Python* API Reference for Intel® Data Analytics Acceleration Library 2019 Update 5

kmeans_csr_distr.py

Deprecation Notice: With the introduction of daal4py, a package that supersedes PyDAAL, Intel is deprecating PyDAAL and will discontinue support starting with Intel® DAAL 2021 and Intel® Distribution for Python 2021. Until then Intel will continue to provide compatible pyDAAL pip and conda packages for newer releases of Intel DAAL and make it available in open source. However, Intel will not add the new features of Intel DAAL to pyDAAL. Intel recommends developers switch to and use daal4py.

Note: To find daal4py examples, refer to daal4py documentation or browse github repository.

1 # file: kmeans_csr_distr.py
2 #===============================================================================
3 # Copyright 2014-2019 Intel Corporation.
4 #
5 # This software and the related documents are Intel copyrighted materials, and
6 # your use of them is governed by the express license under which they were
7 # provided to you (License). Unless the License provides otherwise, you may not
8 # use, modify, copy, publish, distribute, disclose or transmit this software or
9 # the related documents without Intel's prior written permission.
10 #
11 # This software and the related documents are provided as is, with no express
12 # or implied warranties, other than those that are expressly stated in the
13 # License.
14 #===============================================================================
15 
16 ## <a name="DAAL-EXAMPLE-PY-KMEANS_CSR_DISTRIBUTED"></a>
17 ## \example kmeans_csr_distr.py
18 
19 import os
20 import sys
21 
22 import daal.algorithms.kmeans as kmeans
23 import daal.algorithms.kmeans.init as init
24 from daal import step1Local, step2Master
25 
26 utils_folder = os.path.realpath(os.path.abspath(os.path.dirname(os.path.dirname(__file__))))
27 if utils_folder not in sys.path:
28  sys.path.insert(0, utils_folder)
29 from utils import printNumericTable, createSparseTable
30 
31 DAAL_PREFIX = os.path.join('..', 'data')
32 
33 # K-Means algorithm parameters
34 nClusters = 20
35 nIterations = 5
36 nBlocks = 4
37 nVectorsInBlock = 8000
38 
39 dataFileNames = [
40  os.path.join(DAAL_PREFIX, 'batch', 'kmeans_csr.csv'),
41  os.path.join(DAAL_PREFIX, 'batch', 'kmeans_csr.csv'),
42  os.path.join(DAAL_PREFIX, 'batch', 'kmeans_csr.csv'),
43  os.path.join(DAAL_PREFIX, 'batch', 'kmeans_csr.csv')
44 ]
45 
46 dataTable = [0] * nBlocks
47 
48 if __name__ == "__main__":
49 
50  masterAlgorithm = kmeans.Distributed(step2Master, nClusters, method=kmeans.lloydCSR, )
51 
52  centroids = None
53  assignments = [0] * nBlocks
54 
55  masterInitAlgorithm = init.Distributed(step2Master, nClusters, method=init.randomDense)
56 
57  for i in range(nBlocks):
58 
59  # Read dataFileNames and create a numeric table to store the input data
60  dataTable[i] = createSparseTable(dataFileNames[i])
61 
62  # Create an algorithm object for the K-Means algorithm
63  localInit = init.Distributed(step1Local, nClusters, nBlocks * nVectorsInBlock, i * nVectorsInBlock, method=init.randomDense)
64 
65  localInit.input.set(init.data, dataTable[i])
66  # compute and add input for next
67  masterInitAlgorithm.input.add(init.partialResults, localInit.compute())
68 
69  masterInitAlgorithm.compute()
70  res = masterInitAlgorithm.finalizeCompute()
71  centroids = res.get(init.centroids)
72 
73  for it in range(nIterations):
74  for i in range(nBlocks):
75  # Create an algorithm object for the K-Means algorithm
76  localAlgorithm = kmeans.Distributed(step1Local, nClusters, it == nIterations, method=kmeans.lloydCSR)
77 
78  # Set the input data to the algorithm
79  localAlgorithm.input.set(kmeans.data, dataTable[i])
80  localAlgorithm.input.set(kmeans.inputCentroids, centroids)
81 
82  pres = localAlgorithm.compute()
83 
84  masterAlgorithm.input.add(kmeans.partialResults, pres)
85 
86  masterAlgorithm.compute()
87  result = masterAlgorithm.finalizeCompute()
88 
89  centroids = result.get(kmeans.centroids)
90  objectiveFunction = result.get(kmeans.objectiveFunction)
91 
92  for i in range(nBlocks):
93  # Create an algorithm object for the K-Means algorithm
94  localAlgorithm = kmeans.Batch(nClusters, 0, method=kmeans.lloydCSR)
95 
96  # Set the input data to the algorithm
97  localAlgorithm.input.set(kmeans.data, dataTable[i])
98  localAlgorithm.input.set(kmeans.inputCentroids, centroids)
99 
100  res = localAlgorithm.compute()
101 
102  assignments[i] = res.get(kmeans.assignments)
103 
104  # Print the clusterization results
105  printNumericTable(assignments[0], "First 10 cluster assignments from 1st node:", 10)
106  printNumericTable(centroids, "First 10 dimensions of centroids:", 20, 10)
107  printNumericTable(objectiveFunction, "Objective function value:")

For more complete information about compiler optimizations, see our Optimization Notice.