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
|
#
# 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 pyspark import since
from pyspark.ml.util import *
from pyspark.ml.wrapper import JavaEstimator, JavaModel
from pyspark.ml.param.shared import *
from pyspark.mllib.common import inherit_doc
__all__ = ['BisectingKMeans', 'BisectingKMeansModel',
'KMeans', 'KMeansModel']
class KMeansModel(JavaModel, JavaMLWritable, JavaMLReadable):
"""
Model fitted by KMeans.
.. versionadded:: 1.5.0
"""
@since("1.5.0")
def clusterCenters(self):
"""Get the cluster centers, represented as a list of NumPy arrays."""
return [c.toArray() for c in self._call_java("clusterCenters")]
@since("2.0.0")
def computeCost(self, dataset):
"""
Return the K-means cost (sum of squared distances of points to their nearest center)
for this model on the given data.
"""
return self._call_java("computeCost", dataset)
@inherit_doc
class KMeans(JavaEstimator, HasFeaturesCol, HasPredictionCol, HasMaxIter, HasTol, HasSeed,
JavaMLWritable, JavaMLReadable):
"""
K-means clustering with support for multiple parallel runs and a k-means++ like initialization
mode (the k-means|| algorithm by Bahmani et al). When multiple concurrent runs are requested,
they are executed together with joint passes over the data for efficiency.
>>> from pyspark.mllib.linalg import Vectors
>>> data = [(Vectors.dense([0.0, 0.0]),), (Vectors.dense([1.0, 1.0]),),
... (Vectors.dense([9.0, 8.0]),), (Vectors.dense([8.0, 9.0]),)]
>>> df = sqlContext.createDataFrame(data, ["features"])
>>> kmeans = KMeans(k=2, seed=1)
>>> model = kmeans.fit(df)
>>> centers = model.clusterCenters()
>>> len(centers)
2
>>> model.computeCost(df)
2.000...
>>> transformed = model.transform(df).select("features", "prediction")
>>> rows = transformed.collect()
>>> rows[0].prediction == rows[1].prediction
True
>>> rows[2].prediction == rows[3].prediction
True
>>> kmeans_path = temp_path + "/kmeans"
>>> kmeans.save(kmeans_path)
>>> kmeans2 = KMeans.load(kmeans_path)
>>> kmeans2.getK()
2
>>> model_path = temp_path + "/kmeans_model"
>>> model.save(model_path)
>>> model2 = KMeansModel.load(model_path)
>>> model.clusterCenters()[0] == model2.clusterCenters()[0]
array([ True, True], dtype=bool)
>>> model.clusterCenters()[1] == model2.clusterCenters()[1]
array([ True, True], dtype=bool)
.. versionadded:: 1.5.0
"""
k = Param(Params._dummy(), "k", "number of clusters to create",
typeConverter=TypeConverters.toInt)
initMode = Param(Params._dummy(), "initMode",
"the initialization algorithm. This can be either \"random\" to " +
"choose random points as initial cluster centers, or \"k-means||\" " +
"to use a parallel variant of k-means++", TypeConverters.toString)
initSteps = Param(Params._dummy(), "initSteps", "steps for k-means initialization mode",
typeConverter=TypeConverters.toInt)
@keyword_only
def __init__(self, featuresCol="features", predictionCol="prediction", k=2,
initMode="k-means||", initSteps=5, tol=1e-4, maxIter=20, seed=None):
"""
__init__(self, featuresCol="features", predictionCol="prediction", k=2, \
initMode="k-means||", initSteps=5, tol=1e-4, maxIter=20, seed=None)
"""
super(KMeans, self).__init__()
self._java_obj = self._new_java_obj("org.apache.spark.ml.clustering.KMeans", self.uid)
self._setDefault(k=2, initMode="k-means||", initSteps=5, tol=1e-4, maxIter=20)
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
def _create_model(self, java_model):
return KMeansModel(java_model)
@keyword_only
@since("1.5.0")
def setParams(self, featuresCol="features", predictionCol="prediction", k=2,
initMode="k-means||", initSteps=5, tol=1e-4, maxIter=20, seed=None):
"""
setParams(self, featuresCol="features", predictionCol="prediction", k=2, \
initMode="k-means||", initSteps=5, tol=1e-4, maxIter=20, seed=None)
Sets params for KMeans.
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
@since("1.5.0")
def setK(self, value):
"""
Sets the value of :py:attr:`k`.
"""
self._set(k=value)
return self
@since("1.5.0")
def getK(self):
"""
Gets the value of `k`
"""
return self.getOrDefault(self.k)
@since("1.5.0")
def setInitMode(self, value):
"""
Sets the value of :py:attr:`initMode`.
"""
self._set(initMode=value)
return self
@since("1.5.0")
def getInitMode(self):
"""
Gets the value of `initMode`
"""
return self.getOrDefault(self.initMode)
@since("1.5.0")
def setInitSteps(self, value):
"""
Sets the value of :py:attr:`initSteps`.
"""
self._set(initSteps=value)
return self
@since("1.5.0")
def getInitSteps(self):
"""
Gets the value of `initSteps`
"""
return self.getOrDefault(self.initSteps)
class BisectingKMeansModel(JavaModel, JavaMLWritable, JavaMLReadable):
"""
.. note:: Experimental
Model fitted by BisectingKMeans.
.. versionadded:: 2.0.0
"""
@since("2.0.0")
def clusterCenters(self):
"""Get the cluster centers, represented as a list of NumPy arrays."""
return [c.toArray() for c in self._call_java("clusterCenters")]
@since("2.0.0")
def computeCost(self, dataset):
"""
Computes the sum of squared distances between the input points
and their corresponding cluster centers.
"""
return self._call_java("computeCost", dataset)
@inherit_doc
class BisectingKMeans(JavaEstimator, HasFeaturesCol, HasPredictionCol, HasMaxIter, HasSeed,
JavaMLWritable, JavaMLReadable):
"""
.. note:: Experimental
A bisecting k-means algorithm based on the paper "A comparison of document clustering
techniques" by Steinbach, Karypis, and Kumar, with modification to fit Spark.
The algorithm starts from a single cluster that contains all points.
Iteratively it finds divisible clusters on the bottom level and bisects each of them using
k-means, until there are `k` leaf clusters in total or no leaf clusters are divisible.
The bisecting steps of clusters on the same level are grouped together to increase parallelism.
If bisecting all divisible clusters on the bottom level would result more than `k` leaf
clusters, larger clusters get higher priority.
>>> from pyspark.mllib.linalg import Vectors
>>> data = [(Vectors.dense([0.0, 0.0]),), (Vectors.dense([1.0, 1.0]),),
... (Vectors.dense([9.0, 8.0]),), (Vectors.dense([8.0, 9.0]),)]
>>> df = sqlContext.createDataFrame(data, ["features"])
>>> bkm = BisectingKMeans(k=2, minDivisibleClusterSize=1.0)
>>> model = bkm.fit(df)
>>> centers = model.clusterCenters()
>>> len(centers)
2
>>> model.computeCost(df)
2.000...
>>> transformed = model.transform(df).select("features", "prediction")
>>> rows = transformed.collect()
>>> rows[0].prediction == rows[1].prediction
True
>>> rows[2].prediction == rows[3].prediction
True
>>> bkm_path = temp_path + "/bkm"
>>> bkm.save(bkm_path)
>>> bkm2 = BisectingKMeans.load(bkm_path)
>>> bkm2.getK()
2
>>> model_path = temp_path + "/bkm_model"
>>> model.save(model_path)
>>> model2 = BisectingKMeansModel.load(model_path)
>>> model.clusterCenters()[0] == model2.clusterCenters()[0]
array([ True, True], dtype=bool)
>>> model.clusterCenters()[1] == model2.clusterCenters()[1]
array([ True, True], dtype=bool)
.. versionadded:: 2.0.0
"""
k = Param(Params._dummy(), "k", "number of clusters to create",
typeConverter=TypeConverters.toInt)
minDivisibleClusterSize = Param(Params._dummy(), "minDivisibleClusterSize",
"the minimum number of points (if >= 1.0) " +
"or the minimum proportion",
typeConverter=TypeConverters.toFloat)
@keyword_only
def __init__(self, featuresCol="features", predictionCol="prediction", maxIter=20,
seed=None, k=4, minDivisibleClusterSize=1.0):
"""
__init__(self, featuresCol="features", predictionCol="prediction", maxIter=20, \
seed=None, k=4, minDivisibleClusterSize=1.0)
"""
super(BisectingKMeans, self).__init__()
self._java_obj = self._new_java_obj("org.apache.spark.ml.clustering.BisectingKMeans",
self.uid)
self._setDefault(maxIter=20, k=4, minDivisibleClusterSize=1.0)
kwargs = self.__init__._input_kwargs
self.setParams(**kwargs)
@keyword_only
@since("2.0.0")
def setParams(self, featuresCol="features", predictionCol="prediction", maxIter=20,
seed=None, k=4, minDivisibleClusterSize=1.0):
"""
setParams(self, featuresCol="features", predictionCol="prediction", maxIter=20, \
seed=None, k=4, minDivisibleClusterSize=1.0)
Sets params for BisectingKMeans.
"""
kwargs = self.setParams._input_kwargs
return self._set(**kwargs)
@since("2.0.0")
def setK(self, value):
"""
Sets the value of :py:attr:`k`.
"""
self._set(k=value)
return self
@since("2.0.0")
def getK(self):
"""
Gets the value of `k` or its default value.
"""
return self.getOrDefault(self.k)
@since("2.0.0")
def setMinDivisibleClusterSize(self, value):
"""
Sets the value of :py:attr:`minDivisibleClusterSize`.
"""
self._set(minDivisibleClusterSize=value)
return self
@since("2.0.0")
def getMinDivisibleClusterSize(self):
"""
Gets the value of `minDivisibleClusterSize` or its default value.
"""
return self.getOrDefault(self.minDivisibleClusterSize)
def _create_model(self, java_model):
return BisectingKMeansModel(java_model)
if __name__ == "__main__":
import doctest
import pyspark.ml.clustering
from pyspark.context import SparkContext
from pyspark.sql import SQLContext
globs = pyspark.ml.clustering.__dict__.copy()
# The small batch size here ensures that we see multiple batches,
# even in these small test examples:
sc = SparkContext("local[2]", "ml.clustering 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)
|