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

cov_csr_distr.py

1 # file: cov_csr_distr.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 
44 
45 import os
46 import sys
47 
48 from daal import step1Local, step2Master
49 from daal.algorithms import covariance
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 nBlocks = 4
60 
61 datasetFileNames = [
62  os.path.join(DAAL_PREFIX, 'distributed', 'covcormoments_csr_1.csv'),
63  os.path.join(DAAL_PREFIX, 'distributed', 'covcormoments_csr_2.csv'),
64  os.path.join(DAAL_PREFIX, 'distributed', 'covcormoments_csr_3.csv'),
65  os.path.join(DAAL_PREFIX, 'distributed', 'covcormoments_csr_4.csv')
66 ]
67 
68 partialResult = [0] * nBlocks
69 result = None
70 
71 
72 def computestep1Local(block):
73  global partialResult
74 
75  dataTable = createSparseTable(datasetFileNames[block])
76 
77  # Create algorithm objects to compute a variance-covariance matrix in the distributed processing mode using the default method
78  algorithm = covariance.Distributed(step1Local, method=covariance.fastCSR)
79 
80  # Set input objects for the algorithm
81  algorithm.input.set(covariance.data, dataTable)
82 
83  # Compute partial estimates on local nodes
84  partialResult[block] = algorithm.compute() # Get the computed partial estimates
85 
86 
87 def computeOnMasterNode():
88  global result
89 
90  # Create algorithm objects to compute a variance-covariance matrix in the distributed processing mode using the default method
91  algorithm = covariance.Distributed(step2Master, method=covariance.fastCSR)
92 
93  # Set input objects for the algorithm
94  for i in range(nBlocks):
95  algorithm.input.add(covariance.partialResults, partialResult[i])
96 
97  # Compute a partial estimate on the master node from the partial estimates on local nodes
98  algorithm.compute()
99 
100  # Finalize the result in the distributed processing mode and get the computed variance-covariance matrix
101  result = algorithm.finalizeCompute()
102 
103 if __name__ == "__main__":
104 
105  for i in range(nBlocks):
106  computestep1Local(i)
107 
108  computeOnMasterNode()
109 
110  printNumericTable(result.get(covariance.covariance), "Covariance matrix (upper left square 10*10) :", 10, 10)
111  printNumericTable(result.get(covariance.mean), "Mean vector:", 1, 10)

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