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

cov_csr_distr.py

1 # file: cov_csr_distr.py
2 #===============================================================================
3 # Copyright 2014-2018 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 
17 
18 
19 import os
20 import sys
21 
22 from daal import step1Local, step2Master
23 from daal.algorithms import covariance
24 
25 utils_folder = os.path.realpath(os.path.abspath(os.path.dirname(os.path.dirname(__file__))))
26 if utils_folder not in sys.path:
27  sys.path.insert(0, utils_folder)
28 from utils import printNumericTable, createSparseTable
29 
30 DAAL_PREFIX = os.path.join('..', 'data')
31 
32 # Input data set parameters
33 nBlocks = 4
34 
35 datasetFileNames = [
36  os.path.join(DAAL_PREFIX, 'distributed', 'covcormoments_csr_1.csv'),
37  os.path.join(DAAL_PREFIX, 'distributed', 'covcormoments_csr_2.csv'),
38  os.path.join(DAAL_PREFIX, 'distributed', 'covcormoments_csr_3.csv'),
39  os.path.join(DAAL_PREFIX, 'distributed', 'covcormoments_csr_4.csv')
40 ]
41 
42 partialResult = [0] * nBlocks
43 result = None
44 
45 
46 def computestep1Local(block):
47  global partialResult
48 
49  dataTable = createSparseTable(datasetFileNames[block])
50 
51  # Create algorithm objects to compute a variance-covariance matrix in the distributed processing mode using the default method
52  algorithm = covariance.Distributed(step1Local, method=covariance.fastCSR)
53 
54  # Set input objects for the algorithm
55  algorithm.input.set(covariance.data, dataTable)
56 
57  # Compute partial estimates on local nodes
58  partialResult[block] = algorithm.compute() # Get the computed partial estimates
59 
60 
61 def computeOnMasterNode():
62  global result
63 
64  # Create algorithm objects to compute a variance-covariance matrix in the distributed processing mode using the default method
65  algorithm = covariance.Distributed(step2Master, method=covariance.fastCSR)
66 
67  # Set input objects for the algorithm
68  for i in range(nBlocks):
69  algorithm.input.add(covariance.partialResults, partialResult[i])
70 
71  # Compute a partial estimate on the master node from the partial estimates on local nodes
72  algorithm.compute()
73 
74  # Finalize the result in the distributed processing mode and get the computed variance-covariance matrix
75  result = algorithm.finalizeCompute()
76 
77 if __name__ == "__main__":
78 
79  for i in range(nBlocks):
80  computestep1Local(i)
81 
82  computeOnMasterNode()
83 
84  printNumericTable(result.get(covariance.covariance), "Covariance matrix (upper left square 10*10) :", 10, 10)
85  printNumericTable(result.get(covariance.mean), "Mean vector:", 1, 10)

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