In [175]:

    

import sklearn.kernel_ridge as krr
import sklearn.preprocessing as pre
import pandas as pd
import numpy as np


    

In [2]:

    

train = pd.DataFrame.from_csv('data/other/train_df.csv')
val = pd.DataFrame.from_csv('data/other/val_df.csv')
test = pd.DataFrame.from_csv('data/other/test_df.csv')


    

In [124]:

    

train_examples = train.ix[:,0:26]
train_labels = train.ix[:,26]

val_examples = val.ix[:,0:26]
val_labels = val.ix[:,26]

test_examples = test.ix[:,0:26]
test_labels = test.ix[:,26]

X = pre.normalize(train_examples.values)
y = train_labels.values

X_val = pre.normalize(val_examples.values)
y_val = val_labels.values

X_test = pre.normalize(test_examples.values)
y_test = test_labels.values


    

In [127]:

    




    

    
Out[127]:





array([[  6.95467813e-05,   6.95467813e-05,   6.95467813e-05, ...,
          5.05373277e-03,   9.59455803e-03,   1.34546942e-01],
       [  1.38383343e-04,   1.38383343e-04,   1.38383343e-04, ...,
          1.50491886e-03,   1.29446086e-02,   1.52146720e-01],
       [  8.65959679e-05,   8.65959679e-05,   8.65959679e-05, ...,
          1.93036845e-03,   1.30002197e-02,   1.42435935e-01],
       ..., 
       [  5.05185466e-04,   5.47689051e-04,   5.07614243e-04, ...,
          1.09537810e-03,   8.81645790e-03,   6.57226861e-02],
       [  1.89794202e-04,   2.05762464e-04,   1.90706674e-04, ...,
          5.57064232e-04,   1.33243742e-02,   2.48003992e-01],
       [  2.00118306e-02,   2.16955183e-02,   2.01080413e-02, ...,
          1.53456105e-02,   1.14298340e-01,   7.76276226e-01]])

In [73]:

    

# Normalize


    

    
Out[73]:





array([ 0.146,  0.084,  0.122, ...,  0.653,  0.66 ,  0.506])

In [145]:

    

clf = krr.KernelRidge()


    

In [146]:

    

clf.fit(X, y)


    

    
Out[146]:





KernelRidge(alpha=1, coef0=1, degree=3, gamma=None, kernel='linear',
      kernel_params=None)

In [147]:

    

math.sqrt(met.mean_squared_error(y, clf.predict(X)))


    

    
Out[147]:





0.23400804131743755

In [148]:

    

math.sqrt(met.mean_squared_error(y_val, clf.predict(X_val)))


    

    
Out[148]:





0.2070386693834345

In [149]:

    

clf.predict(X_val)


    

    
Out[149]:





array([ 0.26687233,  0.26809746,  0.26691671, ...,  0.26012334,
        0.27752264,  0.24498193])

In [150]:

    

math.sqrt(met.mean_squared_error(y_test, clf.predict(X_test)))


    

    
Out[150]:





0.26048690247546463

In [18]:

    

clf.score(val_examples.as_matrix(), val_labels.as_matrix())


    

    
Out[18]:





-2.8146170752585116

In [20]:

    

k = clf.predict(val_examples.as_matrix())


    

In [23]:

    




    

    
Out[23]:





2.0980579865388392e-07

In [24]:

    

import sklearn.metrics as met
import math
import sklearn.svm as svm


    

In [17]:

    

math.sqrt(met.mean_squared_error(val_labels.as_matrix(), clf.predict(val_examples.as_matrix())))


    

    
Out[17]:





0.3900308396796226

In [167]:

    

clf2 = svm.SVR(kernel='poly')


    

In [168]:

    

ex = clf2.fit(X, y).predict(X_val)


    

In [169]:

    

math.sqrt(met.mean_squared_error(y, clf2.predict(X)))


    

    
Out[169]:





0.239447126122699

In [170]:

    

math.sqrt(met.mean_squared_error(y_val, clf2.predict(X_val)))


    

    
Out[170]:





0.2229353363958645

In [171]:

    

math.sqrt(met.mean_squared_error(y_test, clf2.predict(X_test)))


    

    
Out[171]:





0.2697833225478837

In [43]:

    

math.sqrt(met.mean_squared_error(val_labels.as_matrix(), clf2.predict(val_examples.as_matrix())))


    

    
Out[43]:





0.20725703898361297

In [51]:

    

math.sqrt(met.mean_squared_error(test_labels.as_matrix(), clf2.predict(test_examples.as_matrix())))


    

    
Out[51]:





0.2478506070670278

In [52]:

    

clf2.predict(test_examples.as_matrix())


    

    
Out[52]:





array([ 0.27956674,  0.27956674,  0.27956674, ...,  0.27956674,
        0.27956674,  0.27956674])

In [54]:

    

val_examples.as_matrix()


    

    
Out[54]:





array([[  2.40000000e+01,   2.40000000e+01,   2.40000000e+01, ...,
          1.74400000e+03,   3.31100000e+03,   4.64310000e+04],
       [  2.40000000e+01,   2.40000000e+01,   2.40000000e+01, ...,
          2.61000000e+02,   2.24500000e+03,   2.63870000e+04],
       [  2.40000000e+01,   2.40000000e+01,   2.40000000e+01, ...,
          5.35000000e+02,   3.60300000e+03,   3.94760000e+04],
       ..., 
       [  3.78181818e+01,   4.10000000e+01,   3.80000000e+01, ...,
          8.20000000e+01,   6.60000000e+02,   4.92000000e+03],
       [  3.78181818e+01,   4.10000000e+01,   3.80000000e+01, ...,
          1.11000000e+02,   2.65500000e+03,   4.94170000e+04],
       [  3.78181818e+01,   4.10000000e+01,   3.80000000e+01, ...,
          2.90000000e+01,   2.16000000e+02,   1.46700000e+03]])

In [58]:

    

import sklearn.model_selection as gs
import scipy.stats


    

In [59]:

    

p_dict = {'C': scipy.stats.expon(scale=10000), 'gamma': scipy.stats.expon(scale=.1),
  'kernel': ['rbf'], 'epsilon': scipy.stats.expon(scale=0.001)}


    

In [65]:

    

searcher = gs.RandomizedSearchCV(svm.SVR(), param_distributions=p_dict, n_iter=1)


    

In [66]:

    

searcher.fit(train_examples.as_matrix(), train_labels.as_matrix())


    

    
Out[66]:





RandomizedSearchCV(cv=None, error_score='raise',
          estimator=SVR(C=1.0, cache_size=200, coef0=0.0, degree=3, epsilon=0.1, gamma='auto',
  kernel='rbf', max_iter=-1, shrinking=True, tol=0.001, verbose=False),
          fit_params={}, iid=True, n_iter=1, n_jobs=1,
          param_distributions={'C': <scipy.stats._distn_infrastructure.rv_frozen object at 0x106c22908>, 'gamma': <scipy.stats._distn_infrastructure.rv_frozen object at 0x106c22cc0>, 'kernel': ['rbf'], 'epsilon': <scipy.stats._distn_infrastructure.rv_frozen object at 0x106c30f28>},
          pre_dispatch='2*n_jobs', random_state=None, refit=True,
          return_train_score=True, scoring=None, verbose=0)

In [69]:

    

searcher.predict(val_examples.as_matrix())


    

    
Out[69]:





array([ 0.27983551,  0.27983551,  0.27983551, ...,  0.27983551,
        0.27983551,  0.27983551])

In [185]:

    

kr = gs.GridSearchCV(krr.KernelRidge(kernel='rbf', gamma=0.1), cv=5,
                  param_grid={"alpha": [1e0, 0.1, 1e-2],
                              "gamma": np.logspace(1, 5, 5)}, verbose=10)


    

In [186]:

    

kr.fit(X, y)


    

    




Fitting 5 folds for each of 15 candidates, totalling 75 fits
[CV] alpha=1.0, gamma=10.0 ...........................................
[CV] ............ alpha=1.0, gamma=10.0, score=0.118352, total=  12.3s
[CV] alpha=1.0, gamma=10.0 ...........................................






    




[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:   16.1s remaining:    0.0s






    




[CV] ........... alpha=1.0, gamma=10.0, score=-0.064461, total=  13.8s
[CV] alpha=1.0, gamma=10.0 ...........................................






    




[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:   34.4s remaining:    0.0s






    




[CV] ............ alpha=1.0, gamma=10.0, score=0.069878, total=  15.0s
[CV] alpha=1.0, gamma=10.0 ...........................................






    




[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:   53.5s remaining:    0.0s






    




[CV] ............ alpha=1.0, gamma=10.0, score=0.084519, total=  13.6s
[CV] alpha=1.0, gamma=10.0 ...........................................






    




[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:  1.2min remaining:    0.0s






    




[CV] ............ alpha=1.0, gamma=10.0, score=0.062493, total=  13.0s
[CV] alpha=1.0, gamma=100.0 ..........................................






    




[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:  1.5min remaining:    0.0s






    




[CV] ........... alpha=1.0, gamma=100.0, score=0.150999, total=  12.6s
[CV] alpha=1.0, gamma=100.0 ..........................................






    




[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:  1.7min remaining:    0.0s






    




[CV] .......... alpha=1.0, gamma=100.0, score=-0.088722, total=  12.8s
[CV] alpha=1.0, gamma=100.0 ..........................................






    




[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:  2.0min remaining:    0.0s






    




[CV] ........... alpha=1.0, gamma=100.0, score=0.060110, total=  12.7s
[CV] alpha=1.0, gamma=100.0 ..........................................






    




[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:  2.3min remaining:    0.0s






    




[CV] ........... alpha=1.0, gamma=100.0, score=0.127000, total=  12.4s
[CV] alpha=1.0, gamma=100.0 ..........................................






    




[Parallel(n_jobs=1)]: Done   9 out of   9 | elapsed:  2.6min remaining:    0.0s






    




[CV] ........... alpha=1.0, gamma=100.0, score=0.081555, total=  12.4s
[CV] alpha=1.0, gamma=1000.0 .........................................
[CV] .......... alpha=1.0, gamma=1000.0, score=0.064873, total=  17.2s
[CV] alpha=1.0, gamma=1000.0 .........................................
[CV] ......... alpha=1.0, gamma=1000.0, score=-0.144168, total=  14.4s
[CV] alpha=1.0, gamma=1000.0 .........................................
[CV] ......... alpha=1.0, gamma=1000.0, score=-0.114146, total=  14.3s
[CV] alpha=1.0, gamma=1000.0 .........................................
[CV] .......... alpha=1.0, gamma=1000.0, score=0.026417, total=  13.7s
[CV] alpha=1.0, gamma=1000.0 .........................................
[CV] .......... alpha=1.0, gamma=1000.0, score=0.025133, total=  15.0s
[CV] alpha=1.0, gamma=10000.0 ........................................
[CV] ........ alpha=1.0, gamma=10000.0, score=-0.195577, total=  26.3s
[CV] alpha=1.0, gamma=10000.0 ........................................
[CV] ........ alpha=1.0, gamma=10000.0, score=-0.312111, total=  29.2s
[CV] alpha=1.0, gamma=10000.0 ........................................
[CV] ........ alpha=1.0, gamma=10000.0, score=-0.512290, total=  22.5s
[CV] alpha=1.0, gamma=10000.0 ........................................
[CV] ........ alpha=1.0, gamma=10000.0, score=-0.270818, total=  24.2s
[CV] alpha=1.0, gamma=10000.0 ........................................
[CV] ........ alpha=1.0, gamma=10000.0, score=-0.252810, total=  23.3s
[CV] alpha=1.0, gamma=100000.0 .......................................
[CV] ....... alpha=1.0, gamma=100000.0, score=-0.769446, total=  30.5s
[CV] alpha=1.0, gamma=100000.0 .......................................
[CV] ....... alpha=1.0, gamma=100000.0, score=-0.630932, total=  28.6s
[CV] alpha=1.0, gamma=100000.0 .......................................
[CV] ....... alpha=1.0, gamma=100000.0, score=-1.083636, total=  31.0s
[CV] alpha=1.0, gamma=100000.0 .......................................
[CV] ....... alpha=1.0, gamma=100000.0, score=-0.879237, total=  30.1s
[CV] alpha=1.0, gamma=100000.0 .......................................
[CV] ....... alpha=1.0, gamma=100000.0, score=-0.691568, total=  29.7s
[CV] alpha=0.1, gamma=10.0 ...........................................
[CV] ............ alpha=0.1, gamma=10.0, score=0.193897, total=  12.8s
[CV] alpha=0.1, gamma=10.0 ...........................................
[CV] ........... alpha=0.1, gamma=10.0, score=-0.081386, total=  13.9s
[CV] alpha=0.1, gamma=10.0 ...........................................
[CV] ............ alpha=0.1, gamma=10.0, score=0.097401, total=  13.5s
[CV] alpha=0.1, gamma=10.0 ...........................................
[CV] ............ alpha=0.1, gamma=10.0, score=0.121057, total=  13.2s
[CV] alpha=0.1, gamma=10.0 ...........................................
[CV] ............ alpha=0.1, gamma=10.0, score=0.106056, total=  12.4s
[CV] alpha=0.1, gamma=100.0 ..........................................
[CV] ........... alpha=0.1, gamma=100.0, score=0.243052, total=  12.9s
[CV] alpha=0.1, gamma=100.0 ..........................................
[CV] .......... alpha=0.1, gamma=100.0, score=-0.079294, total=  12.5s
[CV] alpha=0.1, gamma=100.0 ..........................................
[CV] ........... alpha=0.1, gamma=100.0, score=0.045935, total=  12.4s
[CV] alpha=0.1, gamma=100.0 ..........................................
[CV] ........... alpha=0.1, gamma=100.0, score=0.187193, total=  12.6s
[CV] alpha=0.1, gamma=100.0 ..........................................
[CV] ........... alpha=0.1, gamma=100.0, score=0.148834, total=  12.9s
[CV] alpha=0.1, gamma=1000.0 .........................................
[CV] .......... alpha=0.1, gamma=1000.0, score=0.158021, total=  17.8s
[CV] alpha=0.1, gamma=1000.0 .........................................
[CV] ......... alpha=0.1, gamma=1000.0, score=-0.134451, total=  14.0s
[CV] alpha=0.1, gamma=1000.0 .........................................
[CV] ......... alpha=0.1, gamma=1000.0, score=-0.101672, total=  13.9s
[CV] alpha=0.1, gamma=1000.0 .........................................
[CV] .......... alpha=0.1, gamma=1000.0, score=0.096558, total=  13.3s
[CV] alpha=0.1, gamma=1000.0 .........................................
[CV] .......... alpha=0.1, gamma=1000.0, score=0.099233, total=  13.6s
[CV] alpha=0.1, gamma=10000.0 ........................................
[CV] ........ alpha=0.1, gamma=10000.0, score=-0.129101, total=  23.9s
[CV] alpha=0.1, gamma=10000.0 ........................................
[CV] ........ alpha=0.1, gamma=10000.0, score=-0.321992, total=  27.2s
[CV] alpha=0.1, gamma=10000.0 ........................................
[CV] ........ alpha=0.1, gamma=10000.0, score=-0.570719, total=  20.6s
[CV] alpha=0.1, gamma=10000.0 ........................................
[CV] ........ alpha=0.1, gamma=10000.0, score=-0.284257, total=  23.4s
[CV] alpha=0.1, gamma=10000.0 ........................................
[CV] ........ alpha=0.1, gamma=10000.0, score=-0.231309, total=  21.8s
[CV] alpha=0.1, gamma=100000.0 .......................................
[CV] ....... alpha=0.1, gamma=100000.0, score=-0.742079, total=  29.7s
[CV] alpha=0.1, gamma=100000.0 .......................................
[CV] ....... alpha=0.1, gamma=100000.0, score=-0.615172, total=  29.2s
[CV] alpha=0.1, gamma=100000.0 .......................................
[CV] ....... alpha=0.1, gamma=100000.0, score=-1.201545, total=  29.2s
[CV] alpha=0.1, gamma=100000.0 .......................................
[CV] ....... alpha=0.1, gamma=100000.0, score=-0.881265, total=  30.3s
[CV] alpha=0.1, gamma=100000.0 .......................................
[CV] ....... alpha=0.1, gamma=100000.0, score=-0.705271, total=  30.9s
[CV] alpha=0.01, gamma=10.0 ..........................................
[CV] ........... alpha=0.01, gamma=10.0, score=0.268008, total=  13.2s
[CV] alpha=0.01, gamma=10.0 ..........................................
[CV] .......... alpha=0.01, gamma=10.0, score=-0.068484, total=  12.7s
[CV] alpha=0.01, gamma=10.0 ..........................................
[CV] ........... alpha=0.01, gamma=10.0, score=0.114186, total=  12.5s
[CV] alpha=0.01, gamma=10.0 ..........................................
[CV] ........... alpha=0.01, gamma=10.0, score=0.151439, total=  12.6s
[CV] alpha=0.01, gamma=10.0 ..........................................
[CV] ........... alpha=0.01, gamma=10.0, score=0.149585, total=  13.4s
[CV] alpha=0.01, gamma=100.0 .........................................
[CV] .......... alpha=0.01, gamma=100.0, score=0.309940, total=  14.1s
[CV] alpha=0.01, gamma=100.0 .........................................
[CV] ......... alpha=0.01, gamma=100.0, score=-0.055163, total=  13.7s
[CV] alpha=0.01, gamma=100.0 .........................................
[CV] ......... alpha=0.01, gamma=100.0, score=-0.052814, total=  13.5s
[CV] alpha=0.01, gamma=100.0 .........................................
[CV] .......... alpha=0.01, gamma=100.0, score=0.195665, total=  12.7s
[CV] alpha=0.01, gamma=100.0 .........................................
[CV] .......... alpha=0.01, gamma=100.0, score=0.195429, total=  12.8s
[CV] alpha=0.01, gamma=1000.0 ........................................
[CV] ......... alpha=0.01, gamma=1000.0, score=0.173506, total=  17.1s
[CV] alpha=0.01, gamma=1000.0 ........................................
[CV] ........ alpha=0.01, gamma=1000.0, score=-0.135254, total=  13.7s
[CV] alpha=0.01, gamma=1000.0 ........................................
[CV] ........ alpha=0.01, gamma=1000.0, score=-0.181153, total=  13.4s
[CV] alpha=0.01, gamma=1000.0 ........................................
[CV] ......... alpha=0.01, gamma=1000.0, score=0.134063, total=  13.0s
[CV] alpha=0.01, gamma=1000.0 ........................................
[CV] ......... alpha=0.01, gamma=1000.0, score=0.093165, total=  13.2s
[CV] alpha=0.01, gamma=10000.0 .......................................
[CV] ....... alpha=0.01, gamma=10000.0, score=-0.174982, total=  23.2s
[CV] alpha=0.01, gamma=10000.0 .......................................
[CV] ....... alpha=0.01, gamma=10000.0, score=-0.388614, total=  27.2s
[CV] alpha=0.01, gamma=10000.0 .......................................
[CV] ....... alpha=0.01, gamma=10000.0, score=-0.847071, total=  20.4s
[CV] alpha=0.01, gamma=10000.0 .......................................
[CV] ....... alpha=0.01, gamma=10000.0, score=-0.427115, total=  22.3s
[CV] alpha=0.01, gamma=10000.0 .......................................
[CV] ....... alpha=0.01, gamma=10000.0, score=-0.339265, total=  22.9s
[CV] alpha=0.01, gamma=100000.0 ......................................
[CV] ...... alpha=0.01, gamma=100000.0, score=-0.891558, total=  29.6s
[CV] alpha=0.01, gamma=100000.0 ......................................
[CV] ...... alpha=0.01, gamma=100000.0, score=-0.699446, total=  30.7s
[CV] alpha=0.01, gamma=100000.0 ......................................
[CV] ...... alpha=0.01, gamma=100000.0, score=-1.635284, total=  32.6s
[CV] alpha=0.01, gamma=100000.0 ......................................
[CV] ...... alpha=0.01, gamma=100000.0, score=-1.143411, total=  29.9s
[CV] alpha=0.01, gamma=100000.0 ......................................
[CV] ...... alpha=0.01, gamma=100000.0, score=-0.940020, total=  29.9s






    




[Parallel(n_jobs=1)]: Done  75 out of  75 | elapsed: 28.5min finished






    
Out[186]:





GridSearchCV(cv=5, error_score='raise',
       estimator=KernelRidge(alpha=1, coef0=1, degree=3, gamma=0.1, kernel='rbf',
      kernel_params=None),
       fit_params={}, iid=True, n_jobs=1,
       param_grid={'alpha': [1.0, 0.1, 0.01], 'gamma': array([  1.00000e+01,   1.00000e+02,   1.00000e+03,   1.00000e+04,
         1.00000e+05])},
       pre_dispatch='2*n_jobs', refit=True, return_train_score=True,
       scoring=None, verbose=10)

In [187]:

    

kr.best_params_


    

    
Out[187]:





{'alpha': 0.01, 'gamma': 10.0}

In [188]:

    

math.sqrt(met.mean_squared_error(y, kr.predict(X)))


    

    
Out[188]:





0.20863364613006544

In [189]:

    

math.sqrt(met.mean_squared_error(y_val, kr.predict(X_val)))


    

    
Out[189]:





0.1834304519023655

In [190]:

    

math.sqrt(met.mean_squared_error(y_test, kr.predict(X_test)))


    

    
Out[190]:





0.2689755096776497

In [191]:

    

kr.best_params_


    

    
Out[191]:





{'alpha': 0.01, 'gamma': 10.0}

In [194]:

    

svr = gs.GridSearchCV(svm.SVR(kernel='rbf', gamma=0.1), cv=5,
                   param_grid={"C": [1e0, 1e1, 1e2, 1e3],
                               "gamma": np.logspace(-2, 2, 5)}, verbose=10)


    

In [195]:

    

svr.fit(X, y)


    

    




Fitting 5 folds for each of 20 candidates, totalling 100 fits
[CV] C=1.0, gamma=0.01 ...............................................
[CV] ................ C=1.0, gamma=0.01, score=0.008107, total=   6.3s
[CV] C=1.0, gamma=0.01 ...............................................






    




[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    9.9s remaining:    0.0s






    




[CV] ............... C=1.0, gamma=0.01, score=-0.060541, total=   5.9s
[CV] C=1.0, gamma=0.01 ...............................................






    




[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:   19.4s remaining:    0.0s






    




[CV] ............... C=1.0, gamma=0.01, score=-0.063337, total=   6.1s
[CV] C=1.0, gamma=0.01 ...............................................






    




[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:   29.2s remaining:    0.0s






    




[CV] ................ C=1.0, gamma=0.01, score=0.020375, total=   6.6s
[CV] C=1.0, gamma=0.01 ...............................................






    




[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:   39.7s remaining:    0.0s






    




[CV] ............... C=1.0, gamma=0.01, score=-0.039439, total=   6.2s
[CV] C=1.0, gamma=0.1 ................................................






    




[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:   49.8s remaining:    0.0s






    




[CV] ................. C=1.0, gamma=0.1, score=0.030829, total=   6.2s
[CV] C=1.0, gamma=0.1 ................................................






    




[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:   59.8s remaining:    0.0s






    




[CV] ................ C=1.0, gamma=0.1, score=-0.070639, total=   5.9s
[CV] C=1.0, gamma=0.1 ................................................






    




[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:  1.2min remaining:    0.0s






    




[CV] ................ C=1.0, gamma=0.1, score=-0.030182, total=   6.7s
[CV] C=1.0, gamma=0.1 ................................................






    




[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:  1.3min remaining:    0.0s






    




[CV] ................. C=1.0, gamma=0.1, score=0.047875, total=   6.5s
[CV] C=1.0, gamma=0.1 ................................................






    




[Parallel(n_jobs=1)]: Done   9 out of   9 | elapsed:  1.5min remaining:    0.0s






    




[CV] ................ C=1.0, gamma=0.1, score=-0.015380, total=   6.9s
[CV] C=1.0, gamma=1.0 ................................................
[CV] ................. C=1.0, gamma=1.0, score=0.089868, total=   6.6s
[CV] C=1.0, gamma=1.0 ................................................
[CV] ................ C=1.0, gamma=1.0, score=-0.108724, total=   6.3s
[CV] C=1.0, gamma=1.0 ................................................
[CV] ................. C=1.0, gamma=1.0, score=0.008967, total=   6.5s
[CV] C=1.0, gamma=1.0 ................................................
[CV] ................. C=1.0, gamma=1.0, score=0.055787, total=   6.4s
[CV] C=1.0, gamma=1.0 ................................................
[CV] ................. C=1.0, gamma=1.0, score=0.018609, total=   6.2s
[CV] C=1.0, gamma=10.0 ...............................................
[CV] ................ C=1.0, gamma=10.0, score=0.160835, total=   5.9s
[CV] C=1.0, gamma=10.0 ...............................................
[CV] ............... C=1.0, gamma=10.0, score=-0.144667, total=   6.1s
[CV] C=1.0, gamma=10.0 ...............................................
[CV] ................ C=1.0, gamma=10.0, score=0.050787, total=   6.4s
[CV] C=1.0, gamma=10.0 ...............................................
[CV] ................ C=1.0, gamma=10.0, score=0.111105, total=   6.0s
[CV] C=1.0, gamma=10.0 ...............................................
[CV] ................ C=1.0, gamma=10.0, score=0.061344, total=   6.1s
[CV] C=1.0, gamma=100.0 ..............................................
[CV] ............... C=1.0, gamma=100.0, score=0.218014, total=   7.1s
[CV] C=1.0, gamma=100.0 ..............................................
[CV] .............. C=1.0, gamma=100.0, score=-0.139532, total=   6.5s
[CV] C=1.0, gamma=100.0 ..............................................
[CV] ............... C=1.0, gamma=100.0, score=0.065283, total=   6.7s
[CV] C=1.0, gamma=100.0 ..............................................
[CV] ............... C=1.0, gamma=100.0, score=0.174743, total=   7.6s
[CV] C=1.0, gamma=100.0 ..............................................
[CV] ............... C=1.0, gamma=100.0, score=0.116135, total=   7.3s
[CV] C=10.0, gamma=0.01 ..............................................
[CV] ............... C=10.0, gamma=0.01, score=0.027515, total=   6.7s
[CV] C=10.0, gamma=0.01 ..............................................
[CV] .............. C=10.0, gamma=0.01, score=-0.064427, total=   6.1s
[CV] C=10.0, gamma=0.01 ..............................................
[CV] .............. C=10.0, gamma=0.01, score=-0.044469, total=   6.8s
[CV] C=10.0, gamma=0.01 ..............................................
[CV] ............... C=10.0, gamma=0.01, score=0.031419, total=   6.5s
[CV] C=10.0, gamma=0.01 ..............................................
[CV] .............. C=10.0, gamma=0.01, score=-0.023918, total=   6.6s
[CV] C=10.0, gamma=0.1 ...............................................
[CV] ................ C=10.0, gamma=0.1, score=0.060138, total=   6.7s
[CV] C=10.0, gamma=0.1 ...............................................
[CV] ............... C=10.0, gamma=0.1, score=-0.092700, total=   5.8s
[CV] C=10.0, gamma=0.1 ...............................................
[CV] ............... C=10.0, gamma=0.1, score=-0.003730, total=   6.5s
[CV] C=10.0, gamma=0.1 ...............................................
[CV] ................ C=10.0, gamma=0.1, score=0.014720, total=   6.4s
[CV] C=10.0, gamma=0.1 ...............................................
[CV] ................ C=10.0, gamma=0.1, score=0.010790, total=   6.5s
[CV] C=10.0, gamma=1.0 ...............................................
[CV] ................ C=10.0, gamma=1.0, score=0.147792, total=   6.6s
[CV] C=10.0, gamma=1.0 ...............................................
[CV] ............... C=10.0, gamma=1.0, score=-0.138138, total=   6.0s
[CV] C=10.0, gamma=1.0 ...............................................
[CV] ................ C=10.0, gamma=1.0, score=0.046594, total=   6.6s
[CV] C=10.0, gamma=1.0 ...............................................
[CV] ................ C=10.0, gamma=1.0, score=0.066917, total=   6.9s
[CV] C=10.0, gamma=1.0 ...............................................
[CV] ................ C=10.0, gamma=1.0, score=0.058135, total=   6.8s
[CV] C=10.0, gamma=10.0 ..............................................
[CV] ............... C=10.0, gamma=10.0, score=0.241398, total=   6.9s
[CV] C=10.0, gamma=10.0 ..............................................
[CV] .............. C=10.0, gamma=10.0, score=-0.144938, total=   6.8s
[CV] C=10.0, gamma=10.0 ..............................................
[CV] ............... C=10.0, gamma=10.0, score=0.105497, total=   7.4s
[CV] C=10.0, gamma=10.0 ..............................................
[CV] ............... C=10.0, gamma=10.0, score=0.164871, total=   7.5s
[CV] C=10.0, gamma=10.0 ..............................................
[CV] ............... C=10.0, gamma=10.0, score=0.106260, total=   7.2s
[CV] C=10.0, gamma=100.0 .............................................
[CV] .............. C=10.0, gamma=100.0, score=0.291592, total=  11.6s
[CV] C=10.0, gamma=100.0 .............................................
[CV] ............. C=10.0, gamma=100.0, score=-0.111705, total=   9.8s
[CV] C=10.0, gamma=100.0 .............................................
[CV] .............. C=10.0, gamma=100.0, score=0.032269, total=  10.3s
[CV] C=10.0, gamma=100.0 .............................................
[CV] .............. C=10.0, gamma=100.0, score=0.226382, total=  11.6s
[CV] C=10.0, gamma=100.0 .............................................
[CV] .............. C=10.0, gamma=100.0, score=0.186434, total=  11.3s
[CV] C=100.0, gamma=0.01 .............................................
[CV] .............. C=100.0, gamma=0.01, score=0.050463, total=   6.8s
[CV] C=100.0, gamma=0.01 .............................................
[CV] ............. C=100.0, gamma=0.01, score=-0.075194, total=   6.6s
[CV] C=100.0, gamma=0.01 .............................................
[CV] ............. C=100.0, gamma=0.01, score=-0.015815, total=   6.7s
[CV] C=100.0, gamma=0.01 .............................................
[CV] ............. C=100.0, gamma=0.01, score=-0.000061, total=   6.7s
[CV] C=100.0, gamma=0.01 .............................................
[CV] .............. C=100.0, gamma=0.01, score=0.004607, total=   6.9s
[CV] C=100.0, gamma=0.1 ..............................................
[CV] ............... C=100.0, gamma=0.1, score=0.114250, total=   7.2s
[CV] C=100.0, gamma=0.1 ..............................................
[CV] .............. C=100.0, gamma=0.1, score=-0.111059, total=   6.4s
[CV] C=100.0, gamma=0.1 ..............................................
[CV] ............... C=100.0, gamma=0.1, score=0.040049, total=   7.1s
[CV] C=100.0, gamma=0.1 ..............................................
[CV] ............... C=100.0, gamma=0.1, score=0.005875, total=   6.8s
[CV] C=100.0, gamma=0.1 ..............................................
[CV] ............... C=100.0, gamma=0.1, score=0.044773, total=   6.6s
[CV] C=100.0, gamma=1.0 ..............................................
[CV] ............... C=100.0, gamma=1.0, score=0.205332, total=   7.9s
[CV] C=100.0, gamma=1.0 ..............................................
[CV] .............. C=100.0, gamma=1.0, score=-0.132981, total=   7.4s
[CV] C=100.0, gamma=1.0 ..............................................
[CV] ............... C=100.0, gamma=1.0, score=0.091491, total=   7.8s
[CV] C=100.0, gamma=1.0 ..............................................
[CV] ............... C=100.0, gamma=1.0, score=0.109235, total=   7.8s
[CV] C=100.0, gamma=1.0 ..............................................
[CV] ............... C=100.0, gamma=1.0, score=0.099567, total=   8.1s
[CV] C=100.0, gamma=10.0 .............................................
[CV] .............. C=100.0, gamma=10.0, score=0.283579, total=  13.4s
[CV] C=100.0, gamma=10.0 .............................................
[CV] ............. C=100.0, gamma=10.0, score=-0.113777, total=  13.4s
[CV] C=100.0, gamma=10.0 .............................................
[CV] .............. C=100.0, gamma=10.0, score=0.113940, total=  13.8s
[CV] C=100.0, gamma=10.0 .............................................
[CV] .............. C=100.0, gamma=10.0, score=0.145460, total=  13.4s
[CV] C=100.0, gamma=10.0 .............................................
[CV] .............. C=100.0, gamma=10.0, score=0.140119, total=  13.2s
[CV] C=100.0, gamma=100.0 ............................................
[CV] ............. C=100.0, gamma=100.0, score=0.332723, total=  36.8s
[CV] C=100.0, gamma=100.0 ............................................
[CV] ............ C=100.0, gamma=100.0, score=-0.076672, total=  34.8s
[CV] C=100.0, gamma=100.0 ............................................
[CV] ............ C=100.0, gamma=100.0, score=-0.119292, total=  38.2s
[CV] C=100.0, gamma=100.0 ............................................
[CV] ............. C=100.0, gamma=100.0, score=0.177133, total=  37.2s
[CV] C=100.0, gamma=100.0 ............................................
[CV] ............. C=100.0, gamma=100.0, score=0.178340, total=  35.9s
[CV] C=1000.0, gamma=0.01 ............................................
[CV] ............. C=1000.0, gamma=0.01, score=0.073104, total=   7.1s
[CV] C=1000.0, gamma=0.01 ............................................
[CV] ............ C=1000.0, gamma=0.01, score=-0.092889, total=   6.8s
[CV] C=1000.0, gamma=0.01 ............................................
[CV] ............. C=1000.0, gamma=0.01, score=0.008110, total=   7.3s
[CV] C=1000.0, gamma=0.01 ............................................
[CV] ............ C=1000.0, gamma=0.01, score=-0.055454, total=   7.3s
[CV] C=1000.0, gamma=0.01 ............................................
[CV] ............. C=1000.0, gamma=0.01, score=0.022106, total=   6.9s
[CV] C=1000.0, gamma=0.1 .............................................
[CV] .............. C=1000.0, gamma=0.1, score=0.164043, total=   9.2s
[CV] C=1000.0, gamma=0.1 .............................................
[CV] ............. C=1000.0, gamma=0.1, score=-0.118237, total=   8.6s
[CV] C=1000.0, gamma=0.1 .............................................
[CV] .............. C=1000.0, gamma=0.1, score=0.077162, total=  10.3s
[CV] C=1000.0, gamma=0.1 .............................................
[CV] .............. C=1000.0, gamma=0.1, score=0.027891, total=   9.5s
[CV] C=1000.0, gamma=0.1 .............................................
[CV] .............. C=1000.0, gamma=0.1, score=0.071619, total=  11.0s
[CV] C=1000.0, gamma=1.0 .............................................
[CV] .............. C=1000.0, gamma=1.0, score=0.200216, total=  20.9s
[CV] C=1000.0, gamma=1.0 .............................................
[CV] ............. C=1000.0, gamma=1.0, score=-0.090155, total=  19.4s
[CV] C=1000.0, gamma=1.0 .............................................
[CV] .............. C=1000.0, gamma=1.0, score=0.144339, total=  19.3s
[CV] C=1000.0, gamma=1.0 .............................................
[CV] ............. C=1000.0, gamma=1.0, score=-0.036648, total=  18.9s
[CV] C=1000.0, gamma=1.0 .............................................
[CV] .............. C=1000.0, gamma=1.0, score=0.135942, total=  21.2s
[CV] C=1000.0, gamma=10.0 ............................................
[CV] ............. C=1000.0, gamma=10.0, score=0.162697, total= 1.0min
[CV] C=1000.0, gamma=10.0 ............................................
[CV] ............ C=1000.0, gamma=10.0, score=-0.090666, total= 1.0min
[CV] C=1000.0, gamma=10.0 ............................................
[CV] ............. C=1000.0, gamma=10.0, score=0.059787, total=  56.2s
[CV] C=1000.0, gamma=10.0 ............................................
[CV] ............ C=1000.0, gamma=10.0, score=-0.194101, total= 1.0min
[CV] C=1000.0, gamma=10.0 ............................................
[CV] ............. C=1000.0, gamma=10.0, score=0.048385, total= 1.0min
[CV] C=1000.0, gamma=100.0 ...........................................
[CV] ............ C=1000.0, gamma=100.0, score=0.208142, total= 4.2min
[CV] C=1000.0, gamma=100.0 ...........................................
[CV] ........... C=1000.0, gamma=100.0, score=-0.093134, total= 4.2min
[CV] C=1000.0, gamma=100.0 ...........................................
[CV] ........... C=1000.0, gamma=100.0, score=-1.506402, total= 4.1min
[CV] C=1000.0, gamma=100.0 ...........................................
[CV] ........... C=1000.0, gamma=100.0, score=-0.572679, total= 4.5min
[CV] C=1000.0, gamma=100.0 ...........................................
[CV] ............ C=1000.0, gamma=100.0, score=0.026347, total= 4.7min






    




[Parallel(n_jobs=1)]: Done 100 out of 100 | elapsed: 47.4min finished






    
Out[195]:





GridSearchCV(cv=5, error_score='raise',
       estimator=SVR(C=1.0, cache_size=200, coef0=0.0, degree=3, epsilon=0.1, gamma=0.1,
  kernel='rbf', max_iter=-1, shrinking=True, tol=0.001, verbose=False),
       fit_params={}, iid=True, n_jobs=1,
       param_grid={'C': [1.0, 10.0, 100.0, 1000.0], 'gamma': array([  1.00000e-02,   1.00000e-01,   1.00000e+00,   1.00000e+01,
         1.00000e+02])},
       pre_dispatch='2*n_jobs', refit=True, return_train_score=True,
       scoring=None, verbose=10)

In [196]:

    

svr.best_params_


    

    
Out[196]:





{'C': 10.0, 'gamma': 100.0}

In [197]:

    

math.sqrt(met.mean_squared_error(y, svr.predict(X)))


    

    
Out[197]:





0.19836342543877616

In [198]:

    

math.sqrt(met.mean_squared_error(y_val, svr.predict(X_val)))


    

    
Out[198]:





0.1842507245825642

In [199]:

    

math.sqrt(met.mean_squared_error(y_test, svr.predict(X_test)))


    

    
Out[199]:





0.25975025295629367

In [ ]: