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

svm_two_class_csr_batch.py

1 # file: svm_two_class_csr_batch.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.algorithms.svm import training, prediction
23 from daal.algorithms import kernel_function, classifier
24 from daal.data_management import DataSourceIface, FileDataSource
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 printNumericTables, createSparseTable
30 
31 # Input data set parameters
32 DATA_PREFIX = os.path.join('..', 'data', 'batch')
33 
34 trainDatasetFileName = os.path.join(DATA_PREFIX, 'svm_two_class_train_csr.csv')
35 trainLabelsFileName = os.path.join(DATA_PREFIX, 'svm_two_class_train_labels.csv')
36 testDatasetFileName = os.path.join(DATA_PREFIX, 'svm_two_class_test_csr.csv')
37 testLabelsFileName = os.path.join(DATA_PREFIX, 'svm_two_class_test_labels.csv')
38 
39 # Parameters for the SVM kernel function
40 kernel = kernel_function.linear.Batch(method=kernel_function.linear.fastCSR)
41 
42 # Model object for the SVM algorithm
43 trainingResult = None
44 predictionResult = None
45 
46 
47 def trainModel():
48  global trainingResult
49 
50  # Initialize FileDataSource to retrieve the input data from a .csv file
51  trainLabelsDataSource = FileDataSource(
52  trainLabelsFileName, DataSourceIface.doAllocateNumericTable,
53  DataSourceIface.doDictionaryFromContext
54  )
55 
56  # Create numeric table for training data
57  trainData = createSparseTable(trainDatasetFileName)
58 
59  # Retrieve the data from the input file
60  trainLabelsDataSource.loadDataBlock()
61 
62  # Create an algorithm object to train the SVM model
63  algorithm = training.Batch()
64 
65  algorithm.parameter.kernel = kernel
66  algorithm.parameter.cacheSize = 40000000
67 
68  # Pass a training data set and dependent values to the algorithm
69  algorithm.input.set(classifier.training.data, trainData)
70  algorithm.input.set(classifier.training.labels, trainLabelsDataSource.getNumericTable())
71 
72  # Build the SVM model
73  trainingResult = algorithm.compute()
74 
75 
76 def testModel():
77  global predictionResult
78 
79  # Create Numeric Tables for testing data
80  testData = createSparseTable(testDatasetFileName)
81 
82  # Create an algorithm object to predict SVM values
83  algorithm = prediction.Batch()
84 
85  algorithm.parameter.kernel = kernel
86 
87  # Pass a testing data set and the trained model to the algorithm
88  algorithm.input.setTable(classifier.prediction.data, testData)
89 
90  algorithm.input.setModel(classifier.prediction.model, trainingResult.get(classifier.training.model))
91 
92  # Predict SVM values
93  algorithm.compute()
94 
95  # Retrieve the algorithm results
96  predictionResult = algorithm.getResult()
97 
98 
99 def printResults():
100 
101  # Initialize FileDataSource<CSVFeatureManager> to retrieve the test data from a .csv file
102  testLabelsDataSource = FileDataSource(
103  testLabelsFileName, DataSourceIface.doAllocateNumericTable,
104  DataSourceIface.doDictionaryFromContext
105  )
106  # Retrieve the data from input file
107  testLabelsDataSource.loadDataBlock()
108  testGroundTruth = testLabelsDataSource.getNumericTable()
109 
110  printNumericTables(
111  testGroundTruth, predictionResult.get(classifier.prediction.prediction),
112  "Ground truth\t", "Classification results",
113  "SVM classification results (first 20 observations):", 20, flt64=False
114  )
115 
116 if __name__ == "__main__":
117 
118  trainModel()
119  testModel()
120  printResults()

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