1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
|
#
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements. See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
import warnings
from pyspark import since
from pyspark.ml.param.shared import *
from pyspark.ml.util import *
from pyspark.ml.wrapper import JavaEstimator, JavaModel
from pyspark.mllib.common import inherit_doc
__all__ = ['AFTSurvivalRegression', 'AFTSurvivalRegressionModel',
'DecisionTreeRegressor', 'DecisionTreeRegressionModel',
'GBTRegressor', 'GBTRegressionModel',
'IsotonicRegression', 'IsotonicRegressionModel',
'LinearRegression', 'LinearRegressionModel',
'RandomForestRegressor', 'RandomForestRegressionModel']
@inherit_doc
class LinearRegression(JavaEstimator, HasFeaturesCol, HasLabelCol, HasPredictionCol, HasMaxIter,
HasRegParam, HasTol, HasElasticNetParam, HasFitIntercept,
HasStandardization, HasSolver, HasWeightCol, JavaMLWritable, JavaMLReadable):
"""
Linear regression.
The learning objective is to minimize the squared error, with regularization.
The specific squared error loss function used is: L = 1/2n ||A coefficients - y||^2^
This support multiple types of regularization:
- none (a.k.a. ordinary least squares)
- L2 (ridge regression)
- L1 (Lasso)
- L2 + L1 (elastic net)
>>> from pyspark.mllib.linalg import Vectors
>>> df = sqlContext.createDataFrame([
... (1.0, 2.0, Vectors.dense(1.0)),
... (0.0, 2.0, Vectors.sparse(1, [], []))], ["label", "weight", "features"])
>>> lr = LinearRegression(maxIter=5, regParam=0.0, solver="normal", weightCol="weight")
>>> model = lr.fit(df)
>>> test0 = sqlContext.createDataFrame([(Vectors.dense(-1.0),)], ["features"])
>>> abs(model.transform(test0).head().prediction - (-1.0)) < 0.001
True
>>> abs(model.coefficients[0] - 1.0) < 0.001
True
>>> abs(model.intercept - 0.0) < 0.001
True
>>> test1 = sqlContext.createDataFrame([(Vectors.sparse(1, [0], [1.0]),)], ["features"])
>>> abs(model.transform(test1).head().prediction - 1.0) < 0.001
True
>>> lr.setParams("vector")
Traceback (most recent call last):
...
TypeError: Method setParams forces keyword arguments.
>>> lr_path = temp_path + "/lr"
>>> lr.save(lr_path)
>>> lr2 = LinearRegression.load(lr_path)
>>> lr2.getMaxIter()
5
>>> model_path = temp_path + "/lr_model"
>>> model.save(model_path)
>>> model2 = LinearRegressionModel.load(model_path)
>>> model.coefficients[0] == model2.coefficients[0]
True
>>> model.intercept == model2.intercept
True
.. versionadded:: 1.4.0
"""
@keyword_only
def __init__(self, featuresCol="features", labelCol="label", predictionCol="prediction",
maxIter=100, regParam=0.0, elasticNetParam=0.0, tol=1e-6, fitIntercept=True,
standardization=True, solver="auto", weightCol=None):
"""
__init__(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
maxIter=100, regParam=0.0, elasticNetParam=0.0, tol=1e-6, fitIntercept=True, \
standardization=True, solver="auto", weightCol=None)
"""
super(LinearRegression, self).__init__()
self._java_obj = self._new_java_obj(
"org.apache.spark.ml.regression.LinearRegression", self.uid)
self._setDefault(maxIter=100, regParam=0.0, tol=1e-6)
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
@keyword_only
@since("1.4.0")
def setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction",
maxIter=100, regParam=0.0, elasticNetParam=0.0, tol=1e-6, fitIntercept=True,
standardization=True, solver="auto", weightCol=None):
"""
setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
maxIter=100, regParam=0.0, elasticNetParam=0.0, tol=1e-6, fitIntercept=True, \
standardization=True, solver="auto", weightCol=None)
Sets params for linear regression.
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
def _create_model(self, java_model):
return LinearRegressionModel(java_model)
class LinearRegressionModel(JavaModel, JavaMLWritable, JavaMLReadable):
"""
Model fitted by LinearRegression.
.. versionadded:: 1.4.0
"""
@property
@since("1.4.0")
def weights(self):
"""
Model weights.
"""
warnings.warn("weights is deprecated. Use coefficients instead.")
return self._call_java("weights")
@property
@since("1.6.0")
def coefficients(self):
"""
Model coefficients.
"""
return self._call_java("coefficients")
@property
@since("1.4.0")
def intercept(self):
"""
Model intercept.
"""
return self._call_java("intercept")
@inherit_doc
class IsotonicRegression(JavaEstimator, HasFeaturesCol, HasLabelCol, HasPredictionCol,
HasWeightCol, JavaMLWritable, JavaMLReadable):
"""
.. note:: Experimental
Currently implemented using parallelized pool adjacent violators algorithm.
Only univariate (single feature) algorithm supported.
>>> from pyspark.mllib.linalg import Vectors
>>> df = sqlContext.createDataFrame([
... (1.0, Vectors.dense(1.0)),
... (0.0, Vectors.sparse(1, [], []))], ["label", "features"])
>>> ir = IsotonicRegression()
>>> model = ir.fit(df)
>>> test0 = sqlContext.createDataFrame([(Vectors.dense(-1.0),)], ["features"])
>>> model.transform(test0).head().prediction
0.0
>>> model.boundaries
DenseVector([0.0, 1.0])
>>> ir_path = temp_path + "/ir"
>>> ir.save(ir_path)
>>> ir2 = IsotonicRegression.load(ir_path)
>>> ir2.getIsotonic()
True
>>> model_path = temp_path + "/ir_model"
>>> model.save(model_path)
>>> model2 = IsotonicRegressionModel.load(model_path)
>>> model.boundaries == model2.boundaries
True
>>> model.predictions == model2.predictions
True
"""
isotonic = \
Param(Params._dummy(), "isotonic",
"whether the output sequence should be isotonic/increasing (true) or" +
"antitonic/decreasing (false).")
featureIndex = \
Param(Params._dummy(), "featureIndex",
"The index of the feature if featuresCol is a vector column, no effect otherwise.")
@keyword_only
def __init__(self, featuresCol="features", labelCol="label", predictionCol="prediction",
weightCol=None, isotonic=True, featureIndex=0):
"""
__init__(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
weightCol=None, isotonic=True, featureIndex=0):
"""
super(IsotonicRegression, self).__init__()
self._java_obj = self._new_java_obj(
"org.apache.spark.ml.regression.IsotonicRegression", self.uid)
self._setDefault(isotonic=True, featureIndex=0)
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
@keyword_only
def setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction",
weightCol=None, isotonic=True, featureIndex=0):
"""
setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
weightCol=None, isotonic=True, featureIndex=0):
Set the params for IsotonicRegression.
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
def _create_model(self, java_model):
return IsotonicRegressionModel(java_model)
def setIsotonic(self, value):
"""
Sets the value of :py:attr:`isotonic`.
"""
self._paramMap[self.isotonic] = value
return self
def getIsotonic(self):
"""
Gets the value of isotonic or its default value.
"""
return self.getOrDefault(self.isotonic)
def setFeatureIndex(self, value):
"""
Sets the value of :py:attr:`featureIndex`.
"""
self._paramMap[self.featureIndex] = value
return self
def getFeatureIndex(self):
"""
Gets the value of featureIndex or its default value.
"""
return self.getOrDefault(self.featureIndex)
class IsotonicRegressionModel(JavaModel, JavaMLWritable, JavaMLReadable):
"""
.. note:: Experimental
Model fitted by IsotonicRegression.
"""
@property
def boundaries(self):
"""
Model boundaries.
"""
return self._call_java("boundaries")
@property
def predictions(self):
"""
Predictions associated with the boundaries at the same index, monotone because of isotonic
regression.
"""
return self._call_java("predictions")
class TreeEnsembleParams(DecisionTreeParams):
"""
Mixin for Decision Tree-based ensemble algorithms parameters.
"""
subsamplingRate = Param(Params._dummy(), "subsamplingRate", "Fraction of the training data " +
"used for learning each decision tree, in range (0, 1].")
def __init__(self):
super(TreeEnsembleParams, self).__init__()
@since("1.4.0")
def setSubsamplingRate(self, value):
"""
Sets the value of :py:attr:`subsamplingRate`.
"""
self._paramMap[self.subsamplingRate] = value
return self
@since("1.4.0")
def getSubsamplingRate(self):
"""
Gets the value of subsamplingRate or its default value.
"""
return self.getOrDefault(self.subsamplingRate)
class TreeRegressorParams(Params):
"""
Private class to track supported impurity measures.
"""
supportedImpurities = ["variance"]
impurity = Param(Params._dummy(), "impurity",
"Criterion used for information gain calculation (case-insensitive). " +
"Supported options: " +
", ".join(supportedImpurities))
def __init__(self):
super(TreeRegressorParams, self).__init__()
@since("1.4.0")
def setImpurity(self, value):
"""
Sets the value of :py:attr:`impurity`.
"""
self._paramMap[self.impurity] = value
return self
@since("1.4.0")
def getImpurity(self):
"""
Gets the value of impurity or its default value.
"""
return self.getOrDefault(self.impurity)
class RandomForestParams(TreeEnsembleParams):
"""
Private class to track supported random forest parameters.
"""
supportedFeatureSubsetStrategies = ["auto", "all", "onethird", "sqrt", "log2"]
numTrees = Param(Params._dummy(), "numTrees", "Number of trees to train (>= 1).")
featureSubsetStrategy = \
Param(Params._dummy(), "featureSubsetStrategy",
"The number of features to consider for splits at each tree node. Supported " +
"options: " + ", ".join(supportedFeatureSubsetStrategies))
def __init__(self):
super(RandomForestParams, self).__init__()
@since("1.4.0")
def setNumTrees(self, value):
"""
Sets the value of :py:attr:`numTrees`.
"""
self._paramMap[self.numTrees] = value
return self
@since("1.4.0")
def getNumTrees(self):
"""
Gets the value of numTrees or its default value.
"""
return self.getOrDefault(self.numTrees)
@since("1.4.0")
def setFeatureSubsetStrategy(self, value):
"""
Sets the value of :py:attr:`featureSubsetStrategy`.
"""
self._paramMap[self.featureSubsetStrategy] = value
return self
@since("1.4.0")
def getFeatureSubsetStrategy(self):
"""
Gets the value of featureSubsetStrategy or its default value.
"""
return self.getOrDefault(self.featureSubsetStrategy)
class GBTParams(TreeEnsembleParams):
"""
Private class to track supported GBT params.
"""
supportedLossTypes = ["squared", "absolute"]
@inherit_doc
class DecisionTreeRegressor(JavaEstimator, HasFeaturesCol, HasLabelCol, HasPredictionCol,
DecisionTreeParams, TreeRegressorParams, HasCheckpointInterval,
HasSeed):
"""
`http://en.wikipedia.org/wiki/Decision_tree_learning Decision tree`
learning algorithm for regression.
It supports both continuous and categorical features.
>>> from pyspark.mllib.linalg import Vectors
>>> df = sqlContext.createDataFrame([
... (1.0, Vectors.dense(1.0)),
... (0.0, Vectors.sparse(1, [], []))], ["label", "features"])
>>> dt = DecisionTreeRegressor(maxDepth=2)
>>> model = dt.fit(df)
>>> model.depth
1
>>> model.numNodes
3
>>> model.featureImportances
SparseVector(1, {0: 1.0})
>>> test0 = sqlContext.createDataFrame([(Vectors.dense(-1.0),)], ["features"])
>>> model.transform(test0).head().prediction
0.0
>>> test1 = sqlContext.createDataFrame([(Vectors.sparse(1, [0], [1.0]),)], ["features"])
>>> model.transform(test1).head().prediction
1.0
.. versionadded:: 1.4.0
"""
@keyword_only
def __init__(self, featuresCol="features", labelCol="label", predictionCol="prediction",
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, impurity="variance",
seed=None):
"""
__init__(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, \
impurity="variance", seed=None)
"""
super(DecisionTreeRegressor, self).__init__()
self._java_obj = self._new_java_obj(
"org.apache.spark.ml.regression.DecisionTreeRegressor", self.uid)
self._setDefault(maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10,
impurity="variance")
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
@keyword_only
@since("1.4.0")
def setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction",
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10,
impurity="variance", seed=None):
"""
setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, \
impurity="variance", seed=None)
Sets params for the DecisionTreeRegressor.
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
def _create_model(self, java_model):
return DecisionTreeRegressionModel(java_model)
@inherit_doc
class DecisionTreeModel(JavaModel):
"""Abstraction for Decision Tree models.
.. versionadded:: 1.5.0
"""
@property
@since("1.5.0")
def numNodes(self):
"""Return number of nodes of the decision tree."""
return self._call_java("numNodes")
@property
@since("1.5.0")
def depth(self):
"""Return depth of the decision tree."""
return self._call_java("depth")
def __repr__(self):
return self._call_java("toString")
@inherit_doc
class TreeEnsembleModels(JavaModel):
"""Represents a tree ensemble model.
.. versionadded:: 1.5.0
"""
@property
@since("1.5.0")
def treeWeights(self):
"""Return the weights for each tree"""
return list(self._call_java("javaTreeWeights"))
def __repr__(self):
return self._call_java("toString")
@inherit_doc
class DecisionTreeRegressionModel(DecisionTreeModel):
"""
Model fitted by DecisionTreeRegressor.
.. versionadded:: 1.4.0
"""
@property
@since("2.0.0")
def featureImportances(self):
"""
Estimate of the importance of each feature.
This generalizes the idea of "Gini" importance to other losses,
following the explanation of Gini importance from "Random Forests" documentation
by Leo Breiman and Adele Cutler, and following the implementation from scikit-learn.
This feature importance is calculated as follows:
- importance(feature j) = sum (over nodes which split on feature j) of the gain,
where gain is scaled by the number of instances passing through node
- Normalize importances for tree to sum to 1.
Note: Feature importance for single decision trees can have high variance due to
correlated predictor variables. Consider using a :class:`RandomForestRegressor`
to determine feature importance instead.
"""
return self._call_java("featureImportances")
@inherit_doc
class RandomForestRegressor(JavaEstimator, HasFeaturesCol, HasLabelCol, HasPredictionCol, HasSeed,
RandomForestParams, TreeRegressorParams, HasCheckpointInterval):
"""
`http://en.wikipedia.org/wiki/Random_forest Random Forest`
learning algorithm for regression.
It supports both continuous and categorical features.
>>> from numpy import allclose
>>> from pyspark.mllib.linalg import Vectors
>>> df = sqlContext.createDataFrame([
... (1.0, Vectors.dense(1.0)),
... (0.0, Vectors.sparse(1, [], []))], ["label", "features"])
>>> rf = RandomForestRegressor(numTrees=2, maxDepth=2, seed=42)
>>> model = rf.fit(df)
>>> model.featureImportances
SparseVector(1, {0: 1.0})
>>> allclose(model.treeWeights, [1.0, 1.0])
True
>>> test0 = sqlContext.createDataFrame([(Vectors.dense(-1.0),)], ["features"])
>>> model.transform(test0).head().prediction
0.0
>>> test1 = sqlContext.createDataFrame([(Vectors.sparse(1, [0], [1.0]),)], ["features"])
>>> model.transform(test1).head().prediction
0.5
.. versionadded:: 1.4.0
"""
@keyword_only
def __init__(self, featuresCol="features", labelCol="label", predictionCol="prediction",
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10,
impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20,
featureSubsetStrategy="auto"):
"""
__init__(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, \
impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20, \
featureSubsetStrategy="auto")
"""
super(RandomForestRegressor, self).__init__()
self._java_obj = self._new_java_obj(
"org.apache.spark.ml.regression.RandomForestRegressor", self.uid)
self._setDefault(maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10,
impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20,
featureSubsetStrategy="auto")
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
@keyword_only
@since("1.4.0")
def setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction",
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10,
impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20,
featureSubsetStrategy="auto"):
"""
setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, \
impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20, \
featureSubsetStrategy="auto")
Sets params for linear regression.
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
def _create_model(self, java_model):
return RandomForestRegressionModel(java_model)
class RandomForestRegressionModel(TreeEnsembleModels):
"""
Model fitted by RandomForestRegressor.
.. versionadded:: 1.4.0
"""
@property
@since("2.0.0")
def featureImportances(self):
"""
Estimate of the importance of each feature.
This generalizes the idea of "Gini" importance to other losses,
following the explanation of Gini importance from "Random Forests" documentation
by Leo Breiman and Adele Cutler, and following the implementation from scikit-learn.
This feature importance is calculated as follows:
- Average over trees:
- importance(feature j) = sum (over nodes which split on feature j) of the gain,
where gain is scaled by the number of instances passing through node
- Normalize importances for tree to sum to 1.
- Normalize feature importance vector to sum to 1.
"""
return self._call_java("featureImportances")
@inherit_doc
class GBTRegressor(JavaEstimator, HasFeaturesCol, HasLabelCol, HasPredictionCol, HasMaxIter,
GBTParams, HasCheckpointInterval, HasStepSize, HasSeed):
"""
`http://en.wikipedia.org/wiki/Gradient_boosting Gradient-Boosted Trees (GBTs)`
learning algorithm for regression.
It supports both continuous and categorical features.
>>> from numpy import allclose
>>> from pyspark.mllib.linalg import Vectors
>>> df = sqlContext.createDataFrame([
... (1.0, Vectors.dense(1.0)),
... (0.0, Vectors.sparse(1, [], []))], ["label", "features"])
>>> gbt = GBTRegressor(maxIter=5, maxDepth=2)
>>> model = gbt.fit(df)
>>> allclose(model.treeWeights, [1.0, 0.1, 0.1, 0.1, 0.1])
True
>>> test0 = sqlContext.createDataFrame([(Vectors.dense(-1.0),)], ["features"])
>>> model.transform(test0).head().prediction
0.0
>>> test1 = sqlContext.createDataFrame([(Vectors.sparse(1, [0], [1.0]),)], ["features"])
>>> model.transform(test1).head().prediction
1.0
.. versionadded:: 1.4.0
"""
lossType = Param(Params._dummy(), "lossType",
"Loss function which GBT tries to minimize (case-insensitive). " +
"Supported options: " + ", ".join(GBTParams.supportedLossTypes))
@keyword_only
def __init__(self, featuresCol="features", labelCol="label", predictionCol="prediction",
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0,
checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1):
"""
__init__(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \
maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0, \
checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1)
"""
super(GBTRegressor, self).__init__()
self._java_obj = self._new_java_obj("org.apache.spark.ml.regression.GBTRegressor", self.uid)
self._setDefault(maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0,
checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1)
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
@keyword_only
@since("1.4.0")
def setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction",
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0,
maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0,
checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1):
"""
setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \
maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0, \
checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1)
Sets params for Gradient Boosted Tree Regression.
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
def _create_model(self, java_model):
return GBTRegressionModel(java_model)
@since("1.4.0")
def setLossType(self, value):
"""
Sets the value of :py:attr:`lossType`.
"""
self._paramMap[self.lossType] = value
return self
@since("1.4.0")
def getLossType(self):
"""
Gets the value of lossType or its default value.
"""
return self.getOrDefault(self.lossType)
class GBTRegressionModel(TreeEnsembleModels):
"""
Model fitted by GBTRegressor.
.. versionadded:: 1.4.0
"""
@inherit_doc
class AFTSurvivalRegression(JavaEstimator, HasFeaturesCol, HasLabelCol, HasPredictionCol,
HasFitIntercept, HasMaxIter, HasTol, JavaMLWritable, JavaMLReadable):
"""
Accelerated Failure Time (AFT) Model Survival Regression
Fit a parametric AFT survival regression model based on the Weibull distribution
of the survival time.
.. seealso:: `AFT Model <https://en.wikipedia.org/wiki/Accelerated_failure_time_model>`_
>>> from pyspark.mllib.linalg import Vectors
>>> df = sqlContext.createDataFrame([
... (1.0, Vectors.dense(1.0), 1.0),
... (0.0, Vectors.sparse(1, [], []), 0.0)], ["label", "features", "censor"])
>>> aftsr = AFTSurvivalRegression()
>>> model = aftsr.fit(df)
>>> model.predict(Vectors.dense(6.3))
1.0
>>> model.predictQuantiles(Vectors.dense(6.3))
DenseVector([0.0101, 0.0513, 0.1054, 0.2877, 0.6931, 1.3863, 2.3026, 2.9957, 4.6052])
>>> model.transform(df).show()
+-----+---------+------+----------+
|label| features|censor|prediction|
+-----+---------+------+----------+
| 1.0| [1.0]| 1.0| 1.0|
| 0.0|(1,[],[])| 0.0| 1.0|
+-----+---------+------+----------+
...
>>> aftsr_path = temp_path + "/aftsr"
>>> aftsr.save(aftsr_path)
>>> aftsr2 = AFTSurvivalRegression.load(aftsr_path)
>>> aftsr2.getMaxIter()
100
>>> model_path = temp_path + "/aftsr_model"
>>> model.save(model_path)
>>> model2 = AFTSurvivalRegressionModel.load(model_path)
>>> model.coefficients == model2.coefficients
True
>>> model.intercept == model2.intercept
True
>>> model.scale == model2.scale
True
.. versionadded:: 1.6.0
"""
censorCol = Param(Params._dummy(), "censorCol",
"censor column name. The value of this column could be 0 or 1. " +
"If the value is 1, it means the event has occurred i.e. " +
"uncensored; otherwise censored.")
quantileProbabilities = \
Param(Params._dummy(), "quantileProbabilities",
"quantile probabilities array. Values of the quantile probabilities array " +
"should be in the range (0, 1) and the array should be non-empty.")
quantilesCol = Param(Params._dummy(), "quantilesCol",
"quantiles column name. This column will output quantiles of " +
"corresponding quantileProbabilities if it is set.")
@keyword_only
def __init__(self, featuresCol="features", labelCol="label", predictionCol="prediction",
fitIntercept=True, maxIter=100, tol=1E-6, censorCol="censor",
quantileProbabilities=None, quantilesCol=None):
"""
__init__(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
fitIntercept=True, maxIter=100, tol=1E-6, censorCol="censor", \
quantileProbabilities=[0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99], \
quantilesCol=None)
"""
super(AFTSurvivalRegression, self).__init__()
self._java_obj = self._new_java_obj(
"org.apache.spark.ml.regression.AFTSurvivalRegression", self.uid)
self._setDefault(censorCol="censor",
quantileProbabilities=[0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99])
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
@keyword_only
@since("1.6.0")
def setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction",
fitIntercept=True, maxIter=100, tol=1E-6, censorCol="censor",
quantileProbabilities=None, quantilesCol=None):
"""
setParams(self, featuresCol="features", labelCol="label", predictionCol="prediction", \
fitIntercept=True, maxIter=100, tol=1E-6, censorCol="censor", \
quantileProbabilities=[0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99], \
quantilesCol=None):
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
def _create_model(self, java_model):
return AFTSurvivalRegressionModel(java_model)
@since("1.6.0")
def setCensorCol(self, value):
"""
Sets the value of :py:attr:`censorCol`.
"""
self._paramMap[self.censorCol] = value
return self
@since("1.6.0")
def getCensorCol(self):
"""
Gets the value of censorCol or its default value.
"""
return self.getOrDefault(self.censorCol)
@since("1.6.0")
def setQuantileProbabilities(self, value):
"""
Sets the value of :py:attr:`quantileProbabilities`.
"""
self._paramMap[self.quantileProbabilities] = value
return self
@since("1.6.0")
def getQuantileProbabilities(self):
"""
Gets the value of quantileProbabilities or its default value.
"""
return self.getOrDefault(self.quantileProbabilities)
@since("1.6.0")
def setQuantilesCol(self, value):
"""
Sets the value of :py:attr:`quantilesCol`.
"""
self._paramMap[self.quantilesCol] = value
return self
@since("1.6.0")
def getQuantilesCol(self):
"""
Gets the value of quantilesCol or its default value.
"""
return self.getOrDefault(self.quantilesCol)
class AFTSurvivalRegressionModel(JavaModel, JavaMLWritable, JavaMLReadable):
"""
Model fitted by AFTSurvivalRegression.
.. versionadded:: 1.6.0
"""
@property
@since("1.6.0")
def coefficients(self):
"""
Model coefficients.
"""
return self._call_java("coefficients")
@property
@since("1.6.0")
def intercept(self):
"""
Model intercept.
"""
return self._call_java("intercept")
@property
@since("1.6.0")
def scale(self):
"""
Model scale paramter.
"""
return self._call_java("scale")
def predictQuantiles(self, features):
"""
Predicted Quantiles
"""
return self._call_java("predictQuantiles", features)
def predict(self, features):
"""
Predicted value
"""
return self._call_java("predict", features)
if __name__ == "__main__":
import doctest
import pyspark.ml.regression
from pyspark.context import SparkContext
from pyspark.sql import SQLContext
globs = pyspark.ml.regression.__dict__.copy()
# The small batch size here ensures that we see multiple batches,
# even in these small test examples:
sc = SparkContext("local[2]", "ml.regression tests")
sqlContext = SQLContext(sc)
globs['sc'] = sc
globs['sqlContext'] = sqlContext
import tempfile
temp_path = tempfile.mkdtemp()
globs['temp_path'] = temp_path
try:
(failure_count, test_count) = doctest.testmod(globs=globs, optionflags=doctest.ELLIPSIS)
sc.stop()
finally:
from shutil import rmtree
try:
rmtree(temp_path)
except OSError:
pass
if failure_count:
exit(-1)
|