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

assoc_rules_apriori_batch.py

1 # file: assoc_rules_apriori_batch.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 
44 
45 import os
46 import sys
47 
48 from daal.algorithms import association_rules
49 from daal.data_management import FileDataSource, DataSourceIface
50 
51 utils_folder = os.path.realpath(os.path.abspath(os.path.dirname(os.path.dirname(__file__))))
52 if utils_folder not in sys.path:
53  sys.path.insert(0, utils_folder)
54 from utils import printAprioriItemsets, printAprioriRules
55 
56 # Input data set parameters
57 datasetFileName = os.path.join('..','data','batch','apriori.csv')
58 
59 # Apriori algorithm parameters
60 minSupport = 0.001
61 minConfidence = 0.7
62 
63 # Initialize FileDataSource_CSVFeatureManager to retrieve the input data from a .csv file
64 dataSource = FileDataSource(
65  datasetFileName, DataSourceIface.doAllocateNumericTable, DataSourceIface.doDictionaryFromContext
66 )
67 
68 # Retrieve the data from the input file
69 dataSource.loadDataBlock()
70 
71 # Create an algorithm to mine association rules using the Apriori method
72 alg = association_rules.Batch()
73 alg.input.set(association_rules.data, dataSource.getNumericTable())
74 alg.parameter.minSupport = minSupport
75 alg.parameter.minConfidence = minConfidence
76 
77 # Find large item sets and construct association rules
78 res = alg.compute()
79 
80 # Get computed results of the Apriori algorithm
81 nt1 = res.get(association_rules.largeItemsets)
82 nt2 = res.get(association_rules.largeItemsetsSupport)
83 
84 nt3 = res.get(association_rules.antecedentItemsets)
85 nt4 = res.get(association_rules.consequentItemsets)
86 nt5 = res.get(association_rules.confidence)
87 
88 printAprioriItemsets(nt1, nt2)
89 printAprioriRules(nt3, nt4, nt5)

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