      "0.6315338611602783 loss -0.0\n",
      "0.639629602432251 loss -0.0164765540686\n",
      "0.6248371601104736 loss 0.111472953324\n",
      "0.6756877899169922 loss -0.0735786241\n",
      "1.145615577697754 loss -0.0266729176242\n",
      "0.7499613761901855 loss -0.110096376513\n",
      "0.6001913547515869 loss 0.272063703362\n",
      "0.6268424987792969 loss 0.161233878133\n",
      "0.623772382736206 loss 0.0145973735849\n",
      "0.769390344619751 loss -0.151214669382\n",
      "0.7170593738555908 loss 0.0432198439543\n",
      "0.836777925491333 loss -0.0530831893243\n",
      "0.680433988571167 loss -0.139686059154\n",
      "0.5888338088989258 loss -0.0859869362183\n",
      "0.7348072528839111 loss 0.00848555291628\n",
      "0.6514322757720947 loss -0.21146027323\n",
      "0.7759785652160645 loss 0.0344854698225\n",
      "0.6151797771453857 loss 0.0771852276267\n",
      "0.6240737438201904 loss 0.00706439190482\n",
      "0.5938582420349121 loss -0.246753246753\n",
      "0.6287293434143066 loss -0.0916169659172\n",
      "0.721463680267334 loss -0.0379671840329\n",
      "0.6347146034240723 loss 0.0379274579097\n",
      "0.7941563129425049 loss -0.0693894516546\n",
      "0.6449282169342041 loss 0.0176455324594\n",
      "0.7102267742156982 loss -0.0301663738494\n",
      "0.649193286895752 loss 0.231319028645\n",
      "0.8076426982879639 loss -0.0515152545245\n",
      "0.7022056579589844 loss 0.0295927660447\n",
      "0.6741766929626465 loss -0.00205568194581\n",
      "0.9006855487823486 loss 0.00976475175785\n",
      "0.5619826316833496 loss -0.134323635121\n",
      "1.0238103866577148 loss -0.103447736255\n",
      "1.0333888530731201 loss 0.0347302730384\n",
      "0.8595118522644043 loss 0.0261471861472\n",
      "1.1345360279083252 loss 0.0704120357821\n",
      "0.6375558376312256 loss -0.000693000693001\n",
      "0.7497014999389648 loss -0.102987301918\n",
      "0.6595184803009033 loss 0.160006817004\n",
      "0.7520427703857422 loss 0.0521331166179\n",
      "0.7577884197235107 loss -0.0693375245282\n",
      "0.7334339618682861 loss -0.0171969617716\n",
      "1.03696870803833 loss 0.0120955675812\n",
      "0.9062273502349854 loss 0.0415340227693\n",
      "0.5638659000396729 loss -0.116774841624\n",
      "0.7682416439056396 loss -0.0\n",
      "0.8964972496032715 loss -0.0463763541093\n",
      "0.6407692432403564 loss -0.117030570702\n",
      "0.6255383491516113 loss 0.23036964956\n",
      "0.6469650268554688 loss 0.212382500203\n",
      "0.6516819000244141 loss 0.155808450494\n",
      "0.576751708984375 loss 0.0373739976015\n",
      "0.6991660594940186 loss -0.0510809444239\n",
      "1.0338146686553955 loss -0.0995735067501\n",
      "0.8403158187866211 loss -0.0779473896258\n",
      "0.6381821632385254 loss -0.0591777081637\n",
      "0.6832592487335205 loss -0.0421533610677\n",
      "0.6138901710510254 loss 0.0452833412852\n",
      "0.6098330020904541 loss -0.0126533853952\n",
      "0.9266462326049805 loss -0.0324571947271\n",
      "0.8146471977233887 loss -0.0927404027752\n",
      "0.7392101287841797 loss -0.254807692308\n",
      "1.210402250289917 loss -0.158179259849\n",
      "1.0808777809143066 loss 0.0449079879068\n",
      "1.1874070167541504 loss 0.0551468696869\n",
      "0.8618292808532715 loss 0.112296339872\n",
      "0.8307287693023682 loss -0.089436817404\n",
      "0.5953612327575684 loss -0.164319810239\n",
      "1.3439083099365234 loss -0.0581283091233\n",
      "0.656125545501709 loss -0.134265939713\n",
      "0.8931381702423096 loss -0.00543518320687\n",
      "0.8720066547393799 loss -0.0502374933566\n",
      "1.0884425640106201 loss -0.0415105160551\n",
      "0.6721384525299072 loss -0.0404366081933\n",
      "0.598330020904541 loss -0.100342662341\n",
      "0.7014987468719482 loss -0.0\n",
      "0.7393374443054199 loss 0.0214714206574\n",
      "1.031463384628296 loss 0.0106273863205\n",
      "0.6237835884094238 loss 0.0721382259167\n",
      "0.7311820983886719 loss 0.0282524086725\n",
      "0.8257994651794434 loss -0.00726946449676\n",
      "0.6031653881072998 loss -0.151804020962\n",
      "1.0147936344146729 loss -0.0698111713569\n",
      "0.76639723777771 loss -0.0385987768284\n",
      "0.6941409111022949 loss -0.0612528648016\n",
      "0.958491325378418 loss 0.0490778821875\n",
      "0.5955715179443359 loss -0.0363136519601\n",
      "0.5636863708496094 loss -0.0920216361679\n",
      "1.2446448802947998 loss -0.0698890168122\n",
      "0.5374510288238525 loss -0.0227634376287\n",
      "0.6879160404205322 loss 0.106395940703\n",
      "0.7729313373565674 loss 0.143922811122\n",
      "0.5659761428833008 loss -0.00294408548932\n",
      "1.3452141284942627 loss -0.0184478884365\n",
      "0.5716843605041504 loss -0.0564958573541\n",
      "0.7642364501953125 loss 0.0101917673579\n",
      "0.7226831912994385 loss -0.0577091475472\n",
      "0.7755942344665527 loss 0.0398239351288\n",
      "1.976773977279663 loss -0.0552052826941\n",
      "0.6990842819213867 loss -0.0979202913764\n",
      "0.6439402103424072 loss 0.0606908086922\n",
      "1.5841424465179443 loss 0.0434439835331\n",
      "0.6753911972045898 loss -0.133543631652\n",
      "1.0843360424041748 loss 0.0685606663602\n",
      "0.7684245109558105 loss -0.114707866935\n",
      "0.9685087203979492 loss 0.0333767234878\n",
      "0.7372679710388184 loss -0.021868700842\n",
      "0.706218957901001 loss -0.0732223313627\n",
      "0.8422489166259766 loss 0.0442632294458\n",
      "0.6345739364624023 loss 0.0361079545676\n",
      "0.8797464370727539 loss -0.0162641860342\n",
      "0.6848366260528564 loss 0.0282934943341\n",
      "0.7964053153991699 loss -0.0159725220579\n",
      "0.759530782699585 loss -0.0537440666467\n",
      "0.6414358615875244 loss -0.228217732294\n",
      "0.9221184253692627 loss 0.0826271650047\n",
      "0.6078331470489502 loss -0.0574887051656\n",
      "0.5720124244689941 loss 0.00560560661664\n",
      "0.7082974910736084 loss -0.0520307367094\n",
      "0.6930725574493408 loss 0.0778961947682\n",
      "0.6068668365478516 loss -0.178276062434\n",
      "0.8428647518157959 loss -0.0558778585944\n"
     ]
    }
   ],
   "source": [
    "best = fmin(f_dt, space, algo=tpe.suggest, max_evals=1000, trials=trials, verbose=1)\n",
    "print ('best: ')\n",
    "print (best)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T04:27:39.133867Z",
     "start_time": "2017-03-26T12:27:39.130587+08:00"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "564"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(trials.trials)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-04-26T03:02:23.990313Z",
     "start_time": "2017-04-26T11:02:23.859646+08:00"
    }
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'Path' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-1-4a286114c1ae>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mbson\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mjson_util\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      2\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0msave_path\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mPath\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'../output/hyperopt/{}.json'\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrials\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_exp_key\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      4\u001b[0m \u001b[0msave_path\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdirname\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmakedirs_p\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdump\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrials\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrials\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msave_path\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m'w'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdefault\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mjson_util\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdefault\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'Path' is not defined"
     ]
    }
   ],
   "source": [
    "from bson import json_util\n",
    "import json\n",
    "save_path = Path('../output/hyperopt/{}.json'.format(trials._exp_key))\n",
    "save_path.dirname().makedirs_p()\n",
    "json.dump(trials.trials,open(save_path,'w'), default=json_util.default)\n",
    "save_path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-04-26T03:03:01.085510Z",
     "start_time": "2017-04-26T11:03:01.079704+08:00"
    }
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'save_path' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-4-18bff864bd4d>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# load\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mtrial_data\u001b[0m \u001b[0;34m=\u001b[0m\u001b[0mjson\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mload\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msave_path\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m \u001b[0;31m# trial_data =json.load(open(save_path))\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrial_data\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'save_path' is not defined"
     ]
    }
   ],
   "source": [
    "# load\n",
    "trial_data =json.load(open(save_path))\n",
    "# trial_data =json.load(open(save_path))\n",
    "len(trial_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:13:37.293369Z",
     "start_time": "2017-03-26T13:13:36.965767+08:00"
    }
   },
   "outputs": [],
   "source": [
    "# convert to df\n",
    "\n",
    "\n",
    "def flatten(d, parent_key='', sep='_'):\n",
    "    \"\"\"Flatten dicts.\"\"\"\n",
    "    items = []\n",
    "    for k, v in d.items():\n",
    "        new_key = parent_key + sep + k if parent_key else k\n",
    "        if isinstance(v, collections.MutableMapping):\n",
    "            items.extend(flatten(v, new_key, sep=sep).items())\n",
    "        else:\n",
    "            # also flatten lists with one element\n",
    "            if isinstance(v,list) and len(v)==1:\n",
    "                v=v[0]\n",
    "            items.append((new_key, v))\n",
    "    return dict(items)\n",
    "# df = pd.DataFrame([flatten(trial) for trial in trials.trials])\n",
    "df = pd.DataFrame([flatten(trial) for trial in trial_data])\n",
    "# df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-04-26T03:02:36.726770Z",
     "start_time": "2017-04-26T11:02:36.710408+08:00"
    }
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'df' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-3-b47b94b6e60a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mmetric\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'result_metrics_report_f1-score_leak'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;31m# metric='result_metrics_matthews_corrcoef'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 5\u001b[0;31m whitelist = [col for col in df.columns if ('misc_vals' in col) or (metric in col)]+[\n\u001b[0m\u001b[1;32m      6\u001b[0m     \u001b[0;34m'result_dummy_metrics_report_dummy_f1-score_leak'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m     \u001b[0;34m'result_dummy_metrics_report_dummy_support_leak'\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'df' is not defined"
     ]
    }
   ],
   "source": [
    "# lets make a table of major results to record\n",
    "mean_of=50\n",
    "metric='result_metrics_report_f1-score_leak'\n",
    "# metric='result_metrics_matthews_corrcoef'\n",
    "whitelist = [col for col in df.columns if ('misc_vals' in col) or (metric in col)]+[\n",
    "    'result_dummy_metrics_report_dummy_f1-score_leak',\n",
    "    'result_dummy_metrics_report_dummy_support_leak',\n",
    "    \n",
    "#     'result_metrics_report_f1-score_leak',\n",
    "#     'result_metrics_matthews_corrcoef'\n",
    "]\n",
    "df50=df.sort_values(metric, ascending=False)[:mean_of][whitelist]\n",
    "\n",
    "# convert to days\n",
    "df50['misc_vals_timespan_before']=df50['misc_vals_timespan_before']/seconds_in_a_day\n",
    "\n",
    "# columns\n",
    "mean=df50.mean()\n",
    "std=df50.std()\n",
    "corr=df50.corr()[metric]\n",
    "units=['bool', 'frac',\n",
    "       'frac', 'frac',\n",
    "       'bool', 'frac', 'deg',\n",
    "       'days', 'frac',\n",
    "       '','',\n",
    "       'int']\n",
    "\n",
    "# make into df\n",
    "filter_results = pd.DataFrame(collections.OrderedDict(\n",
    "    mean=mean,\n",
    "    std=std,\n",
    "    corr=corr))\n",
    "filter_results['units']=units\n",
    "filter_results=filter_results.round(3)\n",
    "\n",
    "# print\n",
    "print('top {mean_of:} scores, \\n\\n- by metric=\"{metric:}\"\\n- for key=\"{key:}\"\\n- number of trials={n:}\\n'.format(\n",
    "    mean_of=mean_of,\n",
    "    key=save_path, \n",
    "    metric=metric,\n",
    "    n=len(trials.trials)\n",
    "))\n",
    "import tabulate\n",
    "markdown=tabulate.tabulate(filter_results,tablefmt='pipe',headers=filter_results.columns)\n",
    "print(markdown)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T04:30:23.911594Z",
     "start_time": "2017-03-26T12:30:23.698324+08:00"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fee8fd0aef0>"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAx8AAAGJCAYAAADrBI7SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fee90e53518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.result_metrics_matthews_corrcoef.hist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- don't normalise\n",
    "- do balance\n",
    "- shift off 0.005 ish\n",
    "- cloud cover of less than 0.7 (no effect)\n",
    "- rotation eange of ~20 deg\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Best filters (for f1 values) for dataset:\n",
    "- l7_joined\n",
    "    - mean of best 50 f1 values\n",
    "    \n",
    "            misc_vals_balanced_classes                                     0.980000\n",
    "            misc_vals_channel_shift_range                                  0.083127\n",
    "            misc_vals_height_shift_range                                   0.164816\n",
    "            misc_vals_max_cloud_cover                                      0.109005\n",
    "            misc_vals_normalized                                           0.000000\n",
    "            misc_vals_rescale                                              0.329541\n",
    "            misc_vals_rotation_range                                       9.903196\n",
    "            misc_vals_timespan_before                                  97939.315502 (1.13 days)\n",
    "            misc_vals_width_shift_range                                    0.270787\n",
    "            result_metrics_report_support_avg / total                    414.480000\n",
    "            result_metrics_matthews_corrcoef                               0.023856\n",
    "            result_metrics_report_f1-score_avg / total                     0.418200\n",
    "            result_metrics_report_f1-score_leak                            0.678600\n",
    "        \n",
    "    - Correlations (how important each is)\n",
    "    \n",
    "            misc_vals_balanced_classes                      0.623902\n",
    "            misc_vals_channel_shift_range                  -0.052222\n",
    "            misc_vals_max_cloud_cover                      -0.021244\n",
    "            misc_vals_normalized                           -0.196311\n",
    "            misc_vals_rotation_range                       -0.113902\n",
    "            misc_vals_timespan_before                       0.084534\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:16:53.490256Z",
     "start_time": "2017-03-26T13:16:53.485991+08:00"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "filters = ['misc_vals_','result_metrics_','result_dummy_metrics_report_dummy_f1-score_leak','result_run_time']\n",
    "cols = [col for col in df.columns if any([f in col for f in filters])]\n",
    "df1 = df[cols]\n",
    "# df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:16:27.420079Z",
     "start_time": "2017-03-26T13:16:27.416945+08:00"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 300,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-26T05:16:55.421554Z",
     "start_time": "2017-03-26T13:16:54.945059+08:00"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fee9006d860>"
      ]
     },
     "execution_count": 300,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA68AAAJqCAYAAADEynZaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAABB5FK8AAAAAgMj7f9k2jaXN745VAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fee90cfd780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# heat map of correlations\n",
    "import seaborn as sns\n",
    "corr = df1.corr()\n",
    "sns.heatmap(corr, \n",
    "            xticklabels=corr.columns.values,\n",
    "            yticklabels=corr.columns.values)\n",
    "# plt.title('')\n",
    "# plt.savefig('/tmp/cryptocorr.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# rotation range etc\n",
    "\n",
    "**Open the plot below in a new window**\n",
    "\n",
    "We see rotation range above 25 tends to lose predictive power with 20 being ideal\n",
    "\n",
    "Cloud cover restrictions seem to do it below 0.6... but I scraped that data to be below 0.3?\n",
    "\n",
    "Channel shifts>0.01 seems to lose predictive power"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T13:01:29.446749Z",
     "start_time": "2017-03-13T21:01:29.441523+08:00"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['misc_vals_balanced_classes', 'misc_vals_channel_shift_range',\n",
       "       'misc_vals_max_cloud_cover', 'misc_vals_normalized',\n",
       "       'misc_vals_rotation_range', 'misc_vals_timespan_before',\n",
       "       'misc_vals_filter_center_cloudy',\n",
       "       'result_dummy_metrics_matthews_corrcoef_dummy', 'result_loss',\n",
       "       'result_metrics_matthews_corrcoef',\n",
       "       'result_metrics_report_f1-score_leak',\n",
       "       'result_metrics_report_precision_leak',\n",
       "       'result_metrics_report_precision_no leak',\n",
       "       'result_metrics_report_recall_avg / total',\n",
       "       'result_metrics_report_recall_leak',\n",
       "       'result_metrics_report_recall_no leak',\n",
       "       'result_metrics_report_support_avg / total', 'result_run_time'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T13:15:41.051905Z",
     "start_time": "2017-03-13T21:15:31.640899+08:00"
    }
   },
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "max must be larger than min in range parameter.",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-37-2278525d9e86>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     13\u001b[0m \u001b[0mdf5\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'precision_leak'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'result_metrics_report_precision_leak'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     14\u001b[0m \u001b[0mdf5\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'misc_vals_timespan_before'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmisc_vals_timespan_before\u001b[0m\u001b[0;34m/\u001b[0m\u001b[0mseconds_in_a_day\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 15\u001b[0;31m \u001b[0msns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpairplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf5\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0msize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m6\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/home/isisilon/.virtualenvs/py3syspck/lib/python3.4/site-packages/seaborn/linearmodels.py\u001b[0m in \u001b[0;36mpairplot\u001b[0;34m(data, hue, hue_order, palette, vars, x_vars, y_vars, kind, diag_kind, markers, size, aspect, dropna, plot_kws, diag_kws, grid_kws)\u001b[0m\n\u001b[1;32m   1607\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mgrid\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msquare_grid\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1608\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mdiag_kind\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"hist\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1609\u001b[0;31m             \u001b[0mgrid\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap_diag\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhist\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mdiag_kws\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1610\u001b[0m         \u001b[0;32melif\u001b[0m \u001b[0mdiag_kind\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"kde\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1611\u001b[0m             \u001b[0mdiag_kws\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"legend\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/isisilon/.virtualenvs/py3syspck/lib/python3.4/site-packages/seaborn/axisgrid.py\u001b[0m in \u001b[0;36mmap_diag\u001b[0;34m(self, func, **kwargs)\u001b[0m\n\u001b[1;32m   1346\u001b[0m                 \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1347\u001b[0m                     func(vals, color=self.palette, histtype=\"barstacked\",\n\u001b[0;32m-> 1348\u001b[0;31m                          **kwargs)\n\u001b[0m\u001b[1;32m   1349\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1350\u001b[0m                 \u001b[0;32mfor\u001b[0m \u001b[0mk\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabel_k\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhue_names\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/matplotlib/pyplot.py\u001b[0m in \u001b[0;36mhist\u001b[0;34m(x, bins, range, normed, weights, cumulative, bottom, histtype, align, orientation, rwidth, log, color, label, stacked, hold, data, **kwargs)\u001b[0m\n\u001b[1;32m   2956\u001b[0m                       \u001b[0mhisttype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mhisttype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0malign\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0malign\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0morientation\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0morientation\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2957\u001b[0m                       \u001b[0mrwidth\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mrwidth\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlog\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlog\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolor\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcolor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlabel\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2958\u001b[0;31m                       stacked=stacked, data=data, **kwargs)\n\u001b[0m\u001b[1;32m   2959\u001b[0m     \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2960\u001b[0m         \u001b[0max\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhold\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mwashold\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/matplotlib/__init__.py\u001b[0m in \u001b[0;36minner\u001b[0;34m(ax, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1810\u001b[0m                     warnings.warn(msg % (label_namer, func.__name__),\n\u001b[1;32m   1811\u001b[0m                                   RuntimeWarning, stacklevel=2)\n\u001b[0;32m-> 1812\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0max\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1813\u001b[0m         \u001b[0mpre_doc\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0minner\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__doc__\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1814\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mpre_doc\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.4/dist-packages/matplotlib/axes/_axes.py\u001b[0m in \u001b[0;36mhist\u001b[0;34m(self, x, bins, range, normed, weights, cumulative, bottom, histtype, align, orientation, rwidth, log, color, label, stacked, **kwargs)\u001b[0m\n\u001b[1;32m   6008\u001b[0m             \u001b[0;31m# this will automatically overwrite bins,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   6009\u001b[0m             \u001b[0;31m# so that each histogram uses the same bins\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 6010\u001b[0;31m             \u001b[0mm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbins\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhistogram\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbins\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweights\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mw\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mhist_kwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   6011\u001b[0m             \u001b[0mm\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m)\u001b[0m  \u001b[0;31m# causes problems later if it's an int\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   6012\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mmlast\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/home/isisilon/.virtualenvs/py3syspck/lib/python3.4/site-packages/numpy/lib/function_base.py\u001b[0m in \u001b[0;36mhistogram\u001b[0;34m(a, bins, range, normed, weights, density)\u001b[0m\n\u001b[1;32m    664\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mmn\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0mmx\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    665\u001b[0m         raise ValueError(\n\u001b[0;32m--> 666\u001b[0;31m             'max must be larger than min in range parameter.')\n\u001b[0m\u001b[1;32m    667\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mall\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0misfinite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mmn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmx\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    668\u001b[0m         raise ValueError(\n",
      "\u001b[0;31mValueError\u001b[0m: max must be larger than min in range parameter."
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f542852dd30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAADyQAAA8kCAYAAACEsBy3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f54362a0ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# this show the relationship between predictive power and days before\n",
    "# so to get a good correlation we want \n",
    "sns.set()\n",
    "plt.figure(figsize=(15,15))\n",
    "df5=df[[\n",
    "    'result_metrics_matthews_corrcoef',\n",
    "    'misc_vals_channel_shift_range',\n",
    "       'misc_vals_max_cloud_cover',        'misc_vals_rotation_range'\n",
    "]].copy()\n",
    "df5['n']=df['result_metrics_report_support_avg / total']\n",
    "df5['f-score']=df1['result_metrics_report_f1-score_leak']\n",
    "df5['precision*recall_leak']=df['result_metrics_report_recall_leak']*df['result_metrics_report_precision_leak']\n",
    "df5['precision_leak']=df['result_metrics_report_precision_leak']\n",
    "df5['misc_vals_timespan_before']=df.misc_vals_timespan_before/seconds_in_a_day\n",
    "sns.pairplot(df5,size=6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-11T02:40:51.258612Z",
     "start_time": "2017-03-11T10:40:51.255108+08:00"
    }
   },
   "source": [
    "# How many days before?\n",
    "\n",
    "This show how many days before we can go before we lose predictive power (either good or bad)\n",
    "\n",
    "If you look at the top right graph, we lose predictive power at greater than 5 days. Before that its\n",
    "exponential increase but I also get a drop of in data. \n",
    "\n",
    "So: 2 days looks like the sweet spot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-11T02:39:06.095251Z",
     "start_time": "2017-03-11T10:39:06.091162+08:00"
    }
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T13:01:38.829306Z",
     "start_time": "2017-03-13T13:01:30.517Z"
    }
   },
   "outputs": [],
   "source": [
    "# this show the relationship between predictive power and days before\n",
    "# so to get a good correlation we want \n",
    "sns.set()\n",
    "plt.figure(figsize=(15,15))\n",
    "df5=df[[\n",
    "    'result_metrics_matthews_corrcoef'\n",
    "]].copy()\n",
    "df5['n']=df['result_metrics_report_support_avg / total']\n",
    "df5['misc_vals_timespan_before']=df.misc_vals_timespan_before/seconds_in_a_day\n",
    "sns.pairplot(df5,size=6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "start_time": "2017-03-12T05:47:14.443Z"
    },
    "code_folding": [],
    "collapsed": true
   },
   "source": [
    "# Hyperopt for randomforest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-12T06:07:40.971308Z",
     "start_time": "2017-03-12T14:07:40.941865+08:00"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T13:01:52.178324Z",
     "start_time": "2017-03-13T21:01:52.057741+08:00"
    }
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "This experiment will look at the best random forest params\n",
    "\"\"\"\n",
    "\n",
    "from hyperopt import fmin, tpe, hp, STATUS_OK, Trials\n",
    "from sklearn.metrics import roc_auc_score\n",
    "import sys\n",
    "from sklearn.dummy import DummyClassifier\n",
    "import sklearn.ensemble\n",
    "import time\n",
    "seconds_in_a_day = 60*60*24\n",
    "\n",
    "trials2 = Trials(exp_key='data_filters_v3_rf_ATX')\n",
    "# trials = MongoTrials('mongo://localhost:27017/hyperopt_leaks2/jobs', exp_key='data_filters_v2')\n",
    "\n",
    "def hp_int(label,min,max):\n",
    "    return hp.choice(label, np.arange(min, max, dtype=int)) \n",
    "\n",
    "\n",
    "space = {\n",
    "\n",
    "\n",
    "\n",
    "            # rf\n",
    "            'n_estimators': hp_int('n_estimators',2,300),\n",
    "            'criterion': hp.choice('criterion',['gini','entropy']),\n",
    "            'max_features': hp_int('max_features',2,300),\n",
    "            'max_depth': hp_int('max_depth',2,300),\n",
    "            'min_samples_split': hp_int('min_samples_split',1,10),\n",
    "            'min_samples_leaf': hp_int('min_samples_leaf',1,10),\n",
    "            'bootstrap':hp.choice('bootstrap',[False,True]),\n",
    "    \n",
    "    \n",
    "            'min_weight_fraction_leaf': 0, # hp.float('min_weight_fraction_leaf',2,10),\n",
    "            'max_leaf_nodes': None, # hp.int('max_leaf_nodes',2,300),\n",
    "            'min_impurity_split': 1e-7, #hp.float('min_impurity_split',1e-12,1e-3),\n",
    "            \n",
    "            'oob_score': False, # hp.choice('oob_score',[False,True]),\n",
    "            'n_jobs':4,\n",
    "            'random_state':0,\n",
    "            'verbose':0,\n",
    "            'warm_start':False,\n",
    "            'class_weight':None,\n",
    "        \n",
    "    \n",
    "            # data filter\n",
    "            'max_cloud_cover': 0.3,\n",
    "            'timespan_before': seconds_in_a_day*3,\n",
    "            'balanced_classes':True,\n",
    "            'normalized': False,\n",
    "            'filter_center_cloudy': True,\n",
    "    \n",
    "            # std is 5% so max should prob be ~5%\n",
    "            'channel_shift_range': 0.07,\n",
    "            'rotation_range': 25,\n",
    "    \n",
    "            'max_depth': 50,\n",
    "            'batch_size' : 2000*10,\n",
    "            'nb_epochs' :  1,\n",
    "        }\n",
    "\n",
    "\n",
    "def f_rf(params):\n",
    "    t0 = time.time()\n",
    "\n",
    "    # Get data\n",
    "    X_train, y_train, metadata_train, X_val, y_val, metadata_val, X_test, y_test, metadata_test = filter_split_data(\n",
    "        X_raw,\n",
    "        y_raw,\n",
    "        metadatas,\n",
    "        max_cloud_cover=params['max_cloud_cover'], \n",
    "        timespan_before=params['timespan_before'],\n",
    "        test_fraction=test_fraction, \n",
    "        random_seed=random_seed,\n",
    "        balanced_classes=params['balanced_classes'],\n",
    "        normalized=params['normalized'],\n",
    "        filter_center_cloudy=params['filter_center_cloudy']\n",
    "    )   \n",
    "\n",
    "    datagen = ImageDataGenerator(\n",
    "    #     featurewise_center=True,\n",
    "    #     featurewise_std_normalization=True,\n",
    "        channel_shift_range=params['channel_shift_range'],\n",
    "        rotation_range=params['rotation_range'],\n",
    "    #     width_shift_range=0.05,\n",
    "    #     height_shift_range=0.05,\n",
    "        horizontal_flip=True,\n",
    "        vertical_flip=True,\n",
    "       dim_ordering='th',\n",
    "    #     rescale=0.02,\n",
    "    #     zoom_range=0.05,\n",
    "    #     shear_range=0.05,\n",
    "    )\n",
    "\n",
    "    # datagen.fit(X_train)\n",
    "    \n",
    "    gen = datagen.flow(X_train, y_train, batch_size=params['batch_size']*params['nb_epochs'])\n",
    "    X_trainb, y_trainb = next(gen)\n",
    "    X_train2b = X_trainb.reshape((-1,14*25*25))\n",
    "    X_val2 = X_val.reshape((-1,14*25*25))\n",
    "    X_test2 = X_test.reshape((-1,14*25*25))\n",
    "    \n",
    "    t1=time.time()-t0\n",
    "\n",
    "    # Get model    \n",
    "    model = sklearn.ensemble.RandomForestClassifier(\n",
    "        n_estimators=params['n_estimators'],\n",
    "        criterion=params['criterion'],\n",
    "        max_features=params['max_features'],\n",
    "        max_depth=params['max_depth'],\n",
    "        min_samples_split=params['min_samples_split'], \n",
    "        min_samples_leaf=params['min_samples_leaf'],\n",
    "        max_leaf_nodes=params['max_leaf_nodes'],\n",
    "        min_impurity_split=params['min_impurity_split'],\n",
    "        bootstrap=params['bootstrap'],\n",
    "        oob_score=params['oob_score'],\n",
    "        n_jobs=params['n_jobs'],\n",
    "#         random_state=params['random_state'],\n",
    "        verbose=params['verbose'],\n",
    "        warm_start=params['warm_start'],\n",
    "        class_weight=params['class_weight'],\n",
    "        \n",
    "    )\n",
    "    model.fit(X_train2b, y_trainb) \n",
    "    y_pred = model.predict(X_val2)\n",
    "    score = model.score(X_val2, y_val)\n",
    "\n",
    "    # score on X_val\n",
    "    report = parse_classification_report(sklearn.metrics.classification_report(y_val>thresh,y_pred>thresh,target_names=target_names))\n",
    "    matthews_corrcoef = sklearn.metrics.matthews_corrcoef(y_val>thresh, y_pred>thresh)\n",
    "    loss=-matthews_corrcoef\n",
    "    \n",
    "#     sys.stdout.flush()\n",
    "    print(t1,'loss',loss)\n",
    "    return dict(loss=loss,\n",
    "                status=STATUS_OK,\n",
    "                run_time=t1,\n",
    "                metrics=dict(\n",
    "                    report=report.to_dict(),\n",
    "                    matthews_corrcoef=matthews_corrcoef,\n",
    "                ),             \n",
    "                attachments=dict(                \n",
    "#                     model=model.__dict__\n",
    "                )\n",
    "               )\n",
    "            \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T17:45:50.034094Z",
     "start_time": "2017-03-13T21:21:34.385522+08:00"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.272477149963379 loss -0.0831334234745\n",
      "2.0736782550811768 loss -0.0924210974528\n",
      "2.413914918899536 loss -0.078942762829\n",
      "2.1177151203155518 loss -0.152185012789\n",
      "2.129666805267334 loss -0.0765051303491\n",
      "2.2079720497131348 loss -0.0576978278033\n",
      "2.241548776626587 loss -0.0908495972473\n",
      "2.179779052734375 loss -0.103385151515\n",
      "2.162714958190918 loss -0.11771247078\n",
      "2.3401927947998047 loss -0.0884890730573\n",
      "2.8878297805786133 loss -0.0783214847039\n",
      "2.1903300285339355 loss -0.0522391331706\n",
      "2.349170446395874 loss -0.0539767866521\n",
      "2.5587830543518066 loss -0.106040281861\n",
      "2.3124563694000244 loss -0.0747560013207\n",
      "2.6566598415374756 loss -0.092875417357\n",
      "2.196125030517578 loss -0.0610194766288\n",
      "2.644954204559326 loss -0.13901848785\n",
      "2.333570718765259 loss -0.0455991092667\n",
      "2.249497652053833 loss -0.0759148602085\n",
      "2.262049436569214 loss -0.065980782972\n",
      "2.004255533218384 loss -0.10676754388\n",
      "2.469154119491577 loss -0.00437870597846\n",
      "2.393832206726074 loss -0.081317742885\n",
      "2.2206265926361084 loss -0.0785766360178\n",
      "2.216156244277954 loss -0.0505477749992\n",
      "2.2465553283691406 loss -0.113740925451\n",
      "2.3417046070098877 loss -0.0940001068469\n",
      "2.612652540206909 loss -0.117114046469\n",
      "2.2212135791778564 loss -0.0457760601246\n",
      "2.1960608959198 loss -0.0649733170326\n",
      "2.1780216693878174 loss -0.0979144673189\n",
      "2.5605523586273193 loss -0.0686757800803\n",
      "2.5702219009399414 loss -0.105484461654\n",
      "2.2017927169799805 loss -0.101090948864\n",
      "2.198988437652588 loss -0.0634117025284\n",
      "2.497110366821289 loss -0.131897710042\n",
      "2.632697105407715 loss -0.12090793662\n",
      "2.81476092338562 loss -0.0516948720741\n",
      "2.164499521255493 loss -0.037378704278\n",
      "2.207096576690674 loss -0.142844902562\n",
      "2.301442861557007 loss -0.0187473970968\n",
      "2.168757915496826 loss -0.050767683583\n",
      "2.581979274749756 loss -0.0802991350031\n",
      "2.1655192375183105 loss -0.0434177252632\n",
      "2.264699935913086 loss -0.0919981829868\n",
      "2.5627617835998535 loss -0.0858664255344\n",
      "2.2008018493652344 loss -0.0765051303491\n",
      "2.14874529838562 loss -0.107180265742\n",
      "2.184567928314209 loss -0.0535391503585\n",
      "2.2430219650268555 loss -0.0920819515108\n",
      "2.212372303009033 loss -0.0401036900654\n",
      "2.2695999145507812 loss -0.0583675898273\n",
      "2.3197624683380127 loss -0.138453013963\n",
      "2.4439830780029297 loss -0.053038145883\n",
      "2.0221474170684814 loss -0.101650736417\n",
      "1.954366683959961 loss -0.0763788603524\n",
      "1.9254844188690186 loss -0.090657853112\n",
      "1.9014248847961426 loss -0.138804658913\n",
      "1.9059479236602783 loss -0.103385151515\n",
      "1.8416681289672852 loss -0.0678477941487\n",
      "1.8779652118682861 loss -0.120082068543\n",
      "1.9419021606445312 loss -0.0343338366094\n",
      "1.9368500709533691 loss -0.0607694779249\n",
      "1.923565149307251 loss -0.0945299586087\n",
      "1.90217924118042 loss -0.133308692815\n",
      "2.251796245574951 loss -0.101844986261\n",
      "1.9694671630859375 loss -0.034663421535\n",
      "1.879384994506836 loss -0.0945299586087\n",
      "1.9158179759979248 loss -0.0330036793899\n",
      "1.997110366821289 loss -0.0908495972473\n",
      "2.5520095825195312 loss -0.145150835495\n",
      "1.9399936199188232 loss -0.0896524689473\n",
      "2.1461238861083984 loss -0.0577825399784\n",
      "1.891322135925293 loss -0.0576978278033\n",
      "1.9212026596069336 loss -0.0589586235376\n",
      "2.071608543395996 loss -0.0946264229947\n",
      "1.9063045978546143 loss -0.0637288837661\n",
      "2.0294086933135986 loss -0.090657853112\n",
      "1.9250898361206055 loss -0.042349275732\n",
      "1.9384419918060303 loss -0.0615676690749\n",
      "1.9407880306243896 loss -0.0601662612155\n",
      "1.9225695133209229 loss -0.038491734636\n",
      "1.8768441677093506 loss -0.0621714210991\n",
      "2.135751247406006 loss -0.101090948864\n",
      "1.9698553085327148 loss -0.00823135783324\n",
      "1.9567601680755615 loss -0.107776520892\n",
      "2.147087335586548 loss -0.0990058028914\n",
      "2.142184019088745 loss -0.131438427922\n",
      "2.1391549110412598 loss -0.0675782564573\n",
      "2.105844497680664 loss -0.0984156987341\n",
      "1.919830322265625 loss -0.0677352277008\n",
      "2.1734542846679688 loss -0.0642829849834\n",
      "1.9006235599517822 loss -0.0390167952541\n",
      "1.9997014999389648 loss -0.0697978803731\n",
      "1.9252631664276123 loss -0.0808882910299\n",
      "1.8836431503295898 loss -0.122693767174\n",
      "1.9430854320526123 loss -0.0307635798384\n",
      "1.8931622505187988 loss -0.0765051303491\n",
      "1.917811632156372 loss -0.12499727821\n",
      "1.8973944187164307 loss -0.0302186275653\n",
      "1.8989388942718506 loss -0.059334654635\n",
      "1.8455822467803955 loss -0.0550660792952\n",
      "1.9154362678527832 loss -0.0725203518877\n",
      "1.890376329421997 loss -0.160756787017\n",
      "1.8901684284210205 loss -0.108963595926\n",
      "1.9257233142852783 loss -0.0914612730523\n",
      "1.9181849956512451 loss -0.0884890730573\n",
      "2.136522054672241 loss -0.0884890730573\n",
      "1.9116249084472656 loss -0.117113809787\n",
      "2.154127597808838 loss -0.108891098811\n",
      "1.9067628383636475 loss -0.0879195896906\n",
      "2.150751829147339 loss -0.0736249228418\n",
      "2.059560775756836 loss -0.0934504487509\n",
      "1.9569735527038574 loss -0.0686757800803\n",
      "1.9151723384857178 loss -0.130260461722\n",
      "1.9398069381713867 loss -0.0923087985585\n",
      "1.9037628173828125 loss -0.104400409176\n",
      "1.8856403827667236 loss -0.0467579069759\n",
      "1.9639427661895752 loss -0.108282437439\n",
      "2.0048279762268066 loss -0.0703685321551\n",
      "2.2502973079681396 loss -0.115207510486\n",
      "1.8935842514038086 loss -0.137358404247\n",
      "1.8894720077514648 loss -0.0549410471251\n",
      "1.9946928024291992 loss -0.128307047722\n",
      "1.8838300704956055 loss -0.0379430058714\n",
      "1.9047865867614746 loss -0.0827032342273\n",
      "1.867363691329956 loss -0.065980782972\n",
      "2.908477544784546 loss -0.020335762705\n",
      "1.9859187602996826 loss 0.0247354743726\n",
      "2.0023863315582275 loss -0.060159376665\n",
      "1.9454665184020996 loss -0.0615924671148\n",
      "2.0272600650787354 loss -0.0846829672436\n",
      "2.1472408771514893 loss -0.0522391331706\n",
      "2.109426975250244 loss -0.099464477935\n",
      "2.076936721801758 loss -0.0291363936002\n",
      "1.8960051536560059 loss -0.0753318751078\n",
      "1.9494848251342773 loss -0.133895105602\n",
      "1.926264762878418 loss -0.0797175876814\n",
      "1.901444673538208 loss -0.081317742885\n",
      "1.9200818538665771 loss 0.00666507790177\n",
      "1.918950080871582 loss -0.129676549992\n",
      "2.2538864612579346 loss -0.0159346962976\n",
      "1.9304168224334717 loss -0.045710297902\n",
      "2.1862120628356934 loss -0.119771981413\n",
      "1.9967834949493408 loss -0.0500456560705\n",
      "1.964953899383545 loss -0.116509535157\n",
      "2.1903748512268066 loss -0.0543883674915\n",
      "2.0945780277252197 loss -0.0533921467724\n",
      "1.9008405208587646 loss -0.0769181636803\n",
      "1.9007441997528076 loss -0.0984228537846\n",
      "2.03547739982605 loss -0.129443363728\n",
      "1.8992705345153809 loss -0.129676549992\n",
      "1.9064445495605469 loss -0.0527588507085\n",
      "1.9087543487548828 loss -0.0557653745944\n",
      "2.2297580242156982 loss -0.134827545713\n",
      "2.021090269088745 loss -0.0624209766392\n",
      "1.9064786434173584 loss -0.0858664255344\n",
      "1.9093778133392334 loss -0.0132009112551\n",
      "1.9023089408874512 loss -0.0457760601246\n",
      "1.906846523284912 loss -0.111240883313\n",
      "1.9312796592712402 loss -0.0714411015991\n",
      "1.9411375522613525 loss -0.0577825399784\n",
      "1.9291369915008545 loss -0.113337985781\n",
      "1.9406585693359375 loss -0.0319408723268\n",
      "1.911888599395752 loss -0.0528129861336\n",
      "2.207719087600708 loss -0.111563763532\n",
      "1.9166085720062256 loss -0.0566296908362\n",
      "1.975830316543579 loss -0.0950598929015\n",
      "1.9226136207580566 loss -0.0494438002923\n",
      "1.9955275058746338 loss -0.136311298458\n",
      "2.0204832553863525 loss -0.0654523544061\n",
      "1.9152593612670898 loss -0.02691523355\n",
      "2.2059710025787354 loss -0.0759148602085\n",
      "1.972764253616333 loss -0.045047851217\n",
      "1.959801435470581 loss -0.104400409176\n",
      "1.9620394706726074 loss -0.104937885349\n",
      "1.9734573364257812 loss -0.0852706416016\n",
      "1.9744393825531006 loss 0.0220544935899\n",
      "2.031165361404419 loss -0.0927118387897\n",
      "1.9312865734100342 loss -0.117858238196\n",
      "1.929194450378418 loss -0.134827545713\n",
      "1.9032747745513916 loss -0.0329723100498\n",
      "1.938305139541626 loss -0.100540073557\n",
      "1.9479811191558838 loss -0.0318618676535\n",
      "2.008589506149292 loss -0.136828979716\n",
      "1.9159047603607178 loss -0.0582383492343\n",
      "1.8974754810333252 loss -0.0390167952541\n",
      "1.8935036659240723 loss -0.182557198471\n",
      "2.2246391773223877 loss -0.0797316611004\n",
      "2.19687557220459 loss -0.0670383440778\n",
      "1.919950246810913 loss -0.034663421535\n",
      "2.0170676708221436 loss -0.115638843244\n",
      "1.89961576461792 loss -0.0792163562663\n",
      "2.0726208686828613 loss -0.0785766360178\n",
      "1.919705867767334 loss -0.0488984420113\n",
      "1.922788381576538 loss -0.0407901616791\n",
      "1.958472490310669 loss -0.0753318751078\n",
      "1.9718108177185059 loss -0.038491734636\n",
      "1.9257545471191406 loss -0.096457881503\n",
      "1.8958408832550049 loss -0.128128320067\n",
      "1.9814541339874268 loss -0.0681240207462\n",
      "1.9440035820007324 loss -0.112144896181\n",
      "1.874507188796997 loss -0.0689386539477\n",
      "1.9221022129058838 loss -0.0455991092667\n",
      "1.9322073459625244 loss -0.060452284031\n",
      "1.9342303276062012 loss -0.112188472982\n",
      "1.9454269409179688 loss -0.0634117025284\n",
      "1.9201056957244873 loss -0.0329723100498\n",
      "1.9690392017364502 loss -0.0966991564162\n",
      "1.9372045993804932 loss -0.0875635383479\n",
      "2.2621545791625977 loss -0.103385151515\n",
      "2.2119712829589844 loss -0.145160976366\n",
      "1.9101645946502686 loss -0.050767683583\n",
      "1.8958659172058105 loss -0.0252770815233\n",
      "1.8857791423797607 loss -0.0620999761538\n",
      "1.9361152648925781 loss 0.0115294004597\n",
      "1.9150688648223877 loss -0.0467579069759\n",
      "1.9113922119140625 loss -0.113337985781\n",
      "1.9403338432312012 loss -0.103871896467\n",
      "1.98020339012146 loss -0.108963595926\n",
      "1.8617527484893799 loss -0.149094336151\n",
      "2.2284095287323 loss -0.0197801718498\n",
      "2.1718361377716064 loss -0.0291359532429\n",
      "2.258470058441162 loss -0.119574637852\n",
      "2.1982245445251465 loss -0.108963595926\n",
      "2.1535863876342773 loss -0.0841468018363\n",
      "1.9435997009277344 loss -0.0958381413214\n",
      "1.907564640045166 loss -0.0633479082521\n",
      "1.9140510559082031 loss -0.0551632158577\n",
      "1.9043726921081543 loss -0.0499935291897\n",
      "2.254030466079712 loss -0.0406524037444\n",
      "2.008908271789551 loss -0.0719776326581\n",
      "1.95255446434021 loss -0.0979144673189\n",
      "1.9246985912322998 loss -0.12090793662\n",
      "1.9108197689056396 loss -0.131384736749\n",
      "2.0288233757019043 loss -0.103385151515\n",
      "1.9703264236450195 loss -0.035234543452\n",
      "1.9840888977050781 loss -0.0197944557418\n",
      "2.322270631790161 loss -0.0296762789796\n",
      "1.9008126258850098 loss -0.0970871462163\n",
      "1.912318468093872 loss -0.110431465139\n",
      "2.160705804824829 loss -0.0747560013207\n",
      "1.916093349456787 loss -0.0302186275653\n",
      "1.897564172744751 loss -0.0990058028914\n",
      "1.87799072265625 loss -0.107180265742\n",
      "2.053199291229248 loss -0.106040281861\n",
      "1.9825963973999023 loss -0.119574637852\n",
      "1.8810725212097168 loss -0.105802017746\n",
      "1.9084827899932861 loss -0.0924892048676\n",
      "1.9849252700805664 loss -0.0368103788305\n",
      "1.9359517097473145 loss -0.0753318751078\n",
      "1.9162437915802002 loss -0.101650736417\n",
      "2.2428641319274902 loss -0.0966991564162\n",
      "2.1030454635620117 loss -0.115638843244\n",
      "1.9424123764038086 loss -0.0845935181281\n",
      "2.0821516513824463 loss -0.129443363728\n",
      "2.205582618713379 loss -0.0176145929256\n",
      "1.9402730464935303 loss -0.0703685321551\n",
      "2.0109660625457764 loss -0.140186567326\n",
      "1.936178207397461 loss -0.0715298560645\n",
      "2.03877329826355 loss -0.107180265742\n",
      "1.911595106124878 loss -0.0966991564162\n",
      "1.8958818912506104 loss -0.0873580309709\n",
      "2.201850414276123 loss -0.0769181636803\n",
      "2.1555745601654053 loss -0.092875417357\n",
      "1.9671976566314697 loss -0.109578322485\n",
      "2.234816074371338 loss -0.0861136768759\n",
      "2.071816921234131 loss -0.0851880207504\n",
      "1.9329428672790527 loss -0.038491734636\n",
      "1.977942943572998 loss -0.0445006302187\n",
      "2.135798931121826 loss -0.0697978803731\n",
      "2.023602247238159 loss -0.102219590445\n",
      "1.9720029830932617 loss -0.101090948864\n",
      "1.8906338214874268 loss -0.0582383492343\n",
      "1.8680779933929443 loss -0.0648428980937\n",
      "2.1064252853393555 loss -0.175259331341\n",
      "1.9557545185089111 loss -0.00823135783324\n",
      "2.213651180267334 loss -0.100540073557\n",
      "1.9559099674224854 loss -0.0747560013207\n",
      "1.944065809249878 loss -0.107776520892\n",
      "1.9384546279907227 loss -0.0357073349424\n",
      "1.9307761192321777 loss -0.00165473015587\n",
      "1.9483838081359863 loss -0.0610194766288\n",
      "1.896169900894165 loss -0.0488984420113\n",
      "2.0633668899536133 loss -0.134634358055\n",
      "2.0191967487335205 loss -0.0919981829868\n",
      "1.9431431293487549 loss -0.110431465139\n",
      "1.926288366317749 loss -0.0896530595712\n",
      "1.984740972518921 loss -0.0197801718498\n",
      "2.0062997341156006 loss -0.0627565093911\n",
      "1.895641565322876 loss -0.0960072321874\n",
      "1.903278112411499 loss -0.0481804840783\n",
      "1.8935964107513428 loss -0.0549410471251\n",
      "1.9077117443084717 loss -0.085719376627\n",
      "1.942826509475708 loss -0.12801829402\n",
      "2.213685989379883 loss -0.0862580709034\n",
      "1.9019503593444824 loss -0.0566296908362\n",
      "1.9988312721252441 loss -0.042349275732\n",
      "1.9184081554412842 loss -0.0225875383826\n",
      "1.9577953815460205 loss -0.0189465278566\n",
      "2.0706441402435303 loss -0.121535696697\n",
      "1.8973779678344727 loss -0.10987998078\n",
      "2.193061590194702 loss -0.0287357506825\n",
      "1.9655392169952393 loss -0.132049548567\n",
      "2.158751964569092 loss -0.144750492657\n",
      "1.9524664878845215 loss -0.129676549992\n",
      "1.9212746620178223 loss -0.102797672466\n",
      "1.9058544635772705 loss -0.0648428980937\n",
      "1.8945338726043701 loss -0.0829668607001\n",
      "1.8670852184295654 loss -0.107263522845\n",
      "1.9509377479553223 loss 0.0416476125745\n",
      "2.1542348861694336 loss -0.0461545519182\n",
      "1.9038922786712646 loss -0.0986127867169\n",
      "2.217942953109741 loss -0.0621714210991\n",
      "1.9098715782165527 loss -0.0725203518877\n",
      "1.9047234058380127 loss -0.114277792572\n",
      "1.9519076347351074 loss -0.0642829849834\n",
      "2.232672691345215 loss -0.0927118387897\n",
      "1.941432237625122 loss -0.0522247154021\n",
      "1.9372804164886475 loss -0.125291404382\n",
      "1.9442286491394043 loss -0.055498725006\n",
      "1.93833327293396 loss -0.0358304118749\n",
      "1.920067548751831 loss -0.0499935291897\n",
      "1.902808427810669 loss -0.0785766360178\n",
      "1.9718055725097656 loss -0.0870830670117\n",
      "1.9944806098937988 loss 0.01867265903\n",
      "2.024838924407959 loss -0.177196240387\n",
      "1.911667823791504 loss -0.128307047722\n",
      "1.9657883644104004 loss -0.117857285352\n",
      "1.9650189876556396 loss -0.0791434767152\n",
      "1.9580447673797607 loss -0.105802017746\n",
      "2.3080053329467773 loss -0.10880560338\n",
      "1.9423129558563232 loss -0.124722532752\n",
      "2.165111541748047 loss -0.0648428980937\n",
      "1.9804704189300537 loss -0.101650736417\n",
      "1.8774073123931885 loss -0.0745582620265\n",
      "2.0549228191375732 loss -0.10933513042\n",
      "1.9657909870147705 loss -0.0769181636803\n",
      "1.9374985694885254 loss -0.0895965300934\n",
      "1.8997125625610352 loss -0.119771981413\n",
      "1.9880359172821045 loss -0.0587839105193\n",
      "1.9250879287719727 loss -0.0428857992265\n",
      "1.9847872257232666 loss 0.0259319797026\n",
      "1.9815216064453125 loss -0.15176469786\n",
      "1.885209321975708 loss -0.0423213655761\n",
      "2.0495846271514893 loss -0.0972625433635\n",
      "1.9281814098358154 loss -0.0610194766288\n",
      "1.976227045059204 loss -0.0457760601246\n",
      "1.9350345134735107 loss -0.060452284031\n",
      "1.9203529357910156 loss -0.0637288837661\n",
      "2.2321674823760986 loss -0.135217054768\n",
      "2.2005929946899414 loss -0.117625391112\n",
      "2.2187044620513916 loss -0.102219590445\n",
      "1.9159905910491943 loss -0.0654523544061\n",
      "2.177461624145508 loss -0.0550660792952\n",
      "1.9101414680480957 loss -0.0252990589163\n",
      "1.9243059158325195 loss -0.0392200904583\n",
      "1.9497718811035156 loss -0.0109760640518\n",
      "1.948272466659546 loss -0.0824099163152\n",
      "1.9187142848968506 loss -0.0807822270444\n",
      "1.8625762462615967 loss -0.059334654635\n",
      "1.9540917873382568 loss -0.104757384634\n",
      "1.8972125053405762 loss -0.0961443514271\n",
      "1.9835329055786133 loss -0.0858664255344\n",
      "2.5884623527526855 loss -0.0759148602085\n",
      "1.9186582565307617 loss -0.0979144673189\n",
      "1.9250435829162598 loss -0.157979674619\n",
      "1.9216904640197754 loss -0.0703685321551\n",
      "1.9327452182769775 loss -0.104418431604\n",
      "1.9403955936431885 loss -0.0838322838844\n",
      "1.9285871982574463 loss -0.0455991092667\n",
      "1.9811525344848633 loss -0.0412048213264\n",
      "2.2429120540618896 loss -0.0610404684162\n",
      "1.951716423034668 loss -0.0996051609608\n",
      "2.0094964504241943 loss -0.0307635798384\n",
      "2.224017858505249 loss -0.0285764144289\n",
      "2.1986002922058105 loss -0.0362571714083\n",
      "2.25356388092041 loss -0.153320265401\n",
      "1.9188570976257324 loss -0.042349275732\n",
      "1.8817839622497559 loss -0.119219888003\n",
      "1.933318853378296 loss -0.0774641309732\n",
      "1.912841558456421 loss -0.0862580709034\n",
      "1.954021692276001 loss -0.0576978278033\n",
      "2.015620708465576 loss -0.101460668794\n",
      "1.9316892623901367 loss -0.0890666432078\n",
      "1.9657917022705078 loss -0.125871458426\n",
      "1.908172845840454 loss -0.0148529007107\n",
      "2.164991855621338 loss -0.103385151515\n",
      "1.979856252670288 loss -0.0126416694105\n",
      "2.1778740882873535 loss -0.0824099163152\n",
      "1.9427530765533447 loss -0.143404783643\n",
      "2.5551035404205322 loss -0.0818602591107\n",
      "2.0171585083007812 loss -0.111563763532\n",
      "2.02069091796875 loss -0.0771028655743\n",
      "1.9287006855010986 loss -0.092875417357\n",
      "2.00661039352417 loss -0.034663421535\n",
      "1.936842679977417 loss -0.125871458426\n",
      "2.0855844020843506 loss -0.128307047722\n",
      "1.9202523231506348 loss -0.0818602591107\n",
      "1.9583485126495361 loss -0.0516948720741\n",
      "1.9435343742370605 loss -0.0826368878015\n",
      "2.1918535232543945 loss -0.00877785956727\n",
      "1.9735386371612549 loss -0.0693542081034\n",
      "1.9328815937042236 loss -0.108963595926\n",
      "1.9244298934936523 loss -0.0455991092667\n",
      "2.1126062870025635 loss -0.0777082525203\n",
      "2.1559016704559326 loss -0.0258208157727\n",
      "2.116103410720825 loss -0.0858664255344\n",
      "2.204299211502075 loss -0.0528129861336\n",
      "1.9863295555114746 loss -0.0990058028914\n",
      "1.9361588954925537 loss -0.0714411015991\n",
      "1.8644506931304932 loss -0.0274662057428\n",
      "1.967883586883545 loss -0.035234543452\n",
      "2.1228113174438477 loss -0.0428857992265\n",
      "1.8586852550506592 loss -0.0912001663337\n",
      "2.1371073722839355 loss -0.0368103788305\n",
      "1.9206969738006592 loss -0.133504792335\n",
      "1.9185163974761963 loss -0.104589015905\n",
      "1.9055554866790771 loss -0.0733245169438\n",
      "2.1277997493743896 loss -0.100882254023\n",
      "2.2032182216644287 loss -0.0324710786346\n",
      "2.009094715118408 loss -0.0681240207462\n",
      "1.9270055294036865 loss -0.101650736417\n",
      "1.9237165451049805 loss -0.0545670823768\n",
      "1.9313204288482666 loss -0.0489999519538\n",
      "1.8966560363769531 loss -0.0615924671148\n",
      "1.90690016746521 loss -0.113740925451\n",
      "1.9879436492919922 loss -0.0777089298895\n",
      "1.9135658740997314 loss 0.00274324915757\n",
      "1.953009843826294 loss -0.0759148602085\n",
      "1.8932819366455078 loss -0.152185012789\n",
      "1.8465368747711182 loss -0.0950598929015\n",
      "2.141949415206909 loss -0.127487854271\n",
      "2.2162423133850098 loss -0.0703685321551\n",
      "1.923685073852539 loss -0.0120833593845\n",
      "1.8998298645019531 loss -0.130252109708\n",
      "2.039860963821411 loss -0.02691523355\n",
      "1.953913927078247 loss -0.044605665052\n",
      "1.954866647720337 loss -0.00279402893985\n",
      "2.19350266456604 loss -0.184599238564\n",
      "1.8808281421661377 loss -0.0347895213758\n",
      "1.9641449451446533 loss -0.085719376627\n",
      "1.9846813678741455 loss -0.0851880207504\n",
      "1.9189631938934326 loss -0.0739375273237\n",
      "2.1662938594818115 loss -0.0829668607001\n",
      "1.9082891941070557 loss -0.0774641309732\n",
      "2.20949125289917 loss -0.0560615538904\n",
      "1.8774309158325195 loss -0.0582383492343\n",
      "1.991793155670166 loss -0.0670394509301\n",
      "2.1279053688049316 loss -0.118942685103\n",
      "1.8845868110656738 loss -0.0654087776056\n",
      "1.9825890064239502 loss -0.107117842075\n",
      "1.9065778255462646 loss -0.0170522132427\n",
      "1.9010422229766846 loss -0.0677352277008\n",
      "1.9245591163635254 loss -0.132957607018\n",
      "2.220524787902832 loss -0.0511066876889\n",
      "1.8842661380767822 loss 0.00882795411419\n",
      "1.8703839778900146 loss -0.0978346693994\n",
      "1.9612538814544678 loss -0.0340961162028\n",
      "2.179927110671997 loss -0.0654087776056\n",
      "2.109015464782715 loss -0.0721208720098\n",
      "2.0488908290863037 loss -0.0478475160291\n",
      "1.9448528289794922 loss -0.0774641309732\n",
      "1.9026081562042236 loss -0.0852706416016\n",
      "2.0179781913757324 loss -0.0615924671148\n",
      "1.9007294178009033 loss -0.0405751724118\n",
      "1.949885368347168 loss -0.0538405374417\n",
      "1.9186301231384277 loss -0.132049548567\n",
      "1.8672761917114258 loss -0.107768450127\n",
      "1.9501726627349854 loss -0.10776475426\n",
      "1.8864731788635254 loss -0.108455876758\n",
      "1.931056261062622 loss -0.0368103788305\n",
      "1.8590431213378906 loss -0.0115388131193\n",
      "2.0217020511627197 loss -0.0639142974308\n",
      "1.947420358657837 loss -0.138453013963\n",
      "1.9340651035308838 loss -0.0598907286721\n",
      "1.8602228164672852 loss -0.0335324652823\n",
      "1.8460638523101807 loss -0.0841032250358\n",
      "1.9225103855133057 loss -0.135811920839\n",
      "1.8782484531402588 loss -0.0966991564162\n",
      "1.9063780307769775 loss -0.115207510486\n",
      "1.920142412185669 loss -0.0654087776056\n",
      "1.8873217105865479 loss -0.0721208720098\n",
      "1.9113550186157227 loss -0.103385151515\n",
      "2.121018409729004 loss -0.0791434767152\n",
      "1.9271609783172607 loss -0.125871458426\n",
      "2.228130340576172 loss -0.0390167952541\n",
      "1.9099020957946777 loss -0.144659744951\n",
      "1.9287922382354736 loss -0.14140532996\n",
      "1.9719154834747314 loss -0.127065804724\n",
      "1.9241156578063965 loss -0.0440277841537\n",
      "1.9681377410888672 loss -0.055498725006\n",
      "1.8798880577087402 loss 0.00496689036363\n",
      "2.21374249458313 loss -0.0675782564573\n",
      "2.1108546257019043 loss -0.106040281861\n",
      "1.8775737285614014 loss -0.119219888003\n",
      "1.912513256072998 loss -0.149729811755\n",
      "1.9754579067230225 loss -0.0725203518877\n",
      "1.9363734722137451 loss -0.0707983124832\n",
      "1.9536025524139404 loss -0.0384783314586\n",
      "1.9679887294769287 loss -0.0736249228418\n",
      "2.0812721252441406 loss -0.0621714210991\n",
      "1.922593355178833 loss -0.0610404684162\n",
      "2.5439000129699707 loss -0.134063658197\n",
      "2.1816906929016113 loss -0.0631804446647\n",
      "2.215517044067383 loss -0.0629139648498\n",
      "2.2244374752044678 loss -0.0203373592371\n",
      "1.9721968173980713 loss -0.020335762705\n",
      "1.989311695098877 loss -0.0841032250358\n",
      "1.9392085075378418 loss -0.101650736417\n",
      "1.974106788635254 loss -0.103982204388\n",
      "1.9216537475585938 loss -0.12801829402\n",
      "1.973402738571167 loss 0.0116210645483\n",
      "1.9388139247894287 loss -0.0780169055256\n",
      "2.229820966720581 loss -0.132049548567\n",
      "2.009270191192627 loss -0.138127158172\n",
      "1.9998407363891602 loss -0.0774641309732\n",
      "1.9943995475769043 loss -0.0762569644046\n",
      "1.9008476734161377 loss -0.0920819515108\n",
      "2.120903491973877 loss 0.00774652631886\n",
      "1.989898681640625 loss -0.113740925451\n",
      "1.8869378566741943 loss -0.090657853112\n",
      "1.9778459072113037 loss -0.122040563661\n",
      "1.9435186386108398 loss -0.130252109708\n",
      "1.8525416851043701 loss -0.0501729924353\n",
      "1.8730037212371826 loss -0.0732755926162\n",
      "1.9552223682403564 loss -0.0846638713101\n",
      "2.100559711456299 loss -0.125291404382\n",
      "1.9310054779052734 loss -0.0432272198633\n",
      "1.9272971153259277 loss -0.0917504325898\n",
      "2.1740355491638184 loss -0.123311606276\n",
      "1.9930121898651123 loss -0.0337575541548\n",
      "1.8690989017486572 loss -0.0484212447931\n",
      "1.9542129039764404 loss -0.0923087985585\n",
      "1.9032058715820312 loss -0.138804658913\n",
      "2.123246431350708 loss -0.0670383440778\n",
      "1.9052517414093018 loss -0.0171078215042\n",
      "1.9012515544891357 loss -0.101460668794\n",
      "1.9308326244354248 loss -0.0358304118749\n",
      "1.9165618419647217 loss -0.0445006302187\n",
      "2.17138671875 loss -0.0879195896906\n",
      "1.95145583152771 loss -0.0319408723268\n",
      "1.9685394763946533 loss -0.10776475426\n",
      "1.9809017181396484 loss -0.104400409176\n",
      "1.886171579360962 loss -0.134634358055\n",
      "1.8963291645050049 loss -0.0725203518877\n",
      "2.5200464725494385 loss -0.171789453037\n",
      "1.9196479320526123 loss -0.114703402149\n",
      "1.8822154998779297 loss -0.0472785987005\n",
      "1.910759449005127 loss -0.0595558231915\n",
      "1.9118695259094238 loss -0.070945801059\n",
      "1.9161512851715088 loss -0.0923087985585\n",
      "2.1222546100616455 loss -0.154997205803\n",
      "2.282870054244995 loss -0.0160398923931\n",
      "1.9277703762054443 loss -0.0719776326581\n",
      "1.9837000370025635 loss -0.0929877833362\n",
      "1.9141554832458496 loss -0.0829668607001\n",
      "1.9014463424682617 loss -0.155506178767\n",
      "1.917360782623291 loss -0.0401036900654\n",
      "2.1994659900665283 loss -0.0516704211024\n",
      "1.981156349182129 loss -0.0833246830567\n",
      "1.8948731422424316 loss -0.128945759453\n",
      "1.9693033695220947 loss -0.0538405374417\n",
      "2.0849812030792236 loss -0.0615924671148\n",
      "2.2397449016571045 loss -0.0539767866521\n",
      "1.8667850494384766 loss -0.0686757800803\n",
      "2.0220696926116943 loss -0.0763414023954\n",
      "1.9221038818359375 loss -0.0775427471314\n",
      "2.0010218620300293 loss -0.0774641309732\n",
      "1.932893991470337 loss -0.111563763532\n",
      "1.9226999282836914 loss -0.070945801059\n",
      "2.2186150550842285 loss -0.050767683583\n",
      "2.118023157119751 loss -0.0380041139071\n",
      "1.8998048305511475 loss -0.119771981413\n",
      "1.9228711128234863 loss -0.108359136836\n",
      "1.8987383842468262 loss -0.135811920839\n",
      "2.2333552837371826 loss -0.0802991350031\n",
      "1.893449306488037 loss -0.160116892747\n",
      "1.8976693153381348 loss -0.11395031109\n",
      "2.0041027069091797 loss -0.0368812305215\n",
      "1.9702744483947754 loss -0.100540073557\n",
      "1.9883308410644531 loss -0.0769181636803\n",
      "1.914872407913208 loss -0.0274662057428\n",
      "1.8997986316680908 loss -0.0557653745944\n",
      "1.9005825519561768 loss -0.124722532752\n",
      "1.9080641269683838 loss -0.113337985781\n",
      "1.9056317806243896 loss -0.0572032974223\n",
      "1.9151663780212402 loss -0.0472785987005\n",
      "1.9005146026611328 loss -0.0627565093911\n",
      "1.9568562507629395 loss -0.135310689424\n",
      "1.8865666389465332 loss -0.0763788603524\n",
      "1.9712512493133545 loss -0.112736243651\n",
      "1.9189929962158203 loss -0.180301253349\n",
      "2.0096535682678223 loss -0.117113809787\n",
      "1.9266440868377686 loss -0.0455991092667\n",
      "1.9485201835632324 loss -0.0701708865185\n",
      "1.887988805770874 loss -0.175259331341\n",
      "1.9064404964447021 loss -0.0654087776056\n",
      "1.8905978202819824 loss -0.065980782972\n",
      "1.9435796737670898 loss -0.0774641309732\n",
      "1.94435453414917 loss -0.102219590445\n",
      "1.9214532375335693 loss -0.060452284031\n",
      "2.1895952224731445 loss -0.0683445672708\n",
      "2.1737232208251953 loss -0.0879195896906\n",
      "1.9751896858215332 loss -0.16276696151\n",
      "1.9129300117492676 loss -0.0484212447931\n",
      "1.9147801399230957 loss -0.0841468018363\n",
      "1.9469242095947266 loss -0.0627565093911\n",
      "1.9277422428131104 loss -0.0488984420113\n",
      "1.9255149364471436 loss -0.0659755251347\n",
      "2.298863172531128 loss -0.157732972495\n",
      "1.9230260848999023 loss -0.0742855990646\n",
      "1.900740385055542 loss -0.113740925451\n",
      "1.9702301025390625 loss -0.146094193624\n",
      "1.9034309387207031 loss -0.125871458426\n",
      "1.8919222354888916 loss -0.065980782972\n",
      "1.9207649230957031 loss -0.0862580709034\n",
      "2.1885032653808594 loss -0.142297514241\n",
      "1.9089634418487549 loss -0.0807822270444\n",
      "1.8940529823303223 loss -0.0946589434248\n",
      "2.1237030029296875 loss -0.0527588507085\n",
      "1.889650583267212 loss -0.00878315601428\n",
      "1.9303061962127686 loss -0.068333438029\n",
      "1.8998429775238037 loss -0.0274662057428\n",
      "1.90924072265625 loss -0.0499935291897\n",
      "1.936182975769043 loss -0.105205779195\n",
      "2.214877128601074 loss -0.0654523544061\n",
      "2.1542181968688965 loss -0.10987998078\n",
      "2.188228130340576 loss -0.096457881503\n",
      "1.9608469009399414 loss -0.0516704211024\n",
      "2.0128626823425293 loss -0.117082084945\n",
      "1.931058406829834 loss -0.117082084945\n",
      "1.9310872554779053 loss -0.116612578372\n",
      "2.0333054065704346 loss -0.0698673007399\n",
      "1.9133803844451904 loss -0.0197944557418\n",
      "1.9177203178405762 loss -0.0769181636803\n",
      "1.9309663772583008 loss -0.070945801059\n",
      "1.896791934967041 loss -0.0934940255514\n",
      "1.8783197402954102 loss -0.128307047722\n",
      "1.9364402294158936 loss -0.132957607018\n",
      "1.887648582458496 loss -0.114573417068\n",
      "1.9419867992401123 loss -0.0247354743726\n",
      "1.9531035423278809 loss -0.0785766360178\n",
      "1.8920254707336426 loss -0.0934940255514\n",
      "1.9599134922027588 loss -0.0665041450337\n",
      "2.1346490383148193 loss -0.138453013963\n",
      "2.1908118724823 loss -0.0560615538904\n",
      "2.181903839111328 loss 0.0483573115287\n",
      "1.9828460216522217 loss -0.0858664255344\n",
      "1.9887337684631348 loss -0.014825797394\n",
      "2.187124252319336 loss -0.0358096478098\n",
      "1.9084248542785645 loss -0.00437870597846\n",
      "1.8770396709442139 loss -0.0765051303491\n",
      "2.1109042167663574 loss -0.0870830670117\n",
      "1.8766603469848633 loss -0.0870830670117\n",
      "1.8981237411499023 loss -0.0977667544706\n",
      "1.9123046398162842 loss -0.0697978803731\n",
      "1.8892617225646973 loss -0.0851880207504\n",
      "1.9234163761138916 loss -0.0331503719126\n",
      "1.9326987266540527 loss -0.0879275907053\n",
      "1.898040771484375 loss -0.122555772184\n",
      "1.8898491859436035 loss -0.0692336823449\n",
      "1.9590249061584473 loss -0.105484461654\n",
      "2.060631036758423 loss -0.0692336823449\n",
      "2.220353603363037 loss -0.0197801718498\n",
      "1.9747130870819092 loss -0.184058955381\n",
      "1.9225451946258545 loss -0.102797672466\n",
      "2.1493537425994873 loss -0.0797316611004\n",
      "1.9832799434661865 loss -0.117114046469\n",
      "2.1966562271118164 loss -0.0538405374417\n",
      "1.9240188598632812 loss -0.101650736417\n",
      "1.9851324558258057 loss -0.045047851217\n",
      "1.9752049446105957 loss -0.0934504487509\n",
      "1.8904304504394531 loss -0.0742855990646\n",
      "1.995676040649414 loss -0.0792163562663\n",
      "2.182711124420166 loss -0.103982204388\n",
      "2.2595033645629883 loss -0.135805239153\n",
      "1.942941665649414 loss -0.12319274307\n",
      "2.200901746749878 loss -0.0774641309732\n",
      "1.9033658504486084 loss -0.0864705034479\n",
      "1.8680906295776367 loss -0.0955979781264\n",
      "1.910517692565918 loss -0.0358096478098\n",
      "1.9112012386322021 loss -0.038491734636\n",
      "2.0044524669647217 loss -0.0950598929015\n",
      "1.945418357849121 loss -0.035234543452\n",
      "2.188807725906372 loss -0.104418431604\n",
      "2.0596354007720947 loss -0.0841032250358\n",
      "1.9592907428741455 loss -0.162663219073\n",
      "2.5637452602386475 loss -0.0955979781264\n",
      "1.859910249710083 loss -0.0627565093911\n",
      "1.9301397800445557 loss -0.092875417357\n",
      "2.209507703781128 loss -0.0820901514522\n",
      "1.979600429534912 loss -0.126462865313\n",
      "1.9127590656280518 loss -0.0484212447931\n",
      "1.8920214176177979 loss -0.0670383440778\n",
      "2.0446290969848633 loss -0.112813284284\n",
      "1.9070568084716797 loss -0.0197801718498\n",
      "1.9458963871002197 loss -0.0362571714083\n",
      "1.9579033851623535 loss -0.100213959866\n",
      "2.1233935356140137 loss -0.0972625433635\n",
      "1.928635835647583 loss -0.0628469848145\n",
      "1.9141483306884766 loss -0.0984156987341\n",
      "1.9274227619171143 loss -0.0137243023206\n",
      "1.91721773147583 loss -0.0688466684171\n",
      "1.9588894844055176 loss -0.069551069747\n",
      "1.906846046447754 loss -0.0929877833362\n",
      "1.9312717914581299 loss -0.00714100132668\n",
      "2.1439738273620605 loss -0.122040563661\n",
      "1.9114792346954346 loss -0.0927118387897\n",
      "1.9950478076934814 loss -0.059334654635\n",
      "1.9456124305725098 loss -0.0576978278033\n",
      "1.9041047096252441 loss -0.0999979633389\n",
      "1.8824801445007324 loss -0.0126290274086\n",
      "2.095101833343506 loss -0.0675782564573\n",
      "2.25677490234375 loss -0.0501297984862\n",
      "2.2305784225463867 loss -0.0440277841537\n",
      "1.9883482456207275 loss -0.0495838090463\n",
      "1.9838688373565674 loss -0.0302186275653\n",
      "1.8928513526916504 loss -0.0308341800636\n",
      "1.9109294414520264 loss -0.139743244015\n",
      "1.901076078414917 loss -0.0428817617292\n",
      "1.8950722217559814 loss -0.0912001663337\n",
      "1.8585379123687744 loss -0.0763788603524\n",
      "1.9116346836090088 loss -0.129596820645\n",
      "1.8885679244995117 loss -0.100213959866\n",
      "2.1263959407806396 loss -0.0412048213264\n",
      "1.9245975017547607 loss -0.104400409176\n",
      "1.908933162689209 loss -0.14333195396\n",
      "1.9205083847045898 loss -0.0076857861187\n",
      "1.9290952682495117 loss -0.00496689036363\n",
      "2.217933416366577 loss -0.0263668186678\n",
      "2.0220980644226074 loss -0.0375605745343\n",
      "2.146307945251465 loss -0.12090793662\n",
      "1.9362120628356934 loss -0.065911901372\n",
      "1.936361312866211 loss -0.144945536999\n",
      "2.0294015407562256 loss -0.111486150256\n",
      "1.9462153911590576 loss -0.101090948864\n",
      "1.8787472248077393 loss -0.135310689424\n",
      "2.176412343978882 loss -0.0914612730523\n",
      "1.91922926902771 loss -0.018122032655\n",
      "1.8740031719207764 loss -0.176490407062\n",
      "1.9028117656707764 loss -0.0368103788305\n",
      "1.9030230045318604 loss -0.150581585759\n",
      "1.8886864185333252 loss -0.0873580309709\n",
      "2.5322911739349365 loss -0.148354255212\n",
      "2.2620882987976074 loss -0.0950598929015\n",
      "1.906693696975708 loss -0.105484461654\n",
      "1.9980545043945312 loss -0.0697978803731\n",
      "1.9782679080963135 loss -0.055498725006\n",
      "1.8662574291229248 loss -0.056105130691\n",
      "2.166640520095825 loss -0.0908495972473\n",
      "2.0151138305664062 loss -0.130252109708\n",
      "1.9139604568481445 loss 0.000523138021582\n",
      "1.8724334239959717 loss 0.0406524037444\n",
      "1.9618017673492432 loss -0.135811920839\n",
      "1.9107754230499268 loss -0.14140532996\n",
      "1.9472877979278564 loss -0.114277792572\n",
      "1.9288733005523682 loss -0.0934940255514\n",
      "2.3089981079101562 loss -0.153534213601\n",
      "1.8743219375610352 loss -0.0730694000407\n",
      "1.9125115871429443 loss -0.0621714210991\n",
      "1.9360668659210205 loss -0.0759148602085\n",
      "1.878854751586914 loss -0.0390167952541\n",
      "2.059025526046753 loss -0.0132009112551\n",
      "2.119232177734375 loss -0.0384783314586\n",
      "2.095733642578125 loss -0.145160976366\n",
      "1.97562837600708 loss -0.0703685321551\n",
      "1.8942863941192627 loss -0.0313112782991\n",
      "2.209096908569336 loss -0.0390167952541\n",
      "1.986891746520996 loss -0.0808882910299\n",
      "1.9453644752502441 loss -0.0989394837663\n",
      "1.9411883354187012 loss -0.0745582620265\n",
      "2.255173921585083 loss -0.104418431604\n",
      "1.9067716598510742 loss -0.128559795226\n",
      "2.184583902359009 loss -0.132049548567\n",
      "2.1776158809661865 loss -0.109578322485\n",
      "2.159377098083496 loss -0.123081441725\n",
      "1.9820759296417236 loss -0.0543883674915\n",
      "2.178093910217285 loss -0.14852509246\n",
      "1.891432523727417 loss -0.0571622071753\n",
      "1.88694167137146 loss -0.0777089298895\n",
      "1.932133436203003 loss -0.0390167952541\n",
      "1.8945176601409912 loss -0.14333195396\n",
      "1.9700920581817627 loss -0.097726148993\n",
      "1.9506914615631104 loss -0.141746938241\n",
      "1.8519232273101807 loss -0.103871896467\n",
      "1.884427547454834 loss -0.0946264229947\n",
      "2.165431499481201 loss -0.118678183793\n",
      "1.9122178554534912 loss -0.0835312442692\n",
      "1.9782629013061523 loss -0.102280056849\n",
      "1.991415023803711 loss -0.0620101320136\n",
      "2.121047258377075 loss -0.0902467259954\n",
      "2.027583122253418 loss -0.0335324652823\n",
      "1.8798563480377197 loss -0.0557653745944\n",
      "1.949753999710083 loss -0.0727190298522\n",
      "1.894061803817749 loss -0.0319842974191\n",
      "2.0533370971679688 loss -0.0384783314586\n",
      "1.8800013065338135 loss -0.0649345069288\n",
      "2.1620988845825195 loss -0.105205779195\n",
      "1.9774670600891113 loss -0.00612399210195\n",
      "1.9149410724639893 loss -0.068333438029\n",
      "2.2421600818634033 loss -0.128559795226\n",
      "2.1372902393341064 loss -0.0890666432078\n",
      "1.865915298461914 loss -0.0340765998179\n",
      "1.8641622066497803 loss -0.0615676690749\n",
      "2.1246213912963867 loss -0.0865900373488\n",
      "1.9008829593658447 loss -0.0703860647583\n",
      "2.1834468841552734 loss -0.0858664255344\n",
      "2.16695237159729 loss -0.160443506823\n",
      "1.9172377586364746 loss -0.121492119896\n",
      "2.0475220680236816 loss -0.0557653745944\n",
      "2.209203004837036 loss -0.122219873003\n",
      "2.126762628555298 loss -0.0589586235376\n",
      "1.8866541385650635 loss -0.104937885349\n",
      "2.134695529937744 loss -0.148360425709\n",
      "2.2427966594696045 loss -0.0721208720098\n",
      "1.9621937274932861 loss -0.118322030892\n",
      "1.8597986698150635 loss -0.0478475160291\n",
      "1.912693738937378 loss 0.0115939654806\n",
      "1.9137616157531738 loss -0.0225189139807\n",
      "2.180527448654175 loss -0.0884890730573\n",
      "1.914808750152588 loss 0.0335388080564\n",
      "1.9202139377593994 loss -0.0577825399784\n",
      "2.029897451400757 loss -0.060452284031\n",
      "1.8986592292785645 loss -0.153320265401\n",
      "1.8930869102478027 loss -0.101090948864\n",
      "2.0184226036071777 loss 0.0197944557418\n",
      "1.8945832252502441 loss -0.0516948720741\n",
      "1.9525401592254639 loss -0.0862580709034\n",
      "2.2060534954071045 loss -0.0225189139807\n",
      "1.9507296085357666 loss -0.070945801059\n",
      "2.002955198287964 loss -0.115948450709\n",
      "1.9561405181884766 loss -0.081317742885\n",
      "1.8865718841552734 loss -0.159126255471\n",
      "2.2532811164855957 loss -0.115948450709\n",
      "1.8771560192108154 loss -0.050767683583\n",
      "2.019864082336426 loss -0.0505477749992\n",
      "2.122347116470337 loss -0.0626375224311\n",
      "2.032717704772949 loss -0.0120833593845\n",
      "1.8453588485717773 loss -0.108963595926\n",
      "1.876387119293213 loss -0.0885654117926\n",
      "1.8875951766967773 loss -0.0455991092667\n",
      "1.8863112926483154 loss -0.0648428980937\n",
      "1.9795289039611816 loss -0.116525862084\n",
      "1.8863210678100586 loss -0.0714411015991\n",
      "1.9277422428131104 loss -0.0675782564573\n",
      "1.8933625221252441 loss -0.0890772574425\n",
      "1.9100306034088135 loss -0.0718038073954\n",
      "1.9373962879180908 loss -0.0736249228418\n",
      "1.927372694015503 loss -0.115381407307\n",
      "1.852698564529419 loss -0.106605497036\n",
      "1.8847930431365967 loss -0.12090793662\n",
      "1.9624767303466797 loss -0.0940340593793\n",
      "1.8593952655792236 loss -0.112736243651\n",
      "1.852637529373169 loss -0.0802516509777\n",
      "1.9757494926452637 loss -0.148623893226\n",
      "2.1757495403289795 loss -0.0730694000407\n",
      "1.9027879238128662 loss -0.127487854271\n",
      "2.1268630027770996 loss -0.00602081130644\n",
      "1.8953642845153809 loss -0.119771981413\n",
      "1.9058351516723633 loss -0.115638843244\n",
      "1.8558070659637451 loss -0.0769181636803\n",
      "1.885059118270874 loss -0.0395585347883\n",
      "1.8448805809020996 loss -0.060452284031\n",
      "1.877781629562378 loss -0.0406524037444\n",
      "1.9901773929595947 loss -0.121041059743\n",
      "2.0743188858032227 loss -0.0340961162028\n",
      "1.8555965423583984 loss -0.0358096478098\n",
      "1.8756439685821533 loss -0.120788841889\n",
      "1.8925817012786865 loss -0.117625391112\n",
      "2.1787397861480713 loss -0.0802991350031\n",
      "1.9244794845581055 loss -0.0841032250358\n",
      "2.5569405555725098 loss -0.114277792572\n",
      "2.1725404262542725 loss -0.0516704211024\n",
      "1.9036762714385986 loss -0.0407713635208\n",
      "1.8916471004486084 loss -0.10880560338\n",
      "1.863675832748413 loss -0.115948450709\n",
      "1.8976972103118896 loss -0.0533921467724\n",
      "1.8579752445220947 loss -0.134618469241\n",
      "1.8916127681732178 loss -0.0902467259954\n",
      "1.8955557346343994 loss -0.0979144673189\n",
      "1.921990156173706 loss -0.0670383440778\n",
      "2.063081741333008 loss -0.115948450709\n",
      "2.1444132328033447 loss -0.123311606276\n",
      "2.133897304534912 loss -0.0457760601246\n",
      "1.8834035396575928 loss -0.0499935291897\n",
      "1.9205288887023926 loss -0.154997205803\n",
      "1.9174177646636963 loss -0.112696390573\n",
      "1.9570033550262451 loss -0.132957607018\n",
      "1.9284241199493408 loss -0.0419666607379\n",
      "1.9792187213897705 loss -0.0254147068845\n",
      "1.970583200454712 loss -0.0445006302187\n",
      "1.9203853607177734 loss -0.106040281861\n",
      "2.1914303302764893 loss -0.0718038073954\n",
      "2.1740264892578125 loss -0.0467143301753\n",
      "1.919823408126831 loss -0.122555772184\n",
      "2.1652991771698 loss -0.117114046469\n",
      "1.903014898300171 loss -0.121492119896\n",
      "2.0672409534454346 loss -0.110203516252\n",
      "2.216630697250366 loss -0.0390167952541\n",
      "1.945488452911377 loss -0.0374106832607\n",
      "1.9149587154388428 loss -0.148354255212\n",
      "1.9200406074523926 loss -0.00767349595924\n",
      "2.1756551265716553 loss -0.0784571070466\n",
      "1.9760687351226807 loss -0.0461545519182\n",
      "1.9127483367919922 loss -0.0637288837661\n",
      "1.9252471923828125 loss -0.0654087776056\n",
      "2.1900582313537598 loss -0.0814852346285\n",
      "2.2364442348480225 loss -0.148940455883\n",
      "1.9871978759765625 loss -0.0798289259856\n",
      "1.8943121433258057 loss -0.122693767174\n",
      "1.9571926593780518 loss -0.0329307978835\n",
      "1.9539451599121094 loss -0.103871896467\n",
      "2.026322603225708 loss -0.0587839105193\n",
      "2.281940221786499 loss -0.0984156987341\n",
      "2.042287588119507 loss -0.0769181636803\n",
      "1.9403440952301025 loss -0.0511691896873\n",
      "2.1219136714935303 loss -0.0654087776056\n",
      "2.184234142303467 loss -0.0966991564162\n",
      "2.224534749984741 loss -0.0214593655723\n",
      "1.913994550704956 loss -0.0397344293485\n",
      "1.937483787536621 loss -0.147239609672\n",
      "1.9693429470062256 loss -0.11771247078\n",
      "1.9181501865386963 loss -0.0785766360178\n",
      "1.8544762134552002 loss -0.107263522845\n",
      "1.845444917678833 loss -0.0914612730523\n",
      "2.002821683883667 loss -0.105832695994\n",
      "2.111935615539551 loss -0.056105130691\n",
      "2.346569538116455 loss -0.0120833593845\n",
      "1.8831851482391357 loss -0.0335388080564\n",
      "1.934401512145996 loss -0.0528129861336\n",
      "1.8980915546417236 loss -0.128559795226\n",
      "2.0068023204803467 loss -0.0626375224311\n",
      "1.9175426959991455 loss -0.085719376627\n",
      "1.9139907360076904 loss -0.0615924671148\n",
      "2.01072096824646 loss -0.0349325588233\n",
      "1.9846551418304443 loss -0.055498725006\n",
      "1.9855327606201172 loss -0.127974151297\n",
      "1.9402661323547363 loss -0.176972975147\n",
      "2.038374423980713 loss -0.118146728826\n",
      "1.8805246353149414 loss -0.0404222737637\n",
      "2.0239107608795166 loss -0.184058955381\n",
      "1.890681505203247 loss -0.0772189157487\n",
      "1.9313075542449951 loss -0.104937885349\n",
      "1.881563663482666 loss -0.0774641309732\n",
      "2.150602102279663 loss -0.0873580309709\n",
      "1.9293596744537354 loss -0.102219590445\n",
      "2.086696147918701 loss -0.0550660792952\n",
      "1.8965351581573486 loss -0.122555772184\n",
      "1.883742332458496 loss -0.0220544935899\n",
      "2.2133936882019043 loss -0.110992674275\n",
      "1.8922266960144043 loss -0.0241791313382\n",
      "2.0148770809173584 loss -0.103871896467\n",
      "1.9736952781677246 loss -0.100213959866\n",
      "1.9761126041412354 loss -0.0439573019772\n",
      "1.9419691562652588 loss -0.0940340593793\n",
      "2.2537240982055664 loss -0.0439573019772\n",
      "best: \n",
      "{'min_samples_split': 3, 'max_features': 52, 'criterion': 1, 'bootstrap': 1, 'n_estimators': 274, 'min_samples_leaf': 6}\n"
     ]
    }
   ],
   "source": [
    "best = fmin(f_rf, space, algo=tpe.suggest, max_evals=1000, trials=trials2, verbose=1)\n",
    "print ('best: ')\n",
    "print (best)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2017-03-25T03:34:56.906Z"
    }
   },
   "outputs": [],
   "source": [
    "trials._exp_key=str(data_path.basename())\n",
    "len(trials.trials), len(trials2.trials)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2017-03-25T03:34:04.157Z"
    }
   },
   "outputs": [],
   "source": [
    "from bson import json_util\n",
    "import json\n",
    "save_path = Path('../output/hyperopt/{}.json'.format(trials._exp_key))\n",
    "save_path.dirname().makedirs_p()\n",
    "json.dump(trials.trials,open(save_path,'w'), default=json_util.default)\n",
    "save_path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T20:49:01.580712Z",
     "start_time": "2017-03-14T04:49:01.443994+08:00"
    }
   },
   "outputs": [],
   "source": [
    "import collections\n",
    "\n",
    "def flatten(d, parent_key='', sep='_'):\n",
    "    \"\"\"Flatten dicts.\"\"\"\n",
    "    items = []\n",
    "    for k, v in d.items():\n",
    "        new_key = parent_key + sep + k if parent_key else k\n",
    "        if isinstance(v, collections.MutableMapping):\n",
    "            items.extend(flatten(v, new_key, sep=sep).items())\n",
    "        else:\n",
    "            # also flatten lists with one element\n",
    "            if isinstance(v,list) and len(v)==1:\n",
    "                v=v[0]\n",
    "            items.append((new_key, v))\n",
    "    return dict(items)\n",
    "df = pd.DataFrame([flatten(trial) for trial in trials2.trials])\n",
    "# df = pd.DataFrame([flatten(trial) for trial in trial_data])\n",
    "# df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T20:49:02.033997Z",
     "start_time": "2017-03-14T04:49:02.031426+08:00"
    }
   },
   "outputs": [],
   "source": [
    "# df.sort_values('result_metrics_matthews_corrcoef')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T20:49:02.317316Z",
     "start_time": "2017-03-14T04:49:02.307996+08:00"
    }
   },
   "outputs": [],
   "source": [
    "df2=df[[\n",
    "#     'book_time',\n",
    "#  'exp_key',\n",
    "#  'misc_cmd',\n",
    "#  'misc_idxs_bootstrap',\n",
    "#  'misc_idxs_criterion',\n",
    "#  'misc_idxs_max_features',\n",
    "#  'misc_idxs_min_samples_leaf',\n",
    "#  'misc_idxs_min_samples_split',\n",
    "#  'misc_idxs_n_estimators',\n",
    "#  'misc_tid',\n",
    " 'misc_vals_bootstrap',\n",
    " 'misc_vals_criterion',\n",
    " 'misc_vals_max_features',\n",
    " 'misc_vals_min_samples_leaf',\n",
    " 'misc_vals_min_samples_split',\n",
    " 'misc_vals_n_estimators',\n",
    "#  'misc_workdir',\n",
    "#  'owner',\n",
    "#  'refresh_time',\n",
    "#  'result_loss',\n",
    " 'result_metrics_matthews_corrcoef',\n",
    "#  'result_metrics_report_f1-score_avg / total',\n",
    " 'result_metrics_report_f1-score_leak',\n",
    "#  'result_metrics_report_f1-score_no leak',\n",
    "#  'result_metrics_report_precision_avg / total',\n",
    " 'result_metrics_report_precision_leak',\n",
    "#  'result_metrics_report_precision_no leak',\n",
    "#  'result_metrics_report_recall_avg / total',\n",
    " 'result_metrics_report_recall_leak',\n",
    "#  'result_metrics_report_recall_no leak',\n",
    "#  'result_metrics_report_support_avg / total',\n",
    "#  'result_metrics_report_support_leak',\n",
    "#  'result_metrics_report_support_no leak',\n",
    "#  'result_run_time',\n",
    "#  'result_status',\n",
    "#  'spec',\n",
    "#  'state',\n",
    "#  'tid',\n",
    "#  'version'\n",
    "   ]].sort_values('result_metrics_matthews_corrcoef',ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- critenrion = 'entropy'\n",
    "- max_features ~ 100-260\n",
    "- min_samples_leaf=5\n",
    "- min_samples_split=5?\n",
    "- n_estimators=~160\n",
    "- boostrap, yues?\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T20:49:03.599438Z",
     "start_time": "2017-03-14T04:49:03.562607+08:00"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>misc_vals_bootstrap</th>\n",
       "      <th>misc_vals_criterion</th>\n",
       "      <th>misc_vals_max_features</th>\n",
       "      <th>misc_vals_min_samples_leaf</th>\n",
       "      <th>misc_vals_min_samples_split</th>\n",
       "      <th>misc_vals_n_estimators</th>\n",
       "      <th>result_metrics_matthews_corrcoef</th>\n",
       "      <th>result_metrics_report_f1-score_leak</th>\n",
       "      <th>result_metrics_report_precision_leak</th>\n",
       "      <th>result_metrics_report_recall_leak</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>482</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>52</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>274</td>\n",
       "      <td>0.184599</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>708</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>153</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>96</td>\n",
       "      <td>0.184059</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>984</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>170</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>74</td>\n",
       "      <td>0.184059</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>230</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>140</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>132</td>\n",
       "      <td>0.182557</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>635</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>89</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>127</td>\n",
       "      <td>0.180301</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>208</td>\n",
       "      <td>0.179479</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>369</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>142</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>207</td>\n",
       "      <td>0.177196</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.63</td>\n",
       "      <td>0.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>981</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>113</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>208</td>\n",
       "      <td>0.176973</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>783</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>297</td>\n",
       "      <td>0.176490</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.63</td>\n",
       "      <td>0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>318</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>248</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>99</td>\n",
       "      <td>0.175259</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>639</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>107</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>36</td>\n",
       "      <td>0.175259</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>589</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>83</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>286</td>\n",
       "      <td>0.171789</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>647</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>54</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>294</td>\n",
       "      <td>0.162767</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>730</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>274</td>\n",
       "      <td>0.162663</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>238</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>253</td>\n",
       "      <td>0.160757</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>851</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>49</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>136</td>\n",
       "      <td>0.160444</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>620</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>152</td>\n",
       "      <td>0.160117</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>876</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>56</td>\n",
       "      <td>0.159126</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>409</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>190</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>174</td>\n",
       "      <td>0.157980</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>653</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>107</td>\n",
       "      <td>7</td>\n",
       "      <td>2</td>\n",
       "      <td>79</td>\n",
       "      <td>0.157733</td>\n",
       "      <td>0.57</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>600</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>37</td>\n",
       "      <td>0.155506</td>\n",
       "      <td>0.56</td>\n",
       "      <td>0.62</td>\n",
       "      <td>0.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>595</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>131</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>102</td>\n",
       "      <td>0.154997</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>929</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>145</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>57</td>\n",
       "      <td>0.154997</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>801</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>65</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>105</td>\n",
       "      <td>0.153534</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>420</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>141</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>167</td>\n",
       "      <td>0.153320</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>867</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>109</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>196</td>\n",
       "      <td>0.153320</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>473</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>105</td>\n",
       "      <td>0.152185</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>99</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>58</td>\n",
       "      <td>0.152185</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>386</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>75</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>76</td>\n",
       "      <td>0.151765</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>785</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>69</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>212</td>\n",
       "      <td>0.150582</td>\n",
       "      <td>0.59</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>257</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>166</td>\n",
       "      <td>0.008231</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>319</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>107</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>33</td>\n",
       "      <td>0.008231</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>771</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>229</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>270</td>\n",
       "      <td>0.007686</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>945</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>108</td>\n",
       "      <td>0.007673</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>749</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>245</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>165</td>\n",
       "      <td>0.007141</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>842</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>90</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>99</td>\n",
       "      <td>0.006124</td>\n",
       "      <td>0.54</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>901</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>125</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>260</td>\n",
       "      <td>0.006021</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>772</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>254</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>0.004967</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>694</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>157</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>150</td>\n",
       "      <td>0.004379</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>64</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>203</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>58</td>\n",
       "      <td>0.004379</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>481</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>260</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>267</td>\n",
       "      <td>0.002794</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>324</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>248</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>37</td>\n",
       "      <td>0.001655</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>795</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>152</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>-0.000523</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>471</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>255</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>270</td>\n",
       "      <td>-0.002743</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>535</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>151</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>17</td>\n",
       "      <td>-0.004967</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>65</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>38</td>\n",
       "      <td>-0.006665</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>562</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>206</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "      <td>-0.007747</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>499</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>68</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>119</td>\n",
       "      <td>-0.008828</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>207</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>18</td>\n",
       "      <td>-0.011529</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>861</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>193</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>30</td>\n",
       "      <td>-0.011594</td>\n",
       "      <td>0.55</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>555</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>199</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>23</td>\n",
       "      <td>-0.011621</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>368</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>200</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>36</td>\n",
       "      <td>-0.018673</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>869</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>84</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>70</td>\n",
       "      <td>-0.019794</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>109</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>130</td>\n",
       "      <td>-0.022054</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>171</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>266</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>204</td>\n",
       "      <td>-0.024735</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.52</td>\n",
       "      <td>0.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>385</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>58</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.025932</td>\n",
       "      <td>0.39</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>864</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>228</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>-0.033539</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>796</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>160</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>11</td>\n",
       "      <td>-0.040652</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>67</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.041648</td>\n",
       "      <td>0.46</td>\n",
       "      <td>0.51</td>\n",
       "      <td>0.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>690</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>279</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>61</td>\n",
       "      <td>-0.048357</td>\n",
       "      <td>0.49</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.47</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     misc_vals_bootstrap  misc_vals_criterion  misc_vals_max_features  \\\n",
       "482                    1                    1                      52   \n",
       "708                    0                    1                     153   \n",
       "984                    0                    1                     170   \n",
       "230                    0                    1                     140   \n",
       "635                    1                    1                      89   \n",
       "1                      0                    1                      22   \n",
       "369                    0                    1                     142   \n",
       "981                    1                    1                     113   \n",
       "783                    1                    1                      45   \n",
       "318                    1                    1                     248   \n",
       "639                    0                    1                     107   \n",
       "589                    1                    1                      83   \n",
       "647                    0                    1                      54   \n",
       "730                    0                    1                      33   \n",
       "146                    1                    1                     238   \n",
       "851                    0                    1                      49   \n",
       "620                    0                    1                       4   \n",
       "876                    0                    1                       7   \n",
       "409                    1                    1                     190   \n",
       "653                    0                    1                     107   \n",
       "600                    1                    1                       3   \n",
       "595                    1                    1                     131   \n",
       "929                    1                    0                     145   \n",
       "801                    1                    1                      65   \n",
       "420                    1                    1                     141   \n",
       "867                    1                    1                     109   \n",
       "473                    0                    1                      22   \n",
       "45                     1                    1                      99   \n",
       "386                    0                    1                      75   \n",
       "785                    1                    1                      69   \n",
       "..                   ...                  ...                     ...   \n",
       "127                    1                    1                     257   \n",
       "319                    1                    1                     107   \n",
       "771                    1                    1                     229   \n",
       "945                    1                    1                      20   \n",
       "749                    1                    0                     245   \n",
       "842                    0                    1                      90   \n",
       "901                    1                    1                     125   \n",
       "772                    0                    1                     254   \n",
       "694                    0                    0                     157   \n",
       "64                     1                    1                     203   \n",
       "481                    0                    1                     260   \n",
       "324                    0                    0                     248   \n",
       "795                    1                    1                     152   \n",
       "471                    0                    1                     255   \n",
       "535                    0                    1                     151   \n",
       "182                    0                    1                      65   \n",
       "562                    1                    1                     206   \n",
       "499                    0                    1                      68   \n",
       "258                    1                    1                     207   \n",
       "861                    1                    1                     193   \n",
       "555                    0                    1                     199   \n",
       "368                    1                    1                     200   \n",
       "869                    1                    0                      84   \n",
       "220                    1                    1                     109   \n",
       "171                    0                    1                     266   \n",
       "385                    0                    1                      58   \n",
       "864                    0                    1                     228   \n",
       "796                    1                    1                     160   \n",
       "353                    1                    1                      67   \n",
       "690                    0                    1                     279   \n",
       "\n",
       "     misc_vals_min_samples_leaf  misc_vals_min_samples_split  \\\n",
       "482                           6                            3   \n",
       "708                           7                            6   \n",
       "984                           5                            6   \n",
       "230                           6                            6   \n",
       "635                           7                            6   \n",
       "1                             6                            1   \n",
       "369                           5                            0   \n",
       "981                           7                            6   \n",
       "783                           6                            7   \n",
       "318                           7                            3   \n",
       "639                           7                            2   \n",
       "589                           6                            3   \n",
       "647                           7                            2   \n",
       "730                           7                            6   \n",
       "146                           2                            6   \n",
       "851                           5                            3   \n",
       "620                           6                            6   \n",
       "876                           6                            6   \n",
       "409                           6                            3   \n",
       "653                           7                            2   \n",
       "600                           2                            6   \n",
       "595                           6                            3   \n",
       "929                           6                            6   \n",
       "801                           6                            7   \n",
       "420                           6                            3   \n",
       "867                           5                            3   \n",
       "473                           6                            3   \n",
       "45                            7                            1   \n",
       "386                           3                            4   \n",
       "785                           6                            7   \n",
       "..                          ...                          ...   \n",
       "127                           8                            1   \n",
       "319                           7                            3   \n",
       "771                           4                            4   \n",
       "945                           6                            6   \n",
       "749                           1                            1   \n",
       "842                           4                            4   \n",
       "901                           1                            6   \n",
       "772                           6                            0   \n",
       "694                           7                            6   \n",
       "64                            0                            8   \n",
       "481                           1                            4   \n",
       "324                           7                            3   \n",
       "795                           6                            7   \n",
       "471                           2                            0   \n",
       "535                           5                            6   \n",
       "182                           3                            5   \n",
       "562                           7                            3   \n",
       "499                           6                            5   \n",
       "258                           6                            6   \n",
       "861                           7                            7   \n",
       "555                           1                            6   \n",
       "368                           2                            1   \n",
       "869                           1                            6   \n",
       "220                           1                            6   \n",
       "171                           4                            3   \n",
       "385                           1                            2   \n",
       "864                           6                            1   \n",
       "796                           6                            7   \n",
       "353                           4                            8   \n",
       "690                           3                            6   \n",
       "\n",
       "     misc_vals_n_estimators  result_metrics_matthews_corrcoef  \\\n",
       "482                     274                          0.184599   \n",
       "708                      96                          0.184059   \n",
       "984                      74                          0.184059   \n",
       "230                     132                          0.182557   \n",
       "635                     127                          0.180301   \n",
       "1                       208                          0.179479   \n",
       "369                     207                          0.177196   \n",
       "981                     208                          0.176973   \n",
       "783                     297                          0.176490   \n",
       "318                      99                          0.175259   \n",
       "639                      36                          0.175259   \n",
       "589                     286                          0.171789   \n",
       "647                     294                          0.162767   \n",
       "730                     274                          0.162663   \n",
       "146                     253                          0.160757   \n",
       "851                     136                          0.160444   \n",
       "620                     152                          0.160117   \n",
       "876                      56                          0.159126   \n",
       "409                     174                          0.157980   \n",
       "653                      79                          0.157733   \n",
       "600                      37                          0.155506   \n",
       "595                     102                          0.154997   \n",
       "929                      57                          0.154997   \n",
       "801                     105                          0.153534   \n",
       "420                     167                          0.153320   \n",
       "867                     196                          0.153320   \n",
       "473                     105                          0.152185   \n",
       "45                       58                          0.152185   \n",
       "386                      76                          0.151765   \n",
       "785                     212                          0.150582   \n",
       "..                      ...                               ...   \n",
       "127                     166                          0.008231   \n",
       "319                      33                          0.008231   \n",
       "771                     270                          0.007686   \n",
       "945                     108                          0.007673   \n",
       "749                     165                          0.007141   \n",
       "842                      99                          0.006124   \n",
       "901                     260                          0.006021   \n",
       "772                      22                          0.004967   \n",
       "694                     150                          0.004379   \n",
       "64                       58                          0.004379   \n",
       "481                     267                          0.002794   \n",
       "324                      37                          0.001655   \n",
       "795                       6                         -0.000523   \n",
       "471                     270                         -0.002743   \n",
       "535                      17                         -0.004967   \n",
       "182                      38                         -0.006665   \n",
       "562                      19                         -0.007747   \n",
       "499                     119                         -0.008828   \n",
       "258                      18                         -0.011529   \n",
       "861                      30                         -0.011594   \n",
       "555                      23                         -0.011621   \n",
       "368                      36                         -0.018673   \n",
       "869                      70                         -0.019794   \n",
       "220                     130                         -0.022054   \n",
       "171                     204                         -0.024735   \n",
       "385                       0                         -0.025932   \n",
       "864                       3                         -0.033539   \n",
       "796                      11                         -0.040652   \n",
       "353                       4                         -0.041648   \n",
       "690                      61                         -0.048357   \n",
       "\n",
       "     result_metrics_report_f1-score_leak  \\\n",
       "482                                 0.60   \n",
       "708                                 0.60   \n",
       "984                                 0.60   \n",
       "230                                 0.59   \n",
       "635                                 0.62   \n",
       "1                                   0.59   \n",
       "369                                 0.58   \n",
       "981                                 0.59   \n",
       "783                                 0.58   \n",
       "318                                 0.60   \n",
       "639                                 0.60   \n",
       "589                                 0.61   \n",
       "647                                 0.57   \n",
       "730                                 0.59   \n",
       "146                                 0.58   \n",
       "851                                 0.59   \n",
       "620                                 0.58   \n",
       "876                                 0.56   \n",
       "409                                 0.60   \n",
       "653                                 0.57   \n",
       "600                                 0.56   \n",
       "595                                 0.60   \n",
       "929                                 0.60   \n",
       "801                                 0.60   \n",
       "420                                 0.58   \n",
       "867                                 0.58   \n",
       "473                                 0.59   \n",
       "45                                  0.59   \n",
       "386                                 0.61   \n",
       "785                                 0.59   \n",
       "..                                   ...   \n",
       "127                                 0.51   \n",
       "319                                 0.51   \n",
       "771                                 0.51   \n",
       "945                                 0.49   \n",
       "749                                 0.52   \n",
       "842                                 0.54   \n",
       "901                                 0.50   \n",
       "772                                 0.53   \n",
       "694                                 0.51   \n",
       "64                                  0.51   \n",
       "481                                 0.53   \n",
       "324                                 0.52   \n",
       "795                                 0.53   \n",
       "471                                 0.51   \n",
       "535                                 0.50   \n",
       "182                                 0.49   \n",
       "562                                 0.49   \n",
       "499                                 0.50   \n",
       "258                                 0.53   \n",
       "861                                 0.55   \n",
       "555                                 0.49   \n",
       "368                                 0.52   \n",
       "869                                 0.50   \n",
       "220                                 0.49   \n",
       "171                                 0.50   \n",
       "385                                 0.39   \n",
       "864                                 0.49   \n",
       "796                                 0.50   \n",
       "353                                 0.46   \n",
       "690                                 0.49   \n",
       "\n",
       "     result_metrics_report_precision_leak  result_metrics_report_recall_leak  \n",
       "482                                  0.62                               0.59  \n",
       "708                                  0.62                               0.59  \n",
       "984                                  0.62                               0.59  \n",
       "230                                  0.62                               0.56  \n",
       "635                                  0.61                               0.63  \n",
       "1                                    0.62                               0.55  \n",
       "369                                  0.63                               0.53  \n",
       "981                                  0.62                               0.57  \n",
       "783                                  0.63                               0.54  \n",
       "318                                  0.62                               0.58  \n",
       "639                                  0.62                               0.58  \n",
       "589                                  0.61                               0.61  \n",
       "647                                  0.62                               0.53  \n",
       "730                                  0.61                               0.56  \n",
       "146                                  0.61                               0.54  \n",
       "851                                  0.61                               0.58  \n",
       "620                                  0.61                               0.55  \n",
       "876                                  0.62                               0.52  \n",
       "409                                  0.60                               0.60  \n",
       "653                                  0.62                               0.53  \n",
       "600                                  0.62                               0.51  \n",
       "595                                  0.60                               0.59  \n",
       "929                                  0.60                               0.59  \n",
       "801                                  0.60                               0.60  \n",
       "420                                  0.61                               0.56  \n",
       "867                                  0.61                               0.56  \n",
       "473                                  0.61                               0.57  \n",
       "45                                   0.61                               0.57  \n",
       "386                                  0.60                               0.62  \n",
       "785                                  0.60                               0.58  \n",
       "..                                    ...                                ...  \n",
       "127                                  0.53                               0.49  \n",
       "319                                  0.53                               0.49  \n",
       "771                                  0.53                               0.50  \n",
       "945                                  0.53                               0.46  \n",
       "749                                  0.53                               0.50  \n",
       "842                                  0.53                               0.55  \n",
       "901                                  0.53                               0.47  \n",
       "772                                  0.53                               0.52  \n",
       "694                                  0.53                               0.48  \n",
       "64                                   0.53                               0.48  \n",
       "481                                  0.53                               0.54  \n",
       "324                                  0.53                               0.51  \n",
       "795                                  0.53                               0.52  \n",
       "471                                  0.53                               0.50  \n",
       "535                                  0.53                               0.48  \n",
       "182                                  0.53                               0.45  \n",
       "562                                  0.53                               0.46  \n",
       "499                                  0.52                               0.47  \n",
       "258                                  0.52                               0.53  \n",
       "861                                  0.52                               0.57  \n",
       "555                                  0.52                               0.45  \n",
       "368                                  0.52                               0.51  \n",
       "869                                  0.52                               0.48  \n",
       "220                                  0.52                               0.46  \n",
       "171                                  0.52                               0.48  \n",
       "385                                  0.51                               0.32  \n",
       "864                                  0.51                               0.47  \n",
       "796                                  0.51                               0.49  \n",
       "353                                  0.51                               0.41  \n",
       "690                                  0.50                               0.47  \n",
       "\n",
       "[1000 rows x 10 columns]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-03-13T20:49:30.438434Z",
     "start_time": "2017-03-14T04:49:30.432507+08:00"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "misc_vals_bootstrap                       0.540000\n",
       "misc_vals_criterion                       0.980000\n",
       "misc_vals_max_features                  102.040000\n",
       "misc_vals_min_samples_leaf                5.840000\n",
       "misc_vals_min_samples_split               4.380000\n",
       "misc_vals_n_estimators                  160.840000\n",
       "result_metrics_matthews_corrcoef          0.157871\n",
       "result_metrics_report_f1-score_leak       0.584600\n",
       "result_metrics_report_precision_leak      0.609400\n",
       "result_metrics_report_recall_leak         0.564200\n",
       "dtype: float64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2[:50].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-04-26T03:03:28.268144Z",
     "start_time": "2017-04-26T11:03:28.263313+08:00"
    }
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'df' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-5-246e620b3f38>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0miloc\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m482\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'df' is not defined"
     ]
    }
   ],
   "source": [
    "df.iloc[482]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (with sys packages)",
   "language": "python",
   "name": "py3syspck"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.4.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}