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

svm_multi_class_dense_batch.py

1 # file: svm_multi_class_dense_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 ## <a name="DAAL-EXAMPLE-PY-SVM_MULTI_CLASS_DENSE_BATCH"></a>
17 ## \example svm_multi_class_dense_batch.py
18 
19 import os
20 import sys
21 
22 from daal.algorithms.svm import training, prediction
23 from daal.algorithms import classifier, kernel_function, multi_class_classifier
24 from daal.data_management import (
25  FileDataSource, DataSourceIface, HomogenNumericTable, MergedNumericTable, NumericTableIface
26 )
27 
28 utils_folder = os.path.realpath(os.path.abspath(os.path.dirname(os.path.dirname(__file__))))
29 if utils_folder not in sys.path:
30  sys.path.insert(0, utils_folder)
31 from utils import printNumericTables
32 
33 DAAL_PREFIX = os.path.join('..', 'data')
34 
35 # Input data set parameters
36 trainDatasetFileName = os.path.join(DAAL_PREFIX, 'batch', 'svm_multi_class_train_dense.csv')
37 
38 testDatasetFileName = os.path.join(DAAL_PREFIX, 'batch', 'svm_multi_class_test_dense.csv')
39 
40 nFeatures = 20
41 nClasses = 5
42 
43 trainingBatch = training.Batch()
44 predictionBatch = prediction.Batch()
45 
46 trainingResult = None
47 predictionResult = None
48 kernelBatch = kernel_function.linear.Batch()
49 testGroundTruth = None
50 
51 
52 def trainModel():
53  global trainingResult
54 
55  # Initialize FileDataSource<CSVFeatureManager> to retrieve the input data from a .csv file
56  trainDataSource = FileDataSource(
57  trainDatasetFileName,
58  DataSourceIface.notAllocateNumericTable,
59  DataSourceIface.doDictionaryFromContext
60  )
61 
62  # Create Numeric Tables for training data and labels
63  trainData = HomogenNumericTable(nFeatures, 0, NumericTableIface.doNotAllocate)
64  trainGroundTruth = HomogenNumericTable(1, 0, NumericTableIface.doNotAllocate)
65  mergedData = MergedNumericTable(trainData, trainGroundTruth)
66 
67  # Retrieve the data from the input file
68  trainDataSource.loadDataBlock(mergedData)
69 
70  # Create an algorithm object to train the multi-class SVM model
71  algorithm = multi_class_classifier.training.Batch(nClasses)
72 
73  algorithm.parameter.training = trainingBatch
74  algorithm.parameter.prediction = predictionBatch
75 
76  # Pass a training data set and dependent values to the algorithm
77  algorithm.input.set(classifier.training.data, trainData)
78  algorithm.input.set(classifier.training.labels, trainGroundTruth)
79 
80  # Build the multi-class SVM model
81  # and retrieve Result class from multi_class_classifier.training
82  trainingResult = algorithm.compute()
83 
84 
85 def testModel():
86  global predictionResult, testGroundTruth
87 
88  # Initialize FileDataSource<CSVFeatureManager> to retrieve the test data from a .csv file
89  testDataSource = FileDataSource(
90  testDatasetFileName,
91  DataSourceIface.doAllocateNumericTable,
92  DataSourceIface.doDictionaryFromContext
93  )
94 
95  # Create Numeric Tables for testing data and labels
96  testData = HomogenNumericTable(nFeatures, 0, NumericTableIface.doNotAllocate)
97  testGroundTruth = HomogenNumericTable(1, 0, NumericTableIface.doNotAllocate)
98  mergedData = MergedNumericTable(testData, testGroundTruth)
99 
100  # Retrieve the data from input file
101  testDataSource.loadDataBlock(mergedData)
102 
103  # Create an algorithm object to predict multi-class SVM values
104  algorithm = multi_class_classifier.prediction.Batch(nClasses)
105 
106  algorithm.parameter.training = trainingBatch
107  algorithm.parameter.prediction = predictionBatch
108 
109  # Pass a testing data set and the trained model to the algorithm
110  algorithm.input.setTable(classifier.prediction.data, testData)
111  algorithm.input.setModel(classifier.prediction.model,
112  trainingResult.get(classifier.training.model))
113 
114  # Predict multi-class SVM values
115  # and retrieve Result class from classifier.prediction
116  predictionResult = algorithm.compute() # Retrieve the algorithm results
117 
118 
119 def printResults():
120 
121  printNumericTables(
122  testGroundTruth,
123  predictionResult.get(classifier.prediction.prediction),
124  "Ground truth", "Classification results",
125  "Multi-class SVM classification sample program results (first 20 observations):", 20, flt64=False
126  )
127 
128 if __name__ == "__main__":
129 
130  trainingBatch.parameter.cacheSize = 100000000
131  trainingBatch.parameter.kernel = kernelBatch
132  predictionBatch.parameter.kernel = kernelBatch
133 
134  trainModel()
135  testModel()
136  printResults()

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