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

log_reg_binary_dense_batch.py

1 # file: log_reg_binary_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 
17 
18 
19 import os
20 import sys
21 
22 from daal.algorithms import logistic_regression
23 from daal.algorithms.logistic_regression import prediction, training
24 from daal.algorithms import classifier
25 from daal.data_management import (
26  FileDataSource, DataSourceIface, NumericTableIface, HomogenNumericTable,
27  MergedNumericTable, features
28 )
29 
30 utils_folder = os.path.realpath(os.path.abspath(os.path.dirname(os.path.dirname(__file__))))
31 if utils_folder not in sys.path:
32  sys.path.insert(0, utils_folder)
33 from utils import printNumericTable, printNumericTables
34 
35 DAAL_PREFIX = os.path.join('..', 'data')
36 
37 # Input data set parameters
38 trainDatasetFileName = os.path.join(DAAL_PREFIX, 'batch', 'binary_cls_train.csv')
39 testDatasetFileName = os.path.join(DAAL_PREFIX, 'batch', 'binary_cls_test.csv')
40 
41 nFeatures = 20
42 nClasses = 2
43 
44 # Model object for the logistic regression algorithm
45 model = None
46 predictionResult = None
47 testGroundTruth = None
48 
49 def trainModel():
50  global model
51 
52  # Initialize FileDataSource<CSVFeatureManager> to retrieve the input data from a .csv file
53  trainDataSource = FileDataSource(
54  trainDatasetFileName,
55  DataSourceIface.notAllocateNumericTable,
56  DataSourceIface.doDictionaryFromContext
57  )
58 
59  # Create Numeric Tables for training data and labels
60  trainData = HomogenNumericTable(nFeatures, 0, NumericTableIface.notAllocate)
61  trainGroundTruth = HomogenNumericTable(1, 0, NumericTableIface.notAllocate)
62  mergedData = MergedNumericTable(trainData, trainGroundTruth)
63 
64  # Retrieve the data from the input file
65  trainDataSource.loadDataBlock(mergedData)
66 
67  # Create an algorithm object to train the logistic regression model
68  algorithm = training.Batch(nClasses)
69 
70  # Pass the training data set and dependent values to the algorithm
71  algorithm.input.set(classifier.training.data, trainData)
72  algorithm.input.set(classifier.training.labels, trainGroundTruth)
73 
74  # Train the logistic regression model and retrieve the results of the training algorithm
75  trainingResult = algorithm.compute()
76  model = trainingResult.get(classifier.training.model)
77  printNumericTable(model.getBeta(), "Logistic Regression coefficients:")
78 
79 def testModel():
80  global testGroundTruth, predictionResult
81 
82  # Initialize FileDataSource<CSVFeatureManager> to retrieve the test data from a .csv file
83  testDataSource = FileDataSource(
84  testDatasetFileName,
85  DataSourceIface.notAllocateNumericTable,
86  DataSourceIface.doDictionaryFromContext
87  )
88 
89  # Create Numeric Tables for testing data and labels
90  testData = HomogenNumericTable(nFeatures, 0, NumericTableIface.notAllocate)
91  testGroundTruth = HomogenNumericTable(1, 0, NumericTableIface.notAllocate)
92  mergedData = MergedNumericTable(testData, testGroundTruth)
93 
94  # Retrieve the data from input file
95  testDataSource.loadDataBlock(mergedData)
96 
97  # Create algorithm objects for logistic regression prediction with the default method
98  algorithm = prediction.Batch(nClasses)
99 
100  # Pass the testing data set and trained model to the algorithm
101  algorithm.input.setTable(classifier.prediction.data, testData)
102  algorithm.input.setModel(classifier.prediction.model, model)
103 
104  # Compute prediction results and retrieve algorithm results
105  # (Result class from classifier.prediction)
106  predictionResult = algorithm.compute()
107 
108 
109 def printResults():
110 
111  printNumericTable(predictionResult.get(classifier.prediction.prediction),"Logistic regression prediction results (first 10 rows):",10)
112  printNumericTable(testGroundTruth,"Ground truth (first 10 rows):",10)
113 
114 if __name__ == "__main__":
115 
116  trainModel()
117  testModel()
118  printResults()

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