@@ -121,10 +121,11 @@ View the models found by auto-sklearn
121
121
122
122
rank ensemble_weight type cost duration
123
123
model_id
124
- 25 1 0.44 sgd 0.436679 0.599595
125
- 6 2 0.34 ard_regression 0.455042 0.612967
126
- 39 3 0.18 ard_regression 0.474807 0.596258
127
- 7 4 0.04 gradient_boosting 0.518673 1.088064
124
+ 25 1 0.46 sgd 0.436679 0.755302
125
+ 6 2 0.32 ard_regression 0.455042 0.782081
126
+ 27 3 0.14 ard_regression 0.462249 0.746363
127
+ 11 4 0.02 random_forest 0.507400 9.937547
128
+ 7 5 0.06 gradient_boosting 0.518673 1.478772
128
129
129
130
130
131
@@ -151,28 +152,35 @@ Print the final ensemble constructed by auto-sklearn
151
152
152
153
.. code-block :: none
153
154
154
- [(0.440000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'sgd', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False', 'regressor:sgd:alpha': 0.0006517033225329654, 'regressor:sgd:average': 'False', 'regressor:sgd:fit_intercept': 'True', 'regressor:sgd:learning_rate': 'invscaling', 'regressor:sgd:loss': 'epsilon_insensitive', 'regressor:sgd:penalty': 'elasticnet', 'regressor:sgd:tol': 0.002431731981071206, 'regressor:sgd:epsilon': 0.012150149892783745, 'regressor:sgd:eta0': 0.016444224834275295, 'regressor:sgd:l1_ratio': 1.7462342366289323e-09, 'regressor:sgd:power_t': 0.21521743568582094},
155
+ [(0.460000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'sgd', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False', 'regressor:sgd:alpha': 0.0006517033225329654, 'regressor:sgd:average': 'False', 'regressor:sgd:fit_intercept': 'True', 'regressor:sgd:learning_rate': 'invscaling', 'regressor:sgd:loss': 'epsilon_insensitive', 'regressor:sgd:penalty': 'elasticnet', 'regressor:sgd:tol': 0.002431731981071206, 'regressor:sgd:epsilon': 0.012150149892783745, 'regressor:sgd:eta0': 0.016444224834275295, 'regressor:sgd:l1_ratio': 1.7462342366289323e-09, 'regressor:sgd:power_t': 0.21521743568582094},
155
156
dataset_properties={
156
157
'task': 4,
157
158
'sparse': False,
158
159
'multioutput': False,
159
160
'target_type': 'regression',
160
161
'signed': False})),
161
- (0.340000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'ard_regression', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'False', 'regressor:ard_regression:alpha_1': 0.0003701926442639788, 'regressor:ard_regression:alpha_2': 2.2118001735899097e-07, 'regressor:ard_regression:fit_intercept': 'True', 'regressor:ard_regression:lambda_1': 1.2037591637980971e-06, 'regressor:ard_regression:lambda_2': 4.358378124977852e-09, 'regressor:ard_regression:n_iter': 300, 'regressor:ard_regression:threshold_lambda': 1136.5286041327277, 'regressor:ard_regression:tol': 0.021944240404849075},
162
+ (0.320000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'ard_regression', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'False', 'regressor:ard_regression:alpha_1': 0.0003701926442639788, 'regressor:ard_regression:alpha_2': 2.2118001735899097e-07, 'regressor:ard_regression:fit_intercept': 'True', 'regressor:ard_regression:lambda_1': 1.2037591637980971e-06, 'regressor:ard_regression:lambda_2': 4.358378124977852e-09, 'regressor:ard_regression:n_iter': 300, 'regressor:ard_regression:threshold_lambda': 1136.5286041327277, 'regressor:ard_regression:tol': 0.021944240404849075},
162
163
dataset_properties={
163
164
'task': 4,
164
165
'sparse': False,
165
166
'multioutput': False,
166
167
'target_type': 'regression',
167
168
'signed': False})),
168
- (0.180000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'regressor:__choice__': 'ard_regression', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:fast_ica:algorithm': 'parallel', 'feature_preprocessor:fast_ica:fun': 'logcosh', 'feature_preprocessor:fast_ica:whiten': 'False', 'regressor:ard_regression:alpha_1': 0.0005012365297609799, 'regressor:ard_regression:alpha_2': 3.025360750168211e-08, 'regressor:ard_regression:fit_intercept': 'True', 'regressor:ard_regression:lambda_1': 4.9749646614525684e-05, 'regressor:ard_regression:lambda_2': 3.2368037115065363e-10, 'regressor:ard_regression:n_iter': 300, 'regressor:ard_regression:threshold_lambda': 18669.665899307194, 'regressor:ard_regression:tol': 0.0012624032013298571},
169
+ (0.140000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_regression', 'regressor:__choice__': 'ard_regression', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_rates_regression:alpha': 0.3621762718897781, 'feature_preprocessor:select_rates_regression:mode': 'fwe', 'feature_preprocessor:select_rates_regression:score_func': 'f_regression', 'regressor:ard_regression:alpha_1': 2.7664515192592053e-05, 'regressor:ard_regression:alpha_2': 9.504988116581138e-07, 'regressor:ard_regression:fit_intercept': 'True', 'regressor:ard_regression:lambda_1': 6.50650698230178e-09, 'regressor:ard_regression:lambda_2': 4.238533890074848e-07, 'regressor:ard_regression:n_iter': 300, 'regressor:ard_regression:threshold_lambda': 78251.58542976103, 'regressor:ard_regression:tol': 0.0007301343236220855, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0001745391328519669, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8057830372269097, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.24982831110057324},
169
170
dataset_properties={
170
171
'task': 4,
171
172
'sparse': False,
172
173
'multioutput': False,
173
174
'target_type': 'regression',
174
175
'signed': False})),
175
- (0.040000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'regressor:__choice__': 'gradient_boosting', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'regressor:gradient_boosting:early_stop': 'off', 'regressor:gradient_boosting:l2_regularization': 1.8428972335335263e-10, 'regressor:gradient_boosting:learning_rate': 0.012607824914758717, 'regressor:gradient_boosting:loss': 'least_squares', 'regressor:gradient_boosting:max_bins': 255, 'regressor:gradient_boosting:max_depth': 'None', 'regressor:gradient_boosting:max_leaf_nodes': 10, 'regressor:gradient_boosting:min_samples_leaf': 8, 'regressor:gradient_boosting:scoring': 'loss', 'regressor:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 952, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
176
+ (0.060000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'regressor:__choice__': 'gradient_boosting', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'regressor:gradient_boosting:early_stop': 'off', 'regressor:gradient_boosting:l2_regularization': 1.8428972335335263e-10, 'regressor:gradient_boosting:learning_rate': 0.012607824914758717, 'regressor:gradient_boosting:loss': 'least_squares', 'regressor:gradient_boosting:max_bins': 255, 'regressor:gradient_boosting:max_depth': 'None', 'regressor:gradient_boosting:max_leaf_nodes': 10, 'regressor:gradient_boosting:min_samples_leaf': 8, 'regressor:gradient_boosting:scoring': 'loss', 'regressor:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 952, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
177
+ dataset_properties={
178
+ 'task': 4,
179
+ 'sparse': False,
180
+ 'multioutput': False,
181
+ 'target_type': 'regression',
182
+ 'signed': False})),
183
+ (0.020000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'random_forest', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'True', 'regressor:random_forest:bootstrap': 'False', 'regressor:random_forest:criterion': 'mae', 'regressor:random_forest:max_depth': 'None', 'regressor:random_forest:max_features': 0.6277363920171745, 'regressor:random_forest:max_leaf_nodes': 'None', 'regressor:random_forest:min_impurity_decrease': 0.0, 'regressor:random_forest:min_samples_leaf': 6, 'regressor:random_forest:min_samples_split': 15, 'regressor:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0010413452644415357},
176
184
dataset_properties={
177
185
'task': 4,
178
186
'sparse': False,
@@ -213,8 +221,8 @@ predicting the data mean has an R2 score of 0.
213
221
214
222
.. code-block :: none
215
223
216
- Train R2 score: 0.5855373845454157
217
- Test R2 score: 0.39879073225079487
224
+ Train R2 score: 0.5944780427522034
225
+ Test R2 score: 0.3959585042866587
218
226
219
227
220
228
@@ -259,7 +267,7 @@ the true value).
259
267
260
268
.. rst-class :: sphx-glr-timing
261
269
262
- **Total running time of the script: ** ( 2 minutes 12.638 seconds)
270
+ **Total running time of the script: ** ( 1 minutes 56.026 seconds)
263
271
264
272
265
273
.. _sphx_glr_download_examples_20_basic_example_regression.py :
0 commit comments