Python* API Reference for Intel® Data Analytics Acceleration Library 2018 Update 2

datasource_featureextraction.py

1 # file: datasource_featureextraction.py
2 #===============================================================================
3 # Copyright 2014-2018 Intel Corporation
4 # All Rights Reserved.
5 #
6 # If this software was obtained under the Intel Simplified Software License,
7 # the following terms apply:
8 #
9 # The source code, information and material ("Material") contained herein is
10 # owned by Intel Corporation or its suppliers or licensors, and title to such
11 # Material remains with Intel Corporation or its suppliers or licensors. The
12 # Material contains proprietary information of Intel or its suppliers and
13 # licensors. The Material is protected by worldwide copyright laws and treaty
14 # provisions. No part of the Material may be used, copied, reproduced,
15 # modified, published, uploaded, posted, transmitted, distributed or disclosed
16 # in any way without Intel's prior express written permission. No license under
17 # any patent, copyright or other intellectual property rights in the Material
18 # is granted to or conferred upon you, either expressly, by implication,
19 # inducement, estoppel or otherwise. Any license under such intellectual
20 # property rights must be express and approved by Intel in writing.
21 #
22 # Unless otherwise agreed by Intel in writing, you may not remove or alter this
23 # notice or any other notice embedded in Materials by Intel or Intel's
24 # suppliers or licensors in any way.
25 #
26 #
27 # If this software was obtained under the Apache License, Version 2.0 (the
28 # "License"), the following terms apply:
29 #
30 # You may not use this file except in compliance with the License. You may
31 # obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0
32 #
33 #
34 # Unless required by applicable law or agreed to in writing, software
35 # distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
36 # WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
37 #
38 # See the License for the specific language governing permissions and
39 # limitations under the License.
40 #===============================================================================
41 
42 #
43 # ! Content:
44 # ! Python example for using of data source feature extraction
45 # !*****************************************************************************
46 
47 #
48 
51 import os
52 import sys
53 
54 from daal.data_management import FileDataSource, DataSourceIface, ColumnFilter, OneHotEncoder
55 
56 utils_folder = os.path.realpath(os.path.abspath(os.path.dirname(os.path.dirname(__file__))))
57 if utils_folder not in sys.path:
58  sys.path.insert(0, utils_folder)
59 from utils import printNumericTable
60 
61 
62 # Input data set parameters
63 datasetFileName = "../data/batch/kmeans_dense.csv"
64 
65 if __name__ == "__main__":
66 
67  # Initialize FileDataSource to retrieve the input data from a .csv file
68  dataSource = FileDataSource(datasetFileName, DataSourceIface.doAllocateNumericTable)
69 
70  # Create data source dictionary from loading of the first .csv file
71  dataSource.createDictionaryFromContext()
72 
73  # Filter in 3 chosen columns from a .csv file
74  validList = [1, 2, 5]
75 
76  colFilter = ColumnFilter()
77  filterList = colFilter.list(validList)
78  dataSource.getFeatureManager().addModifier(filterList)
79 
80  # Consider column with index 1 as categorical and convert it into 3 binary categorical features
81  dataSource.getFeatureManager().addModifier(OneHotEncoder(1, 3))
82 
83  # Load data from .csv file
84  dataSource.loadDataBlock()
85 
86  # Print result
87  table = dataSource.getNumericTable()
88  printNumericTable(table, "Loaded data", 4, 20)

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