Python* API Reference for Intel® Data Analytics Acceleration Library 2018 Update 2

kmeans_csr_batch.py

1 # file: kmeans_csr_batch.py
2 #===============================================================================
3 # Copyright 2014-2018 Intel Corporation
4 # All Rights Reserved.
5 #
6 # If this software was obtained under the Intel Simplified Software License,
7 # the following terms apply:
8 #
9 # The source code, information and material ("Material") contained herein is
10 # owned by Intel Corporation or its suppliers or licensors, and title to such
11 # Material remains with Intel Corporation or its suppliers or licensors. The
12 # Material contains proprietary information of Intel or its suppliers and
13 # licensors. The Material is protected by worldwide copyright laws and treaty
14 # provisions. No part of the Material may be used, copied, reproduced,
15 # modified, published, uploaded, posted, transmitted, distributed or disclosed
16 # in any way without Intel's prior express written permission. No license under
17 # any patent, copyright or other intellectual property rights in the Material
18 # is granted to or conferred upon you, either expressly, by implication,
19 # inducement, estoppel or otherwise. Any license under such intellectual
20 # property rights must be express and approved by Intel in writing.
21 #
22 # Unless otherwise agreed by Intel in writing, you may not remove or alter this
23 # notice or any other notice embedded in Materials by Intel or Intel's
24 # suppliers or licensors in any way.
25 #
26 #
27 # If this software was obtained under the Apache License, Version 2.0 (the
28 # "License"), the following terms apply:
29 #
30 # You may not use this file except in compliance with the License. You may
31 # obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0
32 #
33 #
34 # Unless required by applicable law or agreed to in writing, software
35 # distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
36 # WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
37 #
38 # See the License for the specific language governing permissions and
39 # limitations under the License.
40 #===============================================================================
41 
42 
43 
44 
45 import os
46 import sys
47 
48 import daal.algorithms.kmeans.init
49 from daal.algorithms import kmeans
50 
51 utils_folder = os.path.realpath(os.path.abspath(os.path.dirname(os.path.dirname(__file__))))
52 if utils_folder not in sys.path:
53  sys.path.insert(0, utils_folder)
54 from utils import printNumericTable, createSparseTable
55 
56 DAAL_PREFIX = os.path.join('..', 'data')
57 
58 # Input data set parameters
59 datasetFileName = os.path.join(DAAL_PREFIX, 'batch', 'kmeans_csr.csv')
60 
61 # K-Means algorithm parameters
62 nClusters = 20
63 nIterations = 5
64 
65 if __name__ == "__main__":
66 
67  # Retrieve the data from the input file
68  dataTable = createSparseTable(datasetFileName)
69 
70  # Get initial clusters for the K-Means algorithm
71  init = kmeans.init.Batch(nClusters, method=kmeans.init.randomDense)
72 
73  init.input.set(kmeans.init.data, dataTable)
74  res = init.compute()
75 
76  centroids = res.get(kmeans.init.centroids)
77 
78  # Create an algorithm object for the K-Means algorithm
79  algorithm = kmeans.Batch(nClusters, nIterations, method=kmeans.lloydCSR)
80 
81  algorithm.input.set(kmeans.data, dataTable)
82  algorithm.input.set(kmeans.inputCentroids, centroids)
83 
84  res = algorithm.compute()
85 
86  # Print the clusterization results
87  printNumericTable(res.get(kmeans.assignments), "First 10 cluster assignments:", 10)
88  printNumericTable(res.get(kmeans.centroids), "First 10 dimensions of centroids:", 20, 10)
89  printNumericTable(res.get(kmeans.objectiveFunction), "Objective function value:")

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