Python* API Reference for Intel® Data Analytics Acceleration Library 2018 Update 3

gbt_reg_dense_batch.py

1 # file: gbt_reg_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 gbt
23 from daal.algorithms.gbt.regression import prediction, training
24 from daal.data_management import (
25  FileDataSource, DataSourceIface, NumericTableIface,
26  HomogenNumericTable, MergedNumericTable, data_feature_utils
27 )
28 
29 utils_folder = os.path.realpath(os.path.abspath(os.path.dirname(os.path.dirname(__file__))))
30 if utils_folder not in sys.path:
31  sys.path.insert(0, utils_folder)
32 from utils import printNumericTable
33 
34 DAAL_PREFIX = os.path.join('..', 'data')
35 
36 # Input data set parameters
37 trainDatasetFileName = os.path.join(DAAL_PREFIX, 'batch', 'df_regression_train.csv')
38 testDatasetFileName = os.path.join(DAAL_PREFIX, 'batch', 'df_regression_test.csv')
39 
40 nFeatures = 13
41 
42 # Gradient boosted trees parameters
43 maxIterations = 40
44 
45 # Model object for the gradient boosted trees regression algorithm
46 model = None
47 predictionResult = None
48 testGroundTruth = None
49 
50 
51 def trainModel():
52  global model
53 
54  # Initialize FileDataSource<CSVFeatureManager> to retrieve the input data from a .csv file
55  trainDataSource = FileDataSource(
56  trainDatasetFileName,
57  DataSourceIface.notAllocateNumericTable,
58  DataSourceIface.doDictionaryFromContext
59  )
60 
61  # Create Numeric Tables for training data and labels
62  trainData = HomogenNumericTable(nFeatures, 0, NumericTableIface.notAllocate)
63  trainGroundTruth = HomogenNumericTable(1, 0, NumericTableIface.notAllocate)
64  mergedData = MergedNumericTable(trainData, trainGroundTruth)
65 
66  # Retrieve the data from the input file
67  trainDataSource.loadDataBlock(mergedData)
68 
69  # Get the dictionary and update it with additional information about data
70  dict = trainData.getDictionary()
71 
72  # Add a feature type to the dictionary
73  dict[3].featureType = data_feature_utils.DAAL_CATEGORICAL
74 
75  # Create an algorithm object to train the gradient boosted trees regression model
76  algorithm = training.Batch()
77  algorithm.parameter().maxIterations = maxIterations
78 
79  # Pass the training data set and dependent values to the algorithm
80  algorithm.input.set(training.data, trainData)
81  algorithm.input.set(training.dependentVariable, trainGroundTruth)
82 
83  # Train the gradient boosted trees regression model and retrieve the results of the training algorithm
84  trainingResult = algorithm.compute()
85  model = trainingResult.get(training.model)
86 
87 def testModel():
88  global testGroundTruth, predictionResult
89 
90  # Initialize FileDataSource<CSVFeatureManager> to retrieve the test data from a .csv file
91  testDataSource = FileDataSource(
92  testDatasetFileName,
93  DataSourceIface.notAllocateNumericTable,
94  DataSourceIface.doDictionaryFromContext
95  )
96 
97  # Create Numeric Tables for testing data and labels
98  testData = HomogenNumericTable(nFeatures, 0, NumericTableIface.notAllocate)
99  testGroundTruth = HomogenNumericTable(1, 0, NumericTableIface.notAllocate)
100  mergedData = MergedNumericTable(testData, testGroundTruth)
101 
102  # Retrieve the data from input file
103  testDataSource.loadDataBlock(mergedData)
104 
105  # Get the dictionary and update it with additional information about data
106  dict = testData.getDictionary()
107 
108  # Add a feature type to the dictionary
109  dict[3].featureType = data_feature_utils.DAAL_CATEGORICAL
110 
111  # Create algorithm objects for the gradient boosted trees regression prediction with the default method
112  algorithm = prediction.Batch()
113 
114  # Pass the testing data set and trained model to the algorithm
115  algorithm.input.setTable(prediction.data, testData)
116  algorithm.input.set(prediction.model, model)
117 
118  # Compute prediction results and retrieve algorithm results
119  predictionResult = algorithm.compute()
120 
121 
122 def printResults():
123 
124  printNumericTable(
125  predictionResult.get(prediction.prediction),
126  "Gradient boosted trees prediction results (first 10 rows):", 10
127  )
128  printNumericTable(
129  testGroundTruth,
130  "Ground truth (first 10 rows):", 10
131  )
132 
133 if __name__ == "__main__":
134 
135  trainModel()
136  testModel()
137  printResults()

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