├── README.md
├── WindTurbine_demo.py
├── WindTurbine.py
└── WindTurbine_fvsff_MIQ-10_cv1-run1.ipynb


/README.md:
--------------------------------------------------------------------------------
1 | # Wind_Turbine
2 | Import, clean, and prepare data and conduct machine learning classification for fault detection in a wind turbine
3 | 
4 | 
5 | The results are published in a <a href='https://github.com/LilyHu/Academic_Papers/blob/master/Diagnosing_Faults_Using_Machine_Learning_Techniques_Applied_to_Operational_Data.pdf'>peer-reviewed article</a>. 
6 | 
7 | 


--------------------------------------------------------------------------------
/WindTurbine_demo.py:
--------------------------------------------------------------------------------
 1 | import WindTurbine as wt
 2 | # Initiate the instance
 3 | edata = wt.EnerconWindTurbineData()
 4 | edata.import_data()
 5 | edata.clean_data()
 6 | edata.create_new_features()
 7 | edata.create_lagged_features(6)
 8 | edata.create_mean_std_features()
 9 | edata.derived_features = pd.concat([edata.new_features, edata.lagged_features, edata.mean_std], axis=1, join='inner')
10 | # To save memory, delete the redundant dataframes
11 | edata.derived_features.dropna(inplace=True)
12 | edata.xdata = pd.concat([edata.derived_features, edata.scada_data], axis=1, join='inner')
13 | edata.xdata.to_pickle('xdata_all_unscaled.p')


--------------------------------------------------------------------------------
/WindTurbine.py:
--------------------------------------------------------------------------------
  1 | import numpy as np
  2 | import datetime as dt
  3 | import pandas as pd
  4 | 
  5 | class EnerconWindTurbineData(object):
  6 |     """
  7 |     Imports the data and returns arrays of SCADA & status data by
  8 |     calling import_data().
  9 | 
 10 |     Parameters
 11 |     ----------
 12 |     scada_data_file: str, optional
 13 |         The raw SCADA data csv file.
 14 |     status_data_wec_file: str, optional
 15 |         The status/fault csv file for the WEC
 16 |     status_data_rtu_file: str, optional
 17 |         The status/fault csv file for the RTU
 18 |     warning_data_wec_file: str, optional
 19 |         The warning/information csv file for the WEC
 20 |     warning_data_rtu_file: str, optional
 21 |         The warning/information csv file for the RTU
 22 |         
 23 |     Notes
 24 |     -----
 25 |     Both status_data_wec.csv & status_data_rtu.csv originally come from
 26 |     pes_extrainfo.csv, filtered according to their plant number.
 27 |     SCADA_data.csv contains the wsd, 03d and 04d data files all
 28 |     combined together.
 29 |     """
 30 | 
 31 |     def __init__(
 32 |             self, 
 33 |             scada_data_file='../Data/Source Data/SCADA_data.csv',
 34 |             status_data_wec_file='../Data/Source Data/status_data_wec.csv',
 35 |             status_data_rtu_file='../Data/Source Data/status_data_rtu.csv',
 36 |             warning_data_wec_file='../Data/Source Data/warning_data_wec.csv',
 37 |             warning_data_rtu_file='../Data/Source Data/warning_data_rtu.csv'):
 38 | 
 39 |         self.scada_data_file = scada_data_file
 40 |         self.status_data_wec_file = status_data_wec_file
 41 |         self.status_data_rtu_file = status_data_rtu_file
 42 |         self.warning_data_wec_file = warning_data_wec_file
 43 |         self.warning_data_rtu_file = warning_data_rtu_file
 44 |         
 45 |         # Pandas dataframe of the above
 46 |         self.status_data_wec = []
 47 |         
 48 |         # Pandas dataframe of all the scada data
 49 |         self.scada_data = []
 50 |         # Pandas dataframe of all the new features
 51 |         self.new_features = []
 52 |         # Pandas Series of labels
 53 |         self.ylabels = []
 54 | 
 55 | 	"""
 56 | 	This imports the data, and returns arrays of SCADA, status &
 57 | 	warning data. Dates are converted to unix time, and strings are
 58 | 	encoded in the correct format (unicode). Two new fields,
 59 | 	"Inverter_averages" and "Inverter_std_dev", are also added to
 60 | 	the SCADA data. These are the average and standard deviation of
 61 | 	all Inverter Temperature fields.
 62 | 
 63 | 	WARNING: The data is not every 10 minutes. There are errors, 
 64 | 	large gaps in data, some values in between 10 minutes, and double values
 65 | 	for the same time
 66 | 
 67 | 	Set's the following fields
 68 | 	-------
 69 | 	self.scada_data: ndarray
 70 | 		The imported and correctly formatted SCADA data
 71 | 	self.status_data_wec: ndarray
 72 | 		The imported and correctly formatted WEC status data
 73 | 	self.status_data_rtu: ndarray
 74 | 		The imported and correctly formatted RTU status data
 75 | 	self.warning_data_wec: ndarray
 76 | 		The imported and correctly formatted WEC warning data
 77 | 	self.warning_data_rtu: ndarray
 78 | 		The imported and correctly formatted RTU warning data
 79 | 	"""
 80 |     def import_data(self):
 81 |         
 82 |         self.scada_data = np.genfromtxt(
 83 |             open(self.scada_data_file, 'rb'), dtype=(
 84 |                 '<U19', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4',
 85 |                 '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4',
 86 |                 '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4',
 87 |                 '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4',
 88 |                 '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4',
 89 |                 '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4',
 90 |                 '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4', '<f4'),
 91 |             delimiter=",", names=True)
 92 |         
 93 |         self.status_data_wec = np.genfromtxt(
 94 |             open(self.status_data_wec_file, 'rb'), dtype=(
 95 |                 '<U19', '<i4', '<i4', '<U9', '<U63', '<i4', '|b1', '|b1',
 96 |                 '<f4'), delimiter=",", names=True)
 97 | 
 98 |         self.status_data_rtu = np.genfromtxt(
 99 |             open(self.status_data_rtu_file, 'rb'), dtype=(
100 |                 '<U19', '<i4', '<i4', '<U9', '<U63', '<i4', '|b1', '|b1',
101 |                 '<f4'), delimiter=",", names=True)
102 | 
103 |         self.warning_data_wec = np.genfromtxt(
104 |             open(self.warning_data_wec_file, 'rb'), dtype=(
105 |                 '<U19', '<i4', '<i4', '<U9', '<U63', '|b1', '<f4'),
106 |             delimiter=",", names=True)
107 | 
108 |         self.warning_data_rtu = np.genfromtxt(
109 |             open(self.warning_data_rtu_file, 'rb'), dtype=(
110 |                 '<U19', '<i4', '<i4', '<U9', '<U63', '|b1', '<f4'),
111 |             delimiter=",", names=True)
112 | 
113 |         self.scada_data = pd.DataFrame(self.scada_data)
114 | 
115 |         # Add 2 extra columns to scada - Inverter_averages and Inverter_std_dev - as features
116 |         inverters = np.array([
117 |             'CS101__Sys_1_inverter_1_cabinet_temp',
118 |             'CS101__Sys_1_inverter_2_cabinet_temp',
119 |             'CS101__Sys_1_inverter_3_cabinet_temp',
120 |             'CS101__Sys_1_inverter_4_cabinet_temp',
121 |             'CS101__Sys_1_inverter_5_cabinet_temp',
122 |             'CS101__Sys_1_inverter_6_cabinet_temp',
123 |             'CS101__Sys_1_inverter_7_cabinet_temp',
124 |             'CS101__Sys_2_inverter_1_cabinet_temp',
125 |             'CS101__Sys_2_inverter_2_cabinet_temp',
126 |             'CS101__Sys_2_inverter_3_cabinet_temp',
127 |             'CS101__Sys_2_inverter_4_cabinet_temp'])
128 |         means = pd.Series(self.scada_data[inverters].mean(axis=1), name='Inverter_averages')
129 |         stds = pd.Series(self.scada_data[inverters].std(axis=1), name='Inverter_std_dev')
130 |     
131 | 	# Save the scada data and the inverter means and standard deviations as "scada_data"
132 |         self.scada_data = pd.concat([self.scada_data, means, stds], axis=1)
133 | 		
134 |         # Convert the unix time in the "Time" column to pandas datetime and set the 'Time' column to be the index
135 |         self.scada_data["Time"] = pd.to_datetime(self.scada_data["Time"], format='%d/%m/%Y %H:%M:%S')- pd.Timedelta('1h')
136 |         self.scada_data = self.scada_data.set_index("Time", drop=True)
137 |         
138 | 	# Sort the data by timestamp
139 |         self.scada_data.sort_index(inplace=True)
140 |         
141 | 	# Convert the unix time in the "Time" column to pandas datetime and set the 'Time' column to be the index
142 |         self.status_data_wec = pd.DataFrame(self.status_data_wec)
143 |         self.status_data_wec["Time"] = pd.to_datetime(self.status_data_wec["Time"], format='%d/%m/%Y %H:%M:%S')- pd.Timedelta('1h')
144 |         self.status_data_rtu = pd.DataFrame(self.status_data_rtu)
145 |         self.status_data_rtu["Time"] = pd.to_datetime(self.status_data_rtu["Time"], format='%d/%m/%Y %H:%M:%S')- pd.Timedelta('1h')
146 |         self.warning_data_wec = pd.DataFrame(self.warning_data_wec)
147 |         self.warning_data_wec["Time"] = pd.to_datetime(self.warning_data_wec["Time"], format='%d/%m/%Y %H:%M:%S')- pd.Timedelta('1h')
148 |         self.warning_data_rtu = pd.DataFrame(self.warning_data_rtu)
149 |         self.warning_data_rtu["Time"] = pd.to_datetime(self.warning_data_rtu["Time"], format='%d/%m/%Y %H:%M:%S')- pd.Timedelta('1h')
150 | 
151 | 	'''
152 | 	Optionally inport the data from pickle files
153 | 	'''
154 |     def import_from_pickle_files(self):
155 |         self.scada_data = pd.read_pickle('scada_data')
156 |         # Reload the expert features, mean features, and std features
157 |         self.new_features = pd.read_pickle('expert_')
158 |         self.mean_std = pd.read_pickle('mean_std')
159 |         pd.DataFrame.dropna(self.mean_std, inplace=True)
160 |         #edata.new_features = pd.concat([edata.new_features, edata.mean_std], axis=1)
161 |         self.create_lagged_features(6)
162 |         # edata.derived_features = pd.concat(axis=1, join='inner')
163 |         self.derived_features = pd.concat([self.new_features, self.mean_std, self.lagged_features], axis=1, join='inner')
164 |         self.status_data_wec = pd.read_pickle('status_data_wec')
165 |         self.status_data_rtu = pd.read_pickle('status_data_rtu')
166 |         self.warning_data_wec = pd.read_pickle('warning_data_wec')
167 |         self.warning_data_rtu = pd.read_pickle('warning_data_rtu')
168 |         
169 |         self.create_labels()
170 |     
171 | 	'''
172 | 	Removes data that is not on the 10 min dot
173 | 	Replaces repetitive timestamps with the mean
174 | 	Removes the known faulty data
175 | 	'''    
176 |     def clean_data(self):
177 |         
178 |         # Keep only the 10 minute samples
179 |         self.scada_data = self.scada_data[self.scada_data.index.minute % 10 == 0]
180 |         
181 |         # Check for identical timestamps and take the average
182 |         self.scada_data.sort_index(inplace=True)
183 |         start_of_dupes = 0
184 |         for i in range(1, len(self.scada_data) + 1):
185 |             if i == len(self.scada_data) or self.scada_data.index[i] != self.scada_data.index[i - 1]:
186 |                 if i - start_of_dupes > 1:
187 |                     average = self.scada_data.iloc[start_of_dupes:i].mean()
188 |                     self.scada_data.iloc[start_of_dupes] = average
189 |                 start_of_dupes = i
190 |         self.scada_data["___INDEX___"] = self.scada_data.index
191 |         self.scada_data.drop_duplicates(subset="___INDEX___", keep="first", inplace=True)
192 |         self.scada_data.drop("___INDEX___", axis=1, inplace=True)
193 |         
194 |         # Remove the faulty blade temps
195 |         self.scada_data.drop(['CS101__Blade_A_temp', 'CS101__Blade_B_temp', 'CS101__Blade_C_temp'], axis=1, inplace=True)
196 |         
197 | 		# Remove the faulty inverters
198 |         self.scada_data.drop([ 'CS101__Sys_2_inverter_5_cabinet_temp', 'CS101__Sys_2_inverter_6_cabinet_temp','CS101__Sys_2_inverter_7_cabinet_temp'], axis=1, inplace=True)
199 |         self.scada_data.drop(['WEC_Operating_Hours', 'WEC_Production_kWh', 'WEC_Production_minutes'], axis=1, inplace=True)
200 |         self.scada_data.drop(['Error'], axis=1, inplace=True)
201 |      
202 | 	'''
203 | 	Create new engineering features from the scada_data
204 | 	'''
205 |     def create_new_features(self):
206 |         # Create new dataframe for new features
207 |         self.new_features = pd.DataFrame(index=self.scada_data.index)
208 | 
209 |         # Calculate averages
210 |         self.new_features['Avg_Sys_1_inverters_cabinet_temp']  = self.scada_data[['CS101__Sys_1_inverter_1_cabinet_temp',
211 |                'CS101__Sys_1_inverter_2_cabinet_temp',
212 |                'CS101__Sys_1_inverter_3_cabinet_temp',
213 |                'CS101__Sys_1_inverter_4_cabinet_temp',
214 |                'CS101__Sys_1_inverter_5_cabinet_temp',
215 |                'CS101__Sys_1_inverter_6_cabinet_temp',
216 |                'CS101__Sys_1_inverter_7_cabinet_temp']].mean(axis=1)
217 | 
218 |         self.new_features['Avg_Sys_2_inverters_cabinet_temp'] = self.scada_data[['CS101__Sys_2_inverter_1_cabinet_temp',
219 |                'CS101__Sys_2_inverter_2_cabinet_temp',
220 |                'CS101__Sys_2_inverter_3_cabinet_temp',
221 |                'CS101__Sys_2_inverter_4_cabinet_temp']].mean(axis=1)
222 | 
223 |         self.new_features['Avg_bearing_temp'] = self.scada_data[['CS101__Front_bearing_temp', 'CS101__Rear_bearing_temp',]].mean(axis=1)
224 | 
225 |         self.new_features['Avg_pitch_cabinet_blade_temp'] = self.scada_data[['CS101__Pitch_cabinet_blade_A_temp',
226 |                'CS101__Pitch_cabinet_blade_B_temp',
227 |                'CS101__Pitch_cabinet_blade_C_temp',]].mean(axis=1)
228 | 
229 |         self.new_features['Avg_rotor_temp'] = self.scada_data[[ 'CS101__Rotor_temp_1',
230 |                'CS101__Rotor_temp_2']].mean(axis=1)
231 | 
232 |         self.new_features['Avg_stator_temp'] = self.scada_data[['CS101__Stator_temp_1', 'CS101__Stator_temp_2',]].mean(axis=1)
233 | 
234 |         self.new_features['Avg_nacelle_ambient_temp'] = self.scada_data[['CS101__Nacelle_ambient_temp_1', 
235 |                                                                            'CS101__Nacelle_ambient_temp_2']].mean(axis=1)
236 |         
237 |         # Create features based on differences
238 | 
239 |         # Max and Min of (wind speed, rotation, power, reactive power)
240 |         self.new_features['Dif_max_min_windspeed'] = self.scada_data['WEC__max_windspeed'] - self.scada_data['WEC__min_windspeed']
241 |         self.new_features['Dif_max_min_rotation'] = self.scada_data['WEC_max_Rotation'] - self.scada_data['WEC_min_Rotation']
242 |         self.new_features['Dif_max_min_Power'] = self.scada_data['WEC_max_Power'] - self.scada_data['WEC_min_Power']
243 |         self.new_features['Dif_max_min_reactive_Power'] = self.scada_data['WEC_max_reactive_Power'] - self.scada_data['WEC_min_reactive_Power']
244 | 
245 |         # Max and Average of (wind speed, rotation, power, reactive power)
246 |         self.new_features['Dif_max_avg_windspeed'] = self.scada_data['WEC__max_windspeed'] - self.scada_data['WEC_ava_windspeed']
247 |         self.new_features['Dif_max_avg_rotation'] = self.scada_data['WEC_max_Rotation'] - self.scada_data['WEC_ava_Rotation']
248 |         self.new_features['Dif_max_avg_Power'] = self.scada_data['WEC_max_Power'] - self.scada_data['WEC_ava_Power']
249 |         self.new_features['Dif_max_avg_reactive_Power'] = self.scada_data['WEC_max_reactive_Power'] - self.scada_data['WEC_ava_reactive_Power']
250 | 
251 |         # Min and Average of (wind speed, rotation, power, reactive power)\\
252 |         self.new_features['Dif_avg_min_windspeed'] = self.scada_data['WEC_ava_windspeed'] - self.scada_data['WEC__min_windspeed']
253 |         self.new_features['Dif_avg_min_rotation'] = self.scada_data['WEC_ava_Rotation'] - self.scada_data['WEC_min_Rotation']
254 |         self.new_features['Dif_avg_min_Power'] = self.scada_data['WEC_ava_Power'] - self.scada_data['WEC_min_Power']
255 |         self.new_features['Dif_avg_min_reactive_Power'] = self.scada_data['WEC_ava_reactive_Power'] - self.scada_data['WEC_min_reactive_Power']
256 | 
257 |         # Available Power (from wind, technical reasons, force majeure reasons, force external reasons)\\
258 |         self.new_features['Diff_P_wind_P_technical'] = self.scada_data['WEC_ava_available_P_from_wind'] - self.scada_data['WEC_ava_available_P_technical_reasons']
259 |         self.new_features['Diff_P_wind_P_majeure'] = self.scada_data['WEC_ava_available_P_from_wind'] - self.scada_data['WEC_ava_Available_P_force_majeure_reasons']
260 |         self.new_features['Diff_P_wind_P_technical'] = self.scada_data['WEC_ava_available_P_from_wind'] - self.scada_data['WEC_ava_Available_P_force_external_reasons']
261 |         self.new_features['Diff_P_technical_P_majeure'] = self.scada_data['WEC_ava_available_P_technical_reasons'] - self.scada_data['WEC_ava_Available_P_force_majeure_reasons']
262 |         self.new_features['Diff_P_technical_P_external'] = self.scada_data['WEC_ava_available_P_technical_reasons'] - self.scada_data['WEC_ava_Available_P_force_external_reasons']
263 |         self.new_features['Diff_P_majeure_P_external'] = self.scada_data['WEC_ava_Available_P_force_majeure_reasons'] - self.scada_data['WEC_ava_Available_P_force_external_reasons']
264 | 
265 |         # Average Power and Available power (from wind, technical reasons, force majeure reasons, force external reasons)\\
266 |         self.new_features['Diff_avg_Power_P_wind'] = self.scada_data['WEC_ava_Power'] - self.scada_data['WEC_ava_available_P_from_wind']
267 |         self.new_features['Diff_avg_Power_P_wind'] = self.scada_data['WEC_ava_Power'] - self.scada_data['WEC_ava_available_P_technical_reasons']
268 |         self.new_features['Diff_avg_Power_P_wind'] = self.scada_data['WEC_ava_Power'] - self.scada_data['WEC_ava_Available_P_force_majeure_reasons']
269 |         self.new_features['Diff_avg_Power_P_wind'] = self.scada_data['WEC_ava_Power'] - self.scada_data['WEC_ava_Available_P_force_external_reasons']
270 | 
271 |         # Inverter Cabinet Temperatures
272 |         # Inverter Cabinet Temperatures and Average Inverter Cabinet Temperature by system
273 |         self.new_features['Diff_Avg_Sys_1_inverter_1']  = self.scada_data['CS101__Sys_1_inverter_1_cabinet_temp'] - self.new_features['Avg_Sys_1_inverters_cabinet_temp']
274 |         self.new_features['Diff_Avg_Sys_1_inverter_2']  = self.scada_data['CS101__Sys_1_inverter_2_cabinet_temp'] - self.new_features['Avg_Sys_1_inverters_cabinet_temp']
275 |         self.new_features['Diff_Avg_Sys_1_inverter_3']  = self.scada_data['CS101__Sys_1_inverter_3_cabinet_temp'] - self.new_features['Avg_Sys_1_inverters_cabinet_temp']
276 |         self.new_features['Diff_Avg_Sys_1_inverter_4']  = self.scada_data['CS101__Sys_1_inverter_4_cabinet_temp'] - self.new_features['Avg_Sys_1_inverters_cabinet_temp']
277 |         self.new_features['Diff_Avg_Sys_1_inverter_5']  = self.scada_data['CS101__Sys_1_inverter_5_cabinet_temp'] - self.new_features['Avg_Sys_1_inverters_cabinet_temp']
278 |         self.new_features['Diff_Avg_Sys_1_inverter_6']  = self.scada_data['CS101__Sys_1_inverter_6_cabinet_temp'] - self.new_features['Avg_Sys_1_inverters_cabinet_temp']
279 |         self.new_features['Diff_Avg_Sys_1_inverter_7']  = self.scada_data['CS101__Sys_1_inverter_7_cabinet_temp'] - self.new_features['Avg_Sys_1_inverters_cabinet_temp']
280 |         
281 |         self.new_features['Diff_Avg_Sys_2_inverter_1']  = self.scada_data['CS101__Sys_2_inverter_1_cabinet_temp'] - self.new_features['Avg_Sys_2_inverters_cabinet_temp']
282 |         self.new_features['Diff_Avg_Sys_2_inverter_2']  = self.scada_data['CS101__Sys_2_inverter_2_cabinet_temp'] - self.new_features['Avg_Sys_2_inverters_cabinet_temp']
283 |         self.new_features['Diff_Avg_Sys_2_inverter_3']  = self.scada_data['CS101__Sys_2_inverter_3_cabinet_temp'] - self.new_features['Avg_Sys_2_inverters_cabinet_temp']
284 |         self.new_features['Diff_Avg_Sys_2_inverter_4']  = self.scada_data['CS101__Sys_2_inverter_4_cabinet_temp'] - self.new_features['Avg_Sys_2_inverters_cabinet_temp']
285 |                                                                           
286 |         # Front and Rear Bearing Temperature
287 |         self.new_features['Diff_font_rear_bearing'] = self.scada_data['CS101__Front_bearing_temp'] - self.scada_data['CS101__Rear_bearing_temp']
288 | 
289 |         # Average Bearing Temperature and (Front and Rear Bearing Temperature)
290 |         self.new_features['Diff_font_avg_bearing'] = self.scada_data['CS101__Front_bearing_temp'] - self.new_features['Avg_bearing_temp']
291 |         self.new_features['Diff_rear_avg_bearing'] = self.scada_data['CS101__Rear_bearing_temp'] - self.new_features['Avg_bearing_temp']
292 | 
293 |         # Pitch Cabinet Blade Temperatures
294 |         self.new_features['Diff_cabinet_A_B_temp']= self.scada_data['CS101__Pitch_cabinet_blade_A_temp'] - self.scada_data['CS101__Pitch_cabinet_blade_B_temp']
295 |         self.new_features['Diff_cabinet_A_C_temp']= self.scada_data['CS101__Pitch_cabinet_blade_A_temp'] - self.scada_data['CS101__Pitch_cabinet_blade_C_temp']
296 |         self.new_features['Diff_cabinet_B_C_temp']= self.scada_data['CS101__Pitch_cabinet_blade_B_temp'] - self.scada_data['CS101__Pitch_cabinet_blade_C_temp']
297 | 
298 |         # Average Pitch Cabinet Blade Temperature and Pitch Cabinet Blade Temperatures
299 |         self.new_features['Diff_cabinet_A_avg_temp'] = self.scada_data['CS101__Pitch_cabinet_blade_A_temp'] - self.new_features['Avg_pitch_cabinet_blade_temp']
300 |         self.new_features['Diff_cabinet_B_avg_temp'] = self.scada_data['CS101__Pitch_cabinet_blade_B_temp'] - self.new_features['Avg_pitch_cabinet_blade_temp']
301 |         self.new_features['Diff_cabinet_C_avg_temp'] = self.scada_data['CS101__Pitch_cabinet_blade_C_temp'] - self.new_features['Avg_pitch_cabinet_blade_temp']
302 | 
303 |         # Rotor Temperatures
304 |         self.new_features['Diff_rotor_temps'] = self.scada_data['CS101__Rotor_temp_1'] - self.scada_data['CS101__Rotor_temp_2']
305 | 
306 |         # Average Rotor Temperature and Rotor Temperatures
307 |         self.new_features['Dif_rotor_1_avg_temps'] = self.scada_data['CS101__Rotor_temp_1'] - self.new_features['Avg_rotor_temp']
308 |         self.new_features['Dif_rotor_2_avg_temps'] = self.scada_data['CS101__Rotor_temp_2'] - self.new_features['Avg_rotor_temp']
309 | 
310 |         # Stator Temperatures
311 |         self.new_features['Diff_stator_temps'] = self.scada_data['CS101__Stator_temp_1'] - self.scada_data['CS101__Stator_temp_2']
312 | 
313 |         # Average Stator Temperature and Stator Temperatures
314 |         self.new_features['Diff_stator_1_avg_temps'] = self.scada_data['CS101__Stator_temp_1'] - self.new_features['Avg_stator_temp']
315 |         self.new_features['Diff_stator_2_avg_temps'] = self.scada_data['CS101__Stator_temp_2'] - self.new_features['Avg_stator_temp']
316 | 
317 |         # Nacelle Ambient Temperatures
318 |         self.new_features['Diff_nacelle_ambient_temps'] = self.scada_data['CS101__Nacelle_ambient_temp_1'] - self.scada_data['CS101__Nacelle_ambient_temp_2']
319 | 
320 |         # Average Nacelle Ambient Temperatures and Nacelle Temperatures
321 |         self.new_features['Diff_avg_nacelle_ambient_temp'] = self.scada_data['CS101__Nacelle_ambient_temp_1'] - self.new_features['Avg_nacelle_ambient_temp']
322 |         self.new_features['Diff_avg_nacelle_ambient_temp'] = self.scada_data['CS101__Nacelle_ambient_temp_2'] - self.new_features['Avg_nacelle_ambient_temp']
323 | 
324 |         #  Nacelle Temperature and Nacelle Cabinet Temperature
325 |         self.new_features['Diff_nacelle_cabinet_temp'] = self.scada_data['CS101__Nacelle_temp'] - self.scada_data['CS101__Nacelle_cabinet_temp']
326 | 
327 |         # Ambient Temperature and (Nacelle Temperature, Nacelle Cabinet Temperatures, Main Carrier Temperature, Rectifier Temperature, Inverter Cabinet Temperature, Tower Temperature, Control Cabinet Temperature, Transformer Temperature)
328 |         self.new_features['Diff_ambient_nacelle_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Nacelle_temp']
329 |         self.new_features['Diff_ambient_nacelle_cabinet_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Nacelle_cabinet_temp']
330 |         self.new_features['Diff_ambient_rectifier_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Rectifier_cabinet_temp']
331 |         self.new_features['Diff_ambient_main_carrier_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Main_carrier_temp']
332 |         self.new_features['Diff_ambient_yaw_inverter_cabinet_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Yaw_inverter_cabinet_temp']
333 |         self.new_features['Diff_ambient_fan_inverter_cabinet_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Fan_inverter_cabinet_temp']
334 |         self.new_features['Diff_ambient_tower_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Tower_temp']
335 |         self.new_features['Diff_ambient_control_cabinet_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Control_cabinet_temp']
336 |         self.new_features['Diff_ambient_transformer_temp'] = self.scada_data['CS101__Ambient_temp'] - self.scada_data['CS101__Transformer_temp']
337 | 
338 |         # Generator Temperature and Nacelle Temperature
339 |         self.new_features['Diff_nacelle_stator_1_temp'] = self.scada_data['CS101__Nacelle_temp'] - self.scada_data['CS101__Stator_temp_1'] 
340 |         self.new_features['Diff_nacelle_stator_2_temp'] = self.scada_data['CS101__Nacelle_temp'] - self.scada_data['CS101__Stator_temp_2'] 
341 |         self.new_features['Diff_nacelle_rotor_1_temp'] = self.scada_data['CS101__Nacelle_temp'] - self.scada_data['CS101__Rotor_temp_1'] 
342 |         self.new_features['Diff_nacelle_rotor_2_temp'] = self.scada_data['CS101__Nacelle_temp'] - self.scada_data['CS101__Rotor_temp_2'] 
343 | 
344 |     	'''
345 | 	Create the mean and standard deviation features. 
346 | 	These calculations take a long time
347 | 	'''
348 |     def create_mean_std_features(self):
349 |         """
350 |         Calculate the 2hr mean and standard deviation for self.scada_data
351 |         Saves the new features in self.mean_std
352 |         """
353 |         # Make a temporary new dataframe
354 |         self.mean_std = pd.DataFrame(index=self.scada_data.index)
355 |         temp_features = self.scada_data.copy()
356 |         temp_features[:] = np.nan
357 |         for time in self.scada_data.index: # Calculate the 2 hr average for the column
358 |             if len(self.scada_data.loc[time-pd.Timedelta('2hr'):time])== 13:
359 |                 temp_features.loc[time] = self.scada_data.loc[time-pd.Timedelta('2hr'):time].mean(axis=0)
360 |         # Rename the new columns
361 |         new_column_names = []
362 |         for name in self.scada_data.columns.values:
363 |             if len(self.scada_data.loc[time-pd.Timedelta('2hr'):time])== 13:
364 |                 new_column_names.append("2hr_mean_" + name) 
365 |         temp_features.columns = new_column_names # Rename the columns
366 |         # Join the temporary features with the new one
367 |         self.mean_std = pd.concat([temp_features, self.mean_std], axis=1)
368 | 
369 |         # Make a temporary new dataframe
370 |         temp_features = self.scada_data.copy()
371 |         temp_features[:] = np.nan
372 |         for time in self.scada_data.index: # Calculate the 2 hr average for the column
373 |             temp_features.loc[time] = self.scada_data.loc[time-pd.Timedelta('2hr'):time].std()
374 |         # Rename the new columns
375 |         new_column_names = []
376 |         for name in self.scada_data.columns.values:
377 |             new_column_names.append("2hr_std_" + name) 
378 |         temp_features.columns = new_column_names # Rename the columns
379 |         # Join the temporary features with the new one
380 |         self.mean_std = pd.concat([temp_features, self.mean_std], axis=1)
381 | 
382 | 	'''
383 | 	Include lagged variables of self.scada_data
384 | 	New features are saved in self.lagged_features
385 | 	'''
386 |     def create_lagged_features(self, n):
387 |         
388 |         self.lagged_features = pd.DataFrame(index=self.scada_data.index)
389 |         
390 |         for i in range(1, n+1):
391 |             self.temp_lagged_features = self.scada_data.copy()
392 | 
393 |             # Introduce the delay or lagged features
394 |             self.temp_lagged_features.index = self.temp_lagged_features.index +pd.Timedelta(str(i*10) + 'm')
395 | 
396 |             # Rename the new columns
397 |             new_column_names = []
398 |             for name in self.scada_data.columns.values:
399 |                 new_column_names.append(str(name) + '_t-' + str(i*10) + 'min') 
400 |             self.temp_lagged_features.columns = new_column_names # Rename the columns
401 |             
402 |             self.lagged_features = pd.concat([self.lagged_features, self.temp_lagged_features], axis=1, join='inner')
403 |     
404 | 	'''
405 | 	Creates a new pandas.Series with the same index as self.scada_data called self.ylabels
406 | 	self.ylabels = 0 except for the 5 main faults, whereby self.ylabels = Main Status
407 | 	
408 | 	Creates:
409 | 		self.ylabels
410 | 	'''    
411 |     def create_labels(self):
412 |         
413 |         fault_main_statuses = (80, 62, 228, 60, 9)
414 |         self.ylabels = pd.Series(index = self.derived_features.index)
415 |         self.ylabels[:] = 0
416 |         for i in range(0, len(self.status_data_wec)):
417 |             current_status = self.status_data_wec.iloc[i]['Main_Status'] 
418 |             if current_status in fault_main_statuses:
419 |                 start_time = self.status_data_wec.iloc[i]['Time']
420 |                 if i == (len(self.status_data_wec)-1):
421 |                     self.ylabels[self.ylabels.index > start_time] = current_status
422 |                 else:
423 |                     end_time = self.status_data_wec.iloc[i+1]['Time']
424 |                     self.ylabels[(self.ylabels.index > start_time) & (self.ylabels.index < end_time)] = current_status
425 |        
426 | 


--------------------------------------------------------------------------------
/WindTurbine_fvsff_MIQ-10_cv1-run1.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |  "cells": [
   3 |   {
   4 |    "cell_type": "code",
   5 |    "execution_count": 1,
   6 |    "metadata": {
   7 |     "collapsed": true
   8 |    },
   9 |    "outputs": [],
  10 |    "source": [
  11 |     "import pandas as pd\n",
  12 |     "import numpy as np\n",
  13 |     "import matplotlib.pyplot as plt\n",
  14 |     "% matplotlib inline\n",
  15 |     "import os\n",
  16 |     "\n",
  17 |     "from sklearn import cross_validation as cval\n",
  18 |     "#from sklearn import utils\n",
  19 |     "from sklearn.grid_search import GridSearchCV, RandomizedSearchCV\n",
  20 |     "from sklearn.metrics import classification_report, confusion_matrix\n",
  21 |     "from sklearn.svm import SVC\n",
  22 |     "import pickle"
  23 |    ]
  24 |   },
  25 |   {
  26 |    "cell_type": "code",
  27 |    "execution_count": 2,
  28 |    "metadata": {
  29 |     "collapsed": true
  30 |    },
  31 |    "outputs": [],
  32 |    "source": [
  33 |     "# Manipulated Variables\n",
  34 |     "num_features = 10\n",
  35 |     "selection_method = 'MIQ'\n",
  36 |     "CV_test_number = 1\n",
  37 |     "run_num = 1\n",
  38 |     "output_filename = 'WindTurbine_fvsff_MIQ-' + str(num_features) + '_cv' + str(CV_test_number) + '-run' + str(run_num)"
  39 |    ]
  40 |   },
  41 |   {
  42 |    "cell_type": "code",
  43 |    "execution_count": 3,
  44 |    "metadata": {
  45 |     "collapsed": false
  46 |    },
  47 |    "outputs": [],
  48 |    "source": [
  49 |     "# Import training data\n",
  50 |     "train_filename = '../Data/PickleFiles/3_labels_cv/training_data_cv' + str(CV_test_number) + '.p'\n",
  51 |     "x_train = pd.read_pickle(train_filename)\n",
  52 |     "\n",
  53 |     "train_filename = '../Data//PickleFiles/3_labels_cv/training_data_cv' + str(CV_test_number) + 'labelweights.p'\n",
  54 |     "labelsweights = pd.read_pickle(train_filename)\n",
  55 |     "x_train['__weight__'] = labelsweights['__weight__']\n",
  56 |     "x_train['__label__'] = labelsweights['__label__']\n",
  57 |     "\n",
  58 |     "# Shuffle the data\n",
  59 |     "x_train = x_train.sample(frac=1, random_state=324351369)\n",
  60 |     "\n",
  61 |     "# Remove out the labels, weights, and x_train data\n",
  62 |     "y_train = x_train['__label__'].copy()\n",
  63 |     "x_train_sample_weights = x_train['__weight__'].copy()\n",
  64 |     "x_train.drop(['__label__', '__weight__'], 1, inplace=True)\n",
  65 |     "\n",
  66 |     "# Import testing data\n",
  67 |     "test_filename = '../Data/PickleFiles/3_labels_cv/testing_data_cv' + str(CV_test_number) +  '.p'\n",
  68 |     "x_test = pd.read_pickle(test_filename)\n",
  69 |     "\n",
  70 |     "test_filename = '../Data/PickleFiles/3_labels_cv/testing_data_cv' + str(CV_test_number) + 'labelweights.p'\n",
  71 |     "labelsweights = pd.read_pickle(test_filename)\n",
  72 |     "y_test = labelsweights['__label__'].copy()\n",
  73 |     "\n",
  74 |     "del labelsweights\n",
  75 |     "\n",
  76 |     "# Import the feature rankings\n",
  77 |     "feature_filepath = '../Code-WindTurbine2/windturbine_fvsff_' + 'MIQ-first100.txt'\n",
  78 |     "features = pd.read_csv(feature_filepath,  delimiter=' \t ', header=2, engine='python')['Name']"
  79 |    ]
  80 |   },
  81 |   {
  82 |    "cell_type": "code",
  83 |    "execution_count": 4,
  84 |    "metadata": {
  85 |     "collapsed": false
  86 |    },
  87 |    "outputs": [
  88 |     {
  89 |      "name": "stdout",
  90 |      "output_type": "stream",
  91 |      "text": [
  92 |       "train_filename is: \n",
  93 |       "../Data//PickleFiles/3_labels_cv/training_data_cv1labelweights.p\n",
  94 |       "\n",
  95 |       "test_filename is: \n",
  96 |       "../Data/PickleFiles/3_labels_cv/testing_data_cv1labelweights.p\n",
  97 |       "\n",
  98 |       "output_filename is: \n",
  99 |       "WindTurbine_fvsff_MIQ-10_cv1-run1\n",
 100 |       "\n"
 101 |      ]
 102 |     }
 103 |    ],
 104 |    "source": [
 105 |     "print(\"train_filename is: \\n%s\\n\" % train_filename)\n",
 106 |     "print(\"test_filename is: \\n%s\\n\" % test_filename)\n",
 107 |     "print(\"output_filename is: \\n%s\\n\" % output_filename)"
 108 |    ]
 109 |   },
 110 |   {
 111 |    "cell_type": "code",
 112 |    "execution_count": 7,
 113 |    "metadata": {
 114 |     "collapsed": false
 115 |    },
 116 |    "outputs": [],
 117 |    "source": [
 118 |     "def CVTrainAndScore(X_train, y_train, X_test, y_test, weights, output_filename, search_type=RandomizedSearchCV,\n",
 119 |     "    tuned_parameters={\n",
 120 |     "        'kernel': ['linear', 'poly', 'rbf', 'sigmoid'], 'gamma': ['auto', 1e-3, 1e-4],\n",
 121 |     "        'C': [0.01, .1, 1, 10,  100, 1000]},\n",
 122 |     "        #'class_weight':['balanced'],\n",
 123 |     "        scores=['f1']):\n",
 124 |     "    \n",
 125 |     "    for score in scores:\n",
 126 |     "        print(\"# Tuning hyper-parameters for %s \\n\" % score)\n",
 127 |     "\n",
 128 |     "        # Find the hyperparameters\n",
 129 |     "        clf = search_type(SVC(), tuned_parameters, fit_params={\"sample_weight\":weights}, cv=10, scoring='%s_weighted' % score, verbose=200)\n",
 130 |     "        \n",
 131 |     "        # Train the SVM\n",
 132 |     "        clf.fit(X_train, y_train)\n",
 133 |     "\n",
 134 |     "        print(\"\\nBest parameters set found on development set: \\n\")\n",
 135 |     "        print(clf.best_params_)\n",
 136 |     "\n",
 137 |     "        print(\"\\nGrid scores on development set:\\n\")\n",
 138 |     "        for params, mean_score, scores in clf.grid_scores_:\n",
 139 |     "            print(\"%0.3f (+/-%0.03f) for %r\"\n",
 140 |     "                  % (mean_score, scores.std() * 2, params))\n",
 141 |     "\n",
 142 |     "        print(\"\\nDetailed classification report:\\n\")\n",
 143 |     "\n",
 144 |     "\n",
 145 |     "        # Make the predictions on training set\n",
 146 |     "        y_true, y_pred = y_train, clf.predict(X_train)\n",
 147 |     "        print('Report on Training Set')\n",
 148 |     "        print(classification_report(y_true, y_pred))\n",
 149 |     "        print()\n",
 150 |     "        # Evaluate the SVM using Confusion Matrix\n",
 151 |     "        cm = confusion_matrix(y_true, y_pred)\n",
 152 |     "        print(cm)\n",
 153 |     "        cm_normalized = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
 154 |     "        # Also print specificity metric\n",
 155 |     "        print(\"Specificity:\", cm[0, 0] / (cm[0, 1] + cm[0, 0]))\n",
 156 |     "\n",
 157 |     "        # Make the predictions on the test set\n",
 158 |     "        y_true, y_pred = y_test, clf.predict(X_test)\n",
 159 |     "        print('Report on Test Set\\n')\n",
 160 |     "        print(classification_report(y_true, y_pred))\n",
 161 |     "        print()\n",
 162 |     "        # Evaluate the SVM using Confusion Matrix\n",
 163 |     "        cm = confusion_matrix(y_true, y_pred)\n",
 164 |     "        print(cm)\n",
 165 |     "        cm_normalized = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
 166 |     "        # Also print specificity metric\n",
 167 |     "        print(\"Specificity:\", cm[0, 0] / (cm[0, 1] + cm[0, 0]))\n",
 168 |     "        \n",
 169 |     "        # plot the confusion matrices\n",
 170 |     "        plot = plot_confusion_matrix(cm_normalized)\n",
 171 |     "        \n",
 172 |     "        print(clf.best_params_)\n",
 173 |     "        f=open(output_filename+score+'.p', 'wb')\n",
 174 |     "        pickle.dump(clf, f)\n",
 175 |     "        f.close()\n",
 176 |     "        \n",
 177 |     "        \n",
 178 |     "def plot_confusion_matrix(cm):\n",
 179 |     "    title='Confusion matrix',\n",
 180 |     "    cmap=plt.cm.Blues\n",
 181 |     "    labels=['No-Fault', 'Fault']\n",
 182 |     "    \n",
 183 |     "    plot = plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
 184 |     "    plt.title(title)\n",
 185 |     "    plt.colorbar()\n",
 186 |     "    tick_marks = np.arange(len(labels))\n",
 187 |     "    plt.xticks(tick_marks, labels, rotation=45)\n",
 188 |     "    plt.yticks(tick_marks, labels)\n",
 189 |     "    # plt.tight_layout()\n",
 190 |     "    plt.ylabel('True label')\n",
 191 |     "    plt.xlabel('Predicted label')\n",
 192 |     "    plt.show()"
 193 |    ]
 194 |   },
 195 |   {
 196 |    "cell_type": "code",
 197 |    "execution_count": 8,
 198 |    "metadata": {
 199 |     "collapsed": false
 200 |    },
 201 |    "outputs": [
 202 |     {
 203 |      "name": "stdout",
 204 |      "output_type": "stream",
 205 |      "text": [
 206 |       "# Tuning hyper-parameters for f1 \n",
 207 |       "\n",
 208 |       "Fitting 10 folds for each of 10 candidates, totalling 100 fits\n",
 209 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 210 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=0.996327 -   0.0s\n",
 211 |       "[Parallel(n_jobs=1)]: Done   1 tasks       | elapsed:    0.0s\n",
 212 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 213 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=0.997961 -   0.0s\n",
 214 |       "[Parallel(n_jobs=1)]: Done   2 tasks       | elapsed:    0.1s\n",
 215 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 216 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=0.994839 -   0.0s\n",
 217 |       "[Parallel(n_jobs=1)]: Done   3 tasks       | elapsed:    0.1s\n",
 218 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 219 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=0.997956 -   0.0s\n",
 220 |       "[Parallel(n_jobs=1)]: Done   4 tasks       | elapsed:    0.2s\n",
 221 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n"
 222 |      ]
 223 |     },
 224 |     {
 225 |      "name": "stderr",
 226 |      "output_type": "stream",
 227 |      "text": [
 228 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/cross_validation.py:516: Warning: The least populated class in y has only 6 members, which is too few. The minimum number of labels for any class cannot be less than n_folds=10.\n",
 229 |       "  % (min_labels, self.n_folds)), Warning)\n"
 230 |      ]
 231 |     },
 232 |     {
 233 |      "name": "stdout",
 234 |      "output_type": "stream",
 235 |      "text": [
 236 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=0.997956 -   0.0s\n",
 237 |       "[Parallel(n_jobs=1)]: Done   5 tasks       | elapsed:    0.2s\n",
 238 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 239 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=1.000000 -   0.0s\n",
 240 |       "[Parallel(n_jobs=1)]: Done   6 tasks       | elapsed:    0.2s\n",
 241 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 242 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=0.996301 -   0.0s\n",
 243 |       "[Parallel(n_jobs=1)]: Done   7 tasks       | elapsed:    0.3s\n",
 244 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 245 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=0.996301 -   0.0s\n",
 246 |       "[Parallel(n_jobs=1)]: Done   8 tasks       | elapsed:    0.3s\n",
 247 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 248 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=1.000000 -   0.0s\n",
 249 |       "[Parallel(n_jobs=1)]: Done   9 tasks       | elapsed:    0.4s\n",
 250 |       "[CV] gamma=auto, kernel=rbf, C=10 ....................................\n",
 251 |       "[CV] ........... gamma=auto, kernel=rbf, C=10, score=0.993820 -   0.0s\n",
 252 |       "[Parallel(n_jobs=1)]: Done  10 tasks       | elapsed:    0.4s\n",
 253 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 254 |      ]
 255 |     },
 256 |     {
 257 |      "name": "stderr",
 258 |      "output_type": "stream",
 259 |      "text": [
 260 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 261 |       "  'recall', 'true', average, warn_for)\n",
 262 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 263 |       "  'recall', 'true', average, warn_for)\n",
 264 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 265 |       "  'recall', 'true', average, warn_for)\n",
 266 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 267 |       "  'precision', 'predicted', average, warn_for)\n"
 268 |      ]
 269 |     },
 270 |     {
 271 |      "name": "stdout",
 272 |      "output_type": "stream",
 273 |      "text": [
 274 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.996334 -   0.5s\n",
 275 |       "[Parallel(n_jobs=1)]: Done  11 tasks       | elapsed:    0.9s\n",
 276 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 277 |      ]
 278 |     },
 279 |     {
 280 |      "name": "stderr",
 281 |      "output_type": "stream",
 282 |      "text": [
 283 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 284 |       "  'precision', 'predicted', average, warn_for)\n"
 285 |      ]
 286 |     },
 287 |     {
 288 |      "name": "stdout",
 289 |      "output_type": "stream",
 290 |      "text": [
 291 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.996334 -   0.5s\n",
 292 |       "[Parallel(n_jobs=1)]: Done  12 tasks       | elapsed:    1.5s\n",
 293 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 294 |      ]
 295 |     },
 296 |     {
 297 |      "name": "stderr",
 298 |      "output_type": "stream",
 299 |      "text": [
 300 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 301 |       "  'precision', 'predicted', average, warn_for)\n"
 302 |      ]
 303 |     },
 304 |     {
 305 |      "name": "stdout",
 306 |      "output_type": "stream",
 307 |      "text": [
 308 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.996325 -   0.5s\n",
 309 |       "[Parallel(n_jobs=1)]: Done  13 tasks       | elapsed:    2.0s\n",
 310 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 311 |      ]
 312 |     },
 313 |     {
 314 |      "name": "stderr",
 315 |      "output_type": "stream",
 316 |      "text": [
 317 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 318 |       "  'precision', 'predicted', average, warn_for)\n"
 319 |      ]
 320 |     },
 321 |     {
 322 |      "name": "stdout",
 323 |      "output_type": "stream",
 324 |      "text": [
 325 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.996325 -   0.5s\n",
 326 |       "[Parallel(n_jobs=1)]: Done  14 tasks       | elapsed:    2.6s\n",
 327 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 328 |      ]
 329 |     },
 330 |     {
 331 |      "name": "stderr",
 332 |      "output_type": "stream",
 333 |      "text": [
 334 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 335 |       "  'precision', 'predicted', average, warn_for)\n"
 336 |      ]
 337 |     },
 338 |     {
 339 |      "name": "stdout",
 340 |      "output_type": "stream",
 341 |      "text": [
 342 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.996325 -   0.5s\n",
 343 |       "[Parallel(n_jobs=1)]: Done  15 tasks       | elapsed:    3.1s\n",
 344 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 345 |      ]
 346 |     },
 347 |     {
 348 |      "name": "stderr",
 349 |      "output_type": "stream",
 350 |      "text": [
 351 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 352 |       "  'precision', 'predicted', average, warn_for)\n"
 353 |      ]
 354 |     },
 355 |     {
 356 |      "name": "stdout",
 357 |      "output_type": "stream",
 358 |      "text": [
 359 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.996325 -   0.5s\n",
 360 |       "[Parallel(n_jobs=1)]: Done  16 tasks       | elapsed:    3.6s\n",
 361 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 362 |      ]
 363 |     },
 364 |     {
 365 |      "name": "stderr",
 366 |      "output_type": "stream",
 367 |      "text": [
 368 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 369 |       "  'precision', 'predicted', average, warn_for)\n",
 370 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 371 |       "  'recall', 'true', average, warn_for)\n"
 372 |      ]
 373 |     },
 374 |     {
 375 |      "name": "stdout",
 376 |      "output_type": "stream",
 377 |      "text": [
 378 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.000000 -   0.6s\n",
 379 |       "[Parallel(n_jobs=1)]: Done  17 tasks       | elapsed:    4.2s\n",
 380 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 381 |      ]
 382 |     },
 383 |     {
 384 |      "name": "stderr",
 385 |      "output_type": "stream",
 386 |      "text": [
 387 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 388 |       "  'precision', 'predicted', average, warn_for)\n",
 389 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 390 |       "  'recall', 'true', average, warn_for)\n"
 391 |      ]
 392 |     },
 393 |     {
 394 |      "name": "stdout",
 395 |      "output_type": "stream",
 396 |      "text": [
 397 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.000000 -   0.6s\n",
 398 |       "[Parallel(n_jobs=1)]: Done  18 tasks       | elapsed:    4.7s\n",
 399 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 400 |      ]
 401 |     },
 402 |     {
 403 |      "name": "stderr",
 404 |      "output_type": "stream",
 405 |      "text": [
 406 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 407 |       "  'precision', 'predicted', average, warn_for)\n",
 408 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 409 |       "  'recall', 'true', average, warn_for)\n"
 410 |      ]
 411 |     },
 412 |     {
 413 |      "name": "stdout",
 414 |      "output_type": "stream",
 415 |      "text": [
 416 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.000000 -   0.6s\n",
 417 |       "[Parallel(n_jobs=1)]: Done  19 tasks       | elapsed:    5.3s\n",
 418 |       "[CV] gamma=auto, kernel=sigmoid, C=1 .................................\n"
 419 |      ]
 420 |     },
 421 |     {
 422 |      "name": "stderr",
 423 |      "output_type": "stream",
 424 |      "text": [
 425 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 426 |       "  'precision', 'predicted', average, warn_for)\n",
 427 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 428 |       "  'recall', 'true', average, warn_for)\n"
 429 |      ]
 430 |     },
 431 |     {
 432 |      "name": "stdout",
 433 |      "output_type": "stream",
 434 |      "text": [
 435 |       "[CV] ........ gamma=auto, kernel=sigmoid, C=1, score=0.000000 -   0.6s\n",
 436 |       "[Parallel(n_jobs=1)]: Done  20 tasks       | elapsed:    5.8s\n",
 437 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n",
 438 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.520482 -   1.0s\n",
 439 |       "[Parallel(n_jobs=1)]: Done  21 tasks       | elapsed:    6.9s\n",
 440 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n",
 441 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.520482 -   1.0s\n",
 442 |       "[Parallel(n_jobs=1)]: Done  22 tasks       | elapsed:    7.9s\n",
 443 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n",
 444 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.529404 -   1.0s\n",
 445 |       "[Parallel(n_jobs=1)]: Done  23 tasks       | elapsed:    8.9s\n",
 446 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n",
 447 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.474507 -   1.0s\n",
 448 |       "[Parallel(n_jobs=1)]: Done  24 tasks       | elapsed:    9.9s\n",
 449 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n",
 450 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.507928 -   1.0s\n",
 451 |       "[Parallel(n_jobs=1)]: Done  25 tasks       | elapsed:   10.9s\n",
 452 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n",
 453 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.499711 -   1.0s\n",
 454 |       "[Parallel(n_jobs=1)]: Done  26 tasks       | elapsed:   12.0s\n",
 455 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n"
 456 |      ]
 457 |     },
 458 |     {
 459 |      "name": "stderr",
 460 |      "output_type": "stream",
 461 |      "text": [
 462 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 463 |       "  'recall', 'true', average, warn_for)\n"
 464 |      ]
 465 |     },
 466 |     {
 467 |      "name": "stdout",
 468 |      "output_type": "stream",
 469 |      "text": [
 470 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.511883 -   1.0s\n",
 471 |       "[Parallel(n_jobs=1)]: Done  27 tasks       | elapsed:   13.0s\n",
 472 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n"
 473 |      ]
 474 |     },
 475 |     {
 476 |      "name": "stderr",
 477 |      "output_type": "stream",
 478 |      "text": [
 479 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 480 |       "  'recall', 'true', average, warn_for)\n"
 481 |      ]
 482 |     },
 483 |     {
 484 |      "name": "stdout",
 485 |      "output_type": "stream",
 486 |      "text": [
 487 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.492593 -   0.9s\n",
 488 |       "[Parallel(n_jobs=1)]: Done  28 tasks       | elapsed:   13.9s\n",
 489 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n"
 490 |      ]
 491 |     },
 492 |     {
 493 |      "name": "stderr",
 494 |      "output_type": "stream",
 495 |      "text": [
 496 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 497 |       "  'recall', 'true', average, warn_for)\n"
 498 |      ]
 499 |     },
 500 |     {
 501 |      "name": "stdout",
 502 |      "output_type": "stream",
 503 |      "text": [
 504 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.530686 -   1.0s\n",
 505 |       "[Parallel(n_jobs=1)]: Done  29 tasks       | elapsed:   14.9s\n",
 506 |       "[CV] gamma=0.001, kernel=rbf, C=0.1 ..................................\n"
 507 |      ]
 508 |     },
 509 |     {
 510 |      "name": "stderr",
 511 |      "output_type": "stream",
 512 |      "text": [
 513 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 514 |       "  'recall', 'true', average, warn_for)\n"
 515 |      ]
 516 |     },
 517 |     {
 518 |      "name": "stdout",
 519 |      "output_type": "stream",
 520 |      "text": [
 521 |       "[CV] ......... gamma=0.001, kernel=rbf, C=0.1, score=0.535971 -   1.0s\n",
 522 |       "[Parallel(n_jobs=1)]: Done  30 tasks       | elapsed:   15.9s\n",
 523 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n",
 524 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.994852 -   0.2s\n",
 525 |       "[Parallel(n_jobs=1)]: Done  31 tasks       | elapsed:   16.0s\n",
 526 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n",
 527 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.993456 -   0.1s\n",
 528 |       "[Parallel(n_jobs=1)]: Done  32 tasks       | elapsed:   16.2s\n",
 529 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n",
 530 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.996317 -   0.1s\n",
 531 |       "[Parallel(n_jobs=1)]: Done  33 tasks       | elapsed:   16.3s\n",
 532 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n",
 533 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.993440 -   0.1s\n",
 534 |       "[Parallel(n_jobs=1)]: Done  34 tasks       | elapsed:   16.5s\n",
 535 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n",
 536 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.996317 -   0.1s\n",
 537 |       "[Parallel(n_jobs=1)]: Done  35 tasks       | elapsed:   16.6s\n",
 538 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n",
 539 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.996317 -   0.1s\n",
 540 |       "[Parallel(n_jobs=1)]: Done  36 tasks       | elapsed:   16.7s\n",
 541 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n",
 542 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.996301 -   0.2s\n",
 543 |       "[Parallel(n_jobs=1)]: Done  37 tasks       | elapsed:   16.9s\n",
 544 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n"
 545 |      ]
 546 |     },
 547 |     {
 548 |      "name": "stderr",
 549 |      "output_type": "stream",
 550 |      "text": [
 551 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 552 |       "  'recall', 'true', average, warn_for)\n",
 553 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 554 |       "  'recall', 'true', average, warn_for)\n"
 555 |      ]
 556 |     },
 557 |     {
 558 |      "name": "stdout",
 559 |      "output_type": "stream",
 560 |      "text": [
 561 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.992574 -   0.2s\n",
 562 |       "[Parallel(n_jobs=1)]: Done  38 tasks       | elapsed:   17.1s\n",
 563 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n",
 564 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.995062 -   0.1s\n",
 565 |       "[Parallel(n_jobs=1)]: Done  39 tasks       | elapsed:   17.3s\n",
 566 |       "[CV] gamma=0.001, kernel=linear, C=1000 ..............................\n"
 567 |      ]
 568 |     },
 569 |     {
 570 |      "name": "stderr",
 571 |      "output_type": "stream",
 572 |      "text": [
 573 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 574 |       "  'recall', 'true', average, warn_for)\n",
 575 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 576 |       "  'recall', 'true', average, warn_for)\n"
 577 |      ]
 578 |     },
 579 |     {
 580 |      "name": "stdout",
 581 |      "output_type": "stream",
 582 |      "text": [
 583 |       "[CV] ..... gamma=0.001, kernel=linear, C=1000, score=0.992574 -   0.1s\n",
 584 |       "[Parallel(n_jobs=1)]: Done  40 tasks       | elapsed:   17.4s\n",
 585 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n",
 586 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.980447 -   0.1s\n",
 587 |       "[Parallel(n_jobs=1)]: Done  41 tasks       | elapsed:   17.5s\n",
 588 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n",
 589 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.985586 -   0.1s\n",
 590 |       "[Parallel(n_jobs=1)]: Done  42 tasks       | elapsed:   17.6s\n",
 591 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n",
 592 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.984261 -   0.1s\n",
 593 |       "[Parallel(n_jobs=1)]: Done  43 tasks       | elapsed:   17.8s\n",
 594 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n",
 595 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.985550 -   0.1s\n",
 596 |       "[Parallel(n_jobs=1)]: Done  44 tasks       | elapsed:   17.9s\n",
 597 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n",
 598 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.982973 -   0.1s\n",
 599 |       "[Parallel(n_jobs=1)]: Done  45 tasks       | elapsed:   18.0s\n",
 600 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n",
 601 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.980398 -   0.1s\n",
 602 |       "[Parallel(n_jobs=1)]: Done  46 tasks       | elapsed:   18.1s\n",
 603 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n",
 604 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.985037 -   0.1s\n",
 605 |       "[Parallel(n_jobs=1)]: Done  47 tasks       | elapsed:   18.3s\n",
 606 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n"
 607 |      ]
 608 |     },
 609 |     {
 610 |      "name": "stderr",
 611 |      "output_type": "stream",
 612 |      "text": [
 613 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 614 |       "  'recall', 'true', average, warn_for)\n",
 615 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 616 |       "  'recall', 'true', average, warn_for)\n"
 617 |      ]
 618 |     },
 619 |     {
 620 |      "name": "stdout",
 621 |      "output_type": "stream",
 622 |      "text": [
 623 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.973518 -   0.1s\n",
 624 |       "[Parallel(n_jobs=1)]: Done  48 tasks       | elapsed:   18.4s\n",
 625 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n",
 626 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.985037 -   0.1s\n",
 627 |       "[Parallel(n_jobs=1)]: Done  49 tasks       | elapsed:   18.5s\n",
 628 |       "[CV] gamma=0.001, kernel=sigmoid, C=100 ..............................\n"
 629 |      ]
 630 |     },
 631 |     {
 632 |      "name": "stderr",
 633 |      "output_type": "stream",
 634 |      "text": [
 635 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 636 |       "  'recall', 'true', average, warn_for)\n",
 637 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 638 |       "  'recall', 'true', average, warn_for)\n"
 639 |      ]
 640 |     },
 641 |     {
 642 |      "name": "stdout",
 643 |      "output_type": "stream",
 644 |      "text": [
 645 |       "[CV] ..... gamma=0.001, kernel=sigmoid, C=100, score=0.976101 -   0.1s\n",
 646 |       "[Parallel(n_jobs=1)]: Done  50 tasks       | elapsed:   18.6s\n",
 647 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 648 |      ]
 649 |     },
 650 |     {
 651 |      "name": "stderr",
 652 |      "output_type": "stream",
 653 |      "text": [
 654 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 655 |       "  'precision', 'predicted', average, warn_for)\n"
 656 |      ]
 657 |     },
 658 |     {
 659 |      "name": "stdout",
 660 |      "output_type": "stream",
 661 |      "text": [
 662 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.996334 -   1.0s\n",
 663 |       "[Parallel(n_jobs=1)]: Done  51 tasks       | elapsed:   19.7s\n",
 664 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 665 |      ]
 666 |     },
 667 |     {
 668 |      "name": "stderr",
 669 |      "output_type": "stream",
 670 |      "text": [
 671 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 672 |       "  'precision', 'predicted', average, warn_for)\n"
 673 |      ]
 674 |     },
 675 |     {
 676 |      "name": "stdout",
 677 |      "output_type": "stream",
 678 |      "text": [
 679 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.996334 -   1.0s\n",
 680 |       "[Parallel(n_jobs=1)]: Done  52 tasks       | elapsed:   20.7s\n",
 681 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 682 |      ]
 683 |     },
 684 |     {
 685 |      "name": "stderr",
 686 |      "output_type": "stream",
 687 |      "text": [
 688 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 689 |       "  'precision', 'predicted', average, warn_for)\n"
 690 |      ]
 691 |     },
 692 |     {
 693 |      "name": "stdout",
 694 |      "output_type": "stream",
 695 |      "text": [
 696 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.996325 -   1.0s\n",
 697 |       "[Parallel(n_jobs=1)]: Done  53 tasks       | elapsed:   21.8s\n",
 698 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 699 |      ]
 700 |     },
 701 |     {
 702 |      "name": "stderr",
 703 |      "output_type": "stream",
 704 |      "text": [
 705 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 706 |       "  'precision', 'predicted', average, warn_for)\n"
 707 |      ]
 708 |     },
 709 |     {
 710 |      "name": "stdout",
 711 |      "output_type": "stream",
 712 |      "text": [
 713 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.996325 -   1.0s\n",
 714 |       "[Parallel(n_jobs=1)]: Done  54 tasks       | elapsed:   22.8s\n",
 715 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 716 |      ]
 717 |     },
 718 |     {
 719 |      "name": "stderr",
 720 |      "output_type": "stream",
 721 |      "text": [
 722 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 723 |       "  'precision', 'predicted', average, warn_for)\n"
 724 |      ]
 725 |     },
 726 |     {
 727 |      "name": "stdout",
 728 |      "output_type": "stream",
 729 |      "text": [
 730 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.996325 -   1.0s\n",
 731 |       "[Parallel(n_jobs=1)]: Done  55 tasks       | elapsed:   23.8s\n",
 732 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 733 |      ]
 734 |     },
 735 |     {
 736 |      "name": "stderr",
 737 |      "output_type": "stream",
 738 |      "text": [
 739 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 740 |       "  'precision', 'predicted', average, warn_for)\n"
 741 |      ]
 742 |     },
 743 |     {
 744 |      "name": "stdout",
 745 |      "output_type": "stream",
 746 |      "text": [
 747 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.996325 -   1.0s\n",
 748 |       "[Parallel(n_jobs=1)]: Done  56 tasks       | elapsed:   24.9s\n",
 749 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 750 |      ]
 751 |     },
 752 |     {
 753 |      "name": "stderr",
 754 |      "output_type": "stream",
 755 |      "text": [
 756 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 757 |       "  'precision', 'predicted', average, warn_for)\n",
 758 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 759 |       "  'recall', 'true', average, warn_for)\n"
 760 |      ]
 761 |     },
 762 |     {
 763 |      "name": "stdout",
 764 |      "output_type": "stream",
 765 |      "text": [
 766 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.000000 -   1.1s\n",
 767 |       "[Parallel(n_jobs=1)]: Done  57 tasks       | elapsed:   26.0s\n",
 768 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 769 |      ]
 770 |     },
 771 |     {
 772 |      "name": "stderr",
 773 |      "output_type": "stream",
 774 |      "text": [
 775 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 776 |       "  'precision', 'predicted', average, warn_for)\n",
 777 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 778 |       "  'recall', 'true', average, warn_for)\n"
 779 |      ]
 780 |     },
 781 |     {
 782 |      "name": "stdout",
 783 |      "output_type": "stream",
 784 |      "text": [
 785 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.000000 -   1.1s\n",
 786 |       "[Parallel(n_jobs=1)]: Done  58 tasks       | elapsed:   27.1s\n",
 787 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 788 |      ]
 789 |     },
 790 |     {
 791 |      "name": "stderr",
 792 |      "output_type": "stream",
 793 |      "text": [
 794 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 795 |       "  'precision', 'predicted', average, warn_for)\n",
 796 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 797 |       "  'recall', 'true', average, warn_for)\n"
 798 |      ]
 799 |     },
 800 |     {
 801 |      "name": "stdout",
 802 |      "output_type": "stream",
 803 |      "text": [
 804 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.000000 -   1.1s\n",
 805 |       "[Parallel(n_jobs=1)]: Done  59 tasks       | elapsed:   28.2s\n",
 806 |       "[CV] gamma=0.0001, kernel=rbf, C=0.1 .................................\n"
 807 |      ]
 808 |     },
 809 |     {
 810 |      "name": "stderr",
 811 |      "output_type": "stream",
 812 |      "text": [
 813 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
 814 |       "  'precision', 'predicted', average, warn_for)\n",
 815 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 816 |       "  'recall', 'true', average, warn_for)\n"
 817 |      ]
 818 |     },
 819 |     {
 820 |      "name": "stdout",
 821 |      "output_type": "stream",
 822 |      "text": [
 823 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=0.1, score=0.000000 -   1.1s\n",
 824 |       "[Parallel(n_jobs=1)]: Done  60 tasks       | elapsed:   29.3s\n",
 825 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 826 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.992103 -   0.0s\n",
 827 |       "[Parallel(n_jobs=1)]: Done  61 tasks       | elapsed:   29.4s\n",
 828 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 829 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.993456 -   0.0s\n",
 830 |       "[Parallel(n_jobs=1)]: Done  62 tasks       | elapsed:   29.4s\n",
 831 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 832 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.996317 -   0.0s\n",
 833 |       "[Parallel(n_jobs=1)]: Done  63 tasks       | elapsed:   29.5s\n",
 834 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 835 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.990754 -   0.0s\n",
 836 |       "[Parallel(n_jobs=1)]: Done  64 tasks       | elapsed:   29.5s\n",
 837 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 838 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.993440 -   0.0s\n",
 839 |       "[Parallel(n_jobs=1)]: Done  65 tasks       | elapsed:   29.6s\n",
 840 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 841 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.994839 -   0.0s\n",
 842 |       "[Parallel(n_jobs=1)]: Done  66 tasks       | elapsed:   29.6s\n",
 843 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 844 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.996301 -   0.0s\n",
 845 |       "[Parallel(n_jobs=1)]: Done  67 tasks       | elapsed:   29.7s\n",
 846 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n"
 847 |      ]
 848 |     },
 849 |     {
 850 |      "name": "stderr",
 851 |      "output_type": "stream",
 852 |      "text": [
 853 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 854 |       "  'recall', 'true', average, warn_for)\n",
 855 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 856 |       "  'recall', 'true', average, warn_for)\n",
 857 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 858 |       "  'recall', 'true', average, warn_for)\n",
 859 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 860 |       "  'recall', 'true', average, warn_for)\n"
 861 |      ]
 862 |     },
 863 |     {
 864 |      "name": "stdout",
 865 |      "output_type": "stream",
 866 |      "text": [
 867 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.991326 -   0.0s\n",
 868 |       "[Parallel(n_jobs=1)]: Done  68 tasks       | elapsed:   29.7s\n",
 869 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 870 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.995062 -   0.0s\n",
 871 |       "[Parallel(n_jobs=1)]: Done  69 tasks       | elapsed:   29.8s\n",
 872 |       "[CV] gamma=0.001, kernel=linear, C=10 ................................\n",
 873 |       "[CV] ....... gamma=0.001, kernel=linear, C=10, score=0.992574 -   0.0s\n",
 874 |       "[Parallel(n_jobs=1)]: Done  70 tasks       | elapsed:   29.8s\n",
 875 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n",
 876 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.980447 -   0.1s\n",
 877 |       "[Parallel(n_jobs=1)]: Done  71 tasks       | elapsed:   29.9s\n",
 878 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n",
 879 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.988167 -   0.1s\n",
 880 |       "[Parallel(n_jobs=1)]: Done  72 tasks       | elapsed:   30.0s\n",
 881 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n",
 882 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.988138 -   0.1s\n",
 883 |       "[Parallel(n_jobs=1)]: Done  73 tasks       | elapsed:   30.1s\n",
 884 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n",
 885 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.985550 -   0.1s\n",
 886 |       "[Parallel(n_jobs=1)]: Done  74 tasks       | elapsed:   30.2s\n",
 887 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n",
 888 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.984261 -   0.1s\n",
 889 |       "[Parallel(n_jobs=1)]: Done  75 tasks       | elapsed:   30.3s\n",
 890 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n",
 891 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.984261 -   0.1s\n",
 892 |       "[Parallel(n_jobs=1)]: Done  76 tasks       | elapsed:   30.4s\n",
 893 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n"
 894 |      ]
 895 |     },
 896 |     {
 897 |      "name": "stderr",
 898 |      "output_type": "stream",
 899 |      "text": [
 900 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 901 |       "  'recall', 'true', average, warn_for)\n",
 902 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 903 |       "  'recall', 'true', average, warn_for)\n",
 904 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 905 |       "  'recall', 'true', average, warn_for)\n"
 906 |      ]
 907 |     },
 908 |     {
 909 |      "name": "stdout",
 910 |      "output_type": "stream",
 911 |      "text": [
 912 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.985037 -   0.1s\n",
 913 |       "[Parallel(n_jobs=1)]: Done  77 tasks       | elapsed:   30.5s\n",
 914 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n",
 915 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.978670 -   0.1s\n",
 916 |       "[Parallel(n_jobs=1)]: Done  78 tasks       | elapsed:   30.6s\n",
 917 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n",
 918 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.991326 -   0.1s\n",
 919 |       "[Parallel(n_jobs=1)]: Done  79 tasks       | elapsed:   30.7s\n",
 920 |       "[CV] gamma=0.001, kernel=rbf, C=100 ..................................\n"
 921 |      ]
 922 |     },
 923 |     {
 924 |      "name": "stderr",
 925 |      "output_type": "stream",
 926 |      "text": [
 927 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 928 |       "  'recall', 'true', average, warn_for)\n"
 929 |      ]
 930 |     },
 931 |     {
 932 |      "name": "stdout",
 933 |      "output_type": "stream",
 934 |      "text": [
 935 |       "[CV] ......... gamma=0.001, kernel=rbf, C=100, score=0.985037 -   0.1s\n",
 936 |       "[Parallel(n_jobs=1)]: Done  80 tasks       | elapsed:   30.8s\n",
 937 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n",
 938 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.869290 -   0.7s\n",
 939 |       "[Parallel(n_jobs=1)]: Done  81 tasks       | elapsed:   31.5s\n",
 940 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n",
 941 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.877026 -   0.7s\n",
 942 |       "[Parallel(n_jobs=1)]: Done  82 tasks       | elapsed:   32.2s\n",
 943 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n",
 944 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.875145 -   0.7s\n",
 945 |       "[Parallel(n_jobs=1)]: Done  83 tasks       | elapsed:   32.9s\n",
 946 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n",
 947 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.861062 -   0.7s\n",
 948 |       "[Parallel(n_jobs=1)]: Done  84 tasks       | elapsed:   33.6s\n",
 949 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n",
 950 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.891892 -   0.7s\n",
 951 |       "[Parallel(n_jobs=1)]: Done  85 tasks       | elapsed:   34.3s\n",
 952 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n",
 953 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.873598 -   0.7s\n",
 954 |       "[Parallel(n_jobs=1)]: Done  86 tasks       | elapsed:   35.0s\n",
 955 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n"
 956 |      ]
 957 |     },
 958 |     {
 959 |      "name": "stderr",
 960 |      "output_type": "stream",
 961 |      "text": [
 962 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 963 |       "  'recall', 'true', average, warn_for)\n"
 964 |      ]
 965 |     },
 966 |     {
 967 |      "name": "stdout",
 968 |      "output_type": "stream",
 969 |      "text": [
 970 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.874136 -   0.7s\n",
 971 |       "[Parallel(n_jobs=1)]: Done  87 tasks       | elapsed:   35.7s\n",
 972 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n"
 973 |      ]
 974 |     },
 975 |     {
 976 |      "name": "stderr",
 977 |      "output_type": "stream",
 978 |      "text": [
 979 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 980 |       "  'recall', 'true', average, warn_for)\n"
 981 |      ]
 982 |     },
 983 |     {
 984 |      "name": "stdout",
 985 |      "output_type": "stream",
 986 |      "text": [
 987 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.858345 -   0.7s\n",
 988 |       "[Parallel(n_jobs=1)]: Done  88 tasks       | elapsed:   36.5s\n",
 989 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n"
 990 |      ]
 991 |     },
 992 |     {
 993 |      "name": "stderr",
 994 |      "output_type": "stream",
 995 |      "text": [
 996 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
 997 |       "  'recall', 'true', average, warn_for)\n"
 998 |      ]
 999 |     },
1000 |     {
1001 |      "name": "stdout",
1002 |      "output_type": "stream",
1003 |      "text": [
1004 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.858345 -   0.7s\n",
1005 |       "[Parallel(n_jobs=1)]: Done  89 tasks       | elapsed:   37.2s\n",
1006 |       "[CV] gamma=0.0001, kernel=sigmoid, C=10 ..............................\n"
1007 |      ]
1008 |     },
1009 |     {
1010 |      "name": "stderr",
1011 |      "output_type": "stream",
1012 |      "text": [
1013 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
1014 |       "  'recall', 'true', average, warn_for)\n"
1015 |      ]
1016 |     },
1017 |     {
1018 |      "name": "stdout",
1019 |      "output_type": "stream",
1020 |      "text": [
1021 |       "[CV] ..... gamma=0.0001, kernel=sigmoid, C=10, score=0.886457 -   0.7s\n",
1022 |       "[Parallel(n_jobs=1)]: Done  90 tasks       | elapsed:   37.9s\n",
1023 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n",
1024 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.967514 -   0.2s\n",
1025 |       "[Parallel(n_jobs=1)]: Done  91 tasks       | elapsed:   38.1s\n",
1026 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n",
1027 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.972713 -   0.2s\n",
1028 |       "[Parallel(n_jobs=1)]: Done  92 tasks       | elapsed:   38.3s\n",
1029 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n",
1030 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.967432 -   0.2s\n",
1031 |       "[Parallel(n_jobs=1)]: Done  93 tasks       | elapsed:   38.5s\n",
1032 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n",
1033 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.972645 -   0.2s\n",
1034 |       "[Parallel(n_jobs=1)]: Done  94 tasks       | elapsed:   38.7s\n",
1035 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n",
1036 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.968739 -   0.2s\n",
1037 |       "[Parallel(n_jobs=1)]: Done  95 tasks       | elapsed:   38.9s\n",
1038 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n",
1039 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.960855 -   0.2s\n",
1040 |       "[Parallel(n_jobs=1)]: Done  96 tasks       | elapsed:   39.1s\n",
1041 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n"
1042 |      ]
1043 |     },
1044 |     {
1045 |      "name": "stderr",
1046 |      "output_type": "stream",
1047 |      "text": [
1048 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
1049 |       "  'recall', 'true', average, warn_for)\n"
1050 |      ]
1051 |     },
1052 |     {
1053 |      "name": "stdout",
1054 |      "output_type": "stream",
1055 |      "text": [
1056 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.964377 -   0.2s\n",
1057 |       "[Parallel(n_jobs=1)]: Done  97 tasks       | elapsed:   39.3s\n",
1058 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n"
1059 |      ]
1060 |     },
1061 |     {
1062 |      "name": "stderr",
1063 |      "output_type": "stream",
1064 |      "text": [
1065 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
1066 |       "  'recall', 'true', average, warn_for)\n"
1067 |      ]
1068 |     },
1069 |     {
1070 |      "name": "stdout",
1071 |      "output_type": "stream",
1072 |      "text": [
1073 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.961735 -   0.2s\n",
1074 |       "[Parallel(n_jobs=1)]: Done  98 tasks       | elapsed:   39.5s\n",
1075 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n",
1076 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.968314 -   0.2s\n",
1077 |       "[Parallel(n_jobs=1)]: Done  99 tasks       | elapsed:   39.7s\n",
1078 |       "[CV] gamma=0.0001, kernel=rbf, C=100 .................................\n"
1079 |      ]
1080 |     },
1081 |     {
1082 |      "name": "stderr",
1083 |      "output_type": "stream",
1084 |      "text": [
1085 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
1086 |       "  'recall', 'true', average, warn_for)\n",
1087 |       "/home/lhu/.local/lib/python3.5/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
1088 |       "  'recall', 'true', average, warn_for)\n"
1089 |      ]
1090 |     },
1091 |     {
1092 |      "name": "stdout",
1093 |      "output_type": "stream",
1094 |      "text": [
1095 |       "[CV] ........ gamma=0.0001, kernel=rbf, C=100, score=0.964377 -   0.2s\n",
1096 |       "[Parallel(n_jobs=1)]: Done 100 tasks       | elapsed:   40.0s\n",
1097 |       "[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed:   40.0s finished\n",
1098 |       "\n",
1099 |       "Best parameters set found on development set: \n",
1100 |       "\n",
1101 |       "{'gamma': 'auto', 'kernel': 'rbf', 'C': 10}\n",
1102 |       "\n",
1103 |       "Grid scores on development set:\n",
1104 |       "\n",
1105 |       "0.997 (+/-0.004) for {'gamma': 'auto', 'kernel': 'rbf', 'C': 10}\n",
1106 |       "0.599 (+/-0.976) for {'gamma': 'auto', 'kernel': 'sigmoid', 'C': 1}\n",
1107 |       "0.512 (+/-0.036) for {'gamma': 0.001, 'kernel': 'rbf', 'C': 0.1}\n",
1108 |       "0.995 (+/-0.003) for {'gamma': 0.001, 'kernel': 'linear', 'C': 1000}\n",
1109 |       "0.982 (+/-0.008) for {'gamma': 0.001, 'kernel': 'sigmoid', 'C': 100}\n",
1110 |       "0.599 (+/-0.976) for {'gamma': 0.0001, 'kernel': 'rbf', 'C': 0.1}\n",
1111 |       "0.994 (+/-0.004) for {'gamma': 0.001, 'kernel': 'linear', 'C': 10}\n",
1112 |       "0.985 (+/-0.007) for {'gamma': 0.001, 'kernel': 'rbf', 'C': 100}\n",
1113 |       "0.873 (+/-0.021) for {'gamma': 0.0001, 'kernel': 'sigmoid', 'C': 10}\n",
1114 |       "0.967 (+/-0.008) for {'gamma': 0.0001, 'kernel': 'rbf', 'C': 100}\n",
1115 |       "\n",
1116 |       "Detailed classification report:\n",
1117 |       "\n",
1118 |       "Report on Training Set\n",
1119 |       "             precision    recall  f1-score   support\n",
1120 |       "\n",
1121 |       "          0       1.00      1.00      1.00      4072\n",
1122 |       "          1       0.27      1.00      0.43         6\n",
1123 |       "\n",
1124 |       "avg / total       1.00      1.00      1.00      4078\n",
1125 |       "\n",
1126 |       "\n",
1127 |       "[[4056   16]\n",
1128 |       " [   0    6]]\n",
1129 |       "Specificity: 0.996070726916\n",
1130 |       "Report on Test Set\n",
1131 |       "\n",
1132 |       "             precision    recall  f1-score   support\n",
1133 |       "\n",
1134 |       "          0       1.00      0.99      0.99      4059\n",
1135 |       "          1       0.00      0.00      0.00        19\n",
1136 |       "\n",
1137 |       "avg / total       0.99      0.99      0.99      4078\n",
1138 |       "\n",
1139 |       "\n",
1140 |       "[[4037   22]\n",
1141 |       " [  19    0]]\n",
1142 |       "Specificity: 0.994579945799\n"
1143 |      ]
1144 |     },
1145 |     {
1146 |      "data": {
1147 |       "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeEAAAGsCAYAAADqjB+8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n1IqWRmxaxt4AAAAASUVORK5CYII=\n",
1148 |       "text/plain": [
1149 |        "<matplotlib.figure.Figure at 0x7fb652018e48>"
1150 |       ]
1151 |      },
1152 |      "metadata": {},
1153 |      "output_type": "display_data"
1154 |     },
1155 |     {
1156 |      "name": "stdout",
1157 |      "output_type": "stream",
1158 |      "text": [
1159 |       "{'gamma': 'auto', 'kernel': 'rbf', 'C': 10}\n"
1160 |      ]
1161 |     }
1162 |    ],
1163 |    "source": [
1164 |     "x_train = x_train.loc[:, features[0:num_features]]\n",
1165 |     "x_test = x_test.loc[:, features[0:num_features]]\n",
1166 |     "CVTrainAndScore(x_train, y_train, x_test, y_test, x_train_sample_weights, output_filename=output_filename)"
1167 |    ]
1168 |   },
1169 |   {
1170 |    "cell_type": "code",
1171 |    "execution_count": 21,
1172 |    "metadata": {
1173 |     "collapsed": false
1174 |    },
1175 |    "outputs": [
1176 |     {
1177 |      "data": {
1178 |       "text/plain": [
1179 |        "Index(['Diff_P_technical_P_majeure_mean', 'Dif_rotor_1_avg_temps_t-4',\n",
1180 |        "       'CS101__Rear_bearing_temp_std', 'Diff_Avg_Sys_2_inverter_4_t-6',\n",
1181 |        "       'Diff_P_technical_P_external_std', 'CS101__Ambient_temp_mean',\n",
1182 |        "       'Diff_P_technical_P_external_mean', 'CS101__Control_cabinet_temp_std',\n",
1183 |        "       'Diff_Avg_Sys_1_inverter_6_std', 'Dif_max_avg_rotation_std'],\n",
1184 |        "      dtype='object')"
1185 |       ]
1186 |      },
1187 |      "execution_count": 21,
1188 |      "metadata": {},
1189 |      "output_type": "execute_result"
1190 |     }
1191 |    ],
1192 |    "source": [
1193 |     "x_train.columns"
1194 |    ]
1195 |   }
1196 |  ],
1197 |  "metadata": {
1198 |   "kernelspec": {
1199 |    "display_name": "Python 3",
1200 |    "language": "python",
1201 |    "name": "python3"
1202 |   },
1203 |   "language_info": {
1204 |    "codemirror_mode": {
1205 |     "name": "ipython",
1206 |     "version": 3
1207 |    },
1208 |    "file_extension": ".py",
1209 |    "mimetype": "text/x-python",
1210 |    "name": "python",
1211 |    "nbconvert_exporter": "python",
1212 |    "pygments_lexer": "ipython3",
1213 |    "version": "3.5.1"
1214 |   }
1215 |  },
1216 |  "nbformat": 4,
1217 |  "nbformat_minor": 0
1218 | }
1219 | 


--------------------------------------------------------------------------------