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
|
#
# 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.
#
from abc import ABCMeta, abstractmethod
from pyspark.ml.param import Param, Params
from pyspark.ml.util import keyword_only
from pyspark.mllib.common import inherit_doc
@inherit_doc
class Estimator(Params):
"""
Abstract class for estimators that fit models to data.
"""
__metaclass__ = ABCMeta
@abstractmethod
def _fit(self, dataset):
"""
Fits a model to the input dataset. This is called by the
default implementation of fit.
:param dataset: input dataset, which is an instance of
:py:class:`pyspark.sql.DataFrame`
:returns: fitted model
"""
raise NotImplementedError()
def fit(self, dataset, params={}):
"""
Fits a model to the input dataset with optional parameters.
:param dataset: input dataset, which is an instance of
:py:class:`pyspark.sql.DataFrame`
:param params: an optional param map that overrides embedded
params. If a list/tuple of param maps is given,
this calls fit on each param map and returns a
list of models.
:returns: fitted model(s)
"""
if isinstance(params, (list, tuple)):
return [self.fit(dataset, paramMap) for paramMap in params]
elif isinstance(params, dict):
if params:
return self.copy(params)._fit(dataset)
else:
return self._fit(dataset)
else:
raise ValueError("Params must be either a param map or a list/tuple of param maps, "
"but got %s." % type(params))
@inherit_doc
class Transformer(Params):
"""
Abstract class for transformers that transform one dataset into
another.
"""
__metaclass__ = ABCMeta
@abstractmethod
def _transform(self, dataset):
"""
Transforms the input dataset with optional parameters.
:param dataset: input dataset, which is an instance of
:py:class:`pyspark.sql.DataFrame`
:returns: transformed dataset
"""
raise NotImplementedError()
def transform(self, dataset, params={}):
"""
Transforms the input dataset with optional parameters.
:param dataset: input dataset, which is an instance of
:py:class:`pyspark.sql.DataFrame`
:param params: an optional param map that overrides embedded
params.
:returns: transformed dataset
"""
if isinstance(params, dict):
if params:
return self.copy(params,)._transform(dataset)
else:
return self._transform(dataset)
else:
raise ValueError("Params must be either a param map but got %s." % type(params))
@inherit_doc
class Model(Transformer):
"""
Abstract class for models that are fitted by estimators.
"""
__metaclass__ = ABCMeta
@inherit_doc
class Pipeline(Estimator):
"""
A simple pipeline, which acts as an estimator. A Pipeline consists
of a sequence of stages, each of which is either an
:py:class:`Estimator` or a :py:class:`Transformer`. When
:py:meth:`Pipeline.fit` is called, the stages are executed in
order. If a stage is an :py:class:`Estimator`, its
:py:meth:`Estimator.fit` method will be called on the input
dataset to fit a model. Then the model, which is a transformer,
will be used to transform the dataset as the input to the next
stage. If a stage is a :py:class:`Transformer`, its
:py:meth:`Transformer.transform` method will be called to produce
the dataset for the next stage. The fitted model from a
:py:class:`Pipeline` is an :py:class:`PipelineModel`, which
consists of fitted models and transformers, corresponding to the
pipeline stages. If there are no stages, the pipeline acts as an
identity transformer.
"""
@keyword_only
def __init__(self, stages=[]):
"""
__init__(self, stages=[])
"""
super(Pipeline, self).__init__()
#: Param for pipeline stages.
self.stages = Param(self, "stages", "pipeline stages")
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
def setStages(self, value):
"""
Set pipeline stages.
:param value: a list of transformers or estimators
:return: the pipeline instance
"""
self._paramMap[self.stages] = value
return self
def getStages(self):
"""
Get pipeline stages.
"""
if self.stages in self._paramMap:
return self._paramMap[self.stages]
@keyword_only
def setParams(self, stages=[]):
"""
setParams(self, stages=[])
Sets params for Pipeline.
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
def _fit(self, dataset):
stages = self.getStages()
for stage in stages:
if not (isinstance(stage, Estimator) or isinstance(stage, Transformer)):
raise TypeError(
"Cannot recognize a pipeline stage of type %s." % type(stage))
indexOfLastEstimator = -1
for i, stage in enumerate(stages):
if isinstance(stage, Estimator):
indexOfLastEstimator = i
transformers = []
for i, stage in enumerate(stages):
if i <= indexOfLastEstimator:
if isinstance(stage, Transformer):
transformers.append(stage)
dataset = stage.transform(dataset)
else: # must be an Estimator
model = stage.fit(dataset)
transformers.append(model)
if i < indexOfLastEstimator:
dataset = model.transform(dataset)
else:
transformers.append(stage)
return PipelineModel(transformers)
def copy(self, extra={}):
that = Params.copy(self, extra)
stages = [stage.copy(extra) for stage in that.getStages()]
return that.setStages(stages)
@inherit_doc
class PipelineModel(Model):
"""
Represents a compiled pipeline with transformers and fitted models.
"""
def __init__(self, stages):
super(PipelineModel, self).__init__()
self.stages = stages
def _transform(self, dataset):
for t in self.stages:
dataset = t.transform(dataset)
return dataset
def copy(self, extra={}):
stages = [stage.copy(extra) for stage in self.stages]
return PipelineModel(stages)
|