Package quickml.supervised.classifier.randomForest

Examples of quickml.supervised.classifier.randomForest.RandomForest

74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
       
            TreeBuilder treeBuilder = new TreeBuilder()
                .ignoreAttributeAtNodeProbability(0.7);
            RandomForestBuilder randomForestBuilder = new RandomForestBuilder(treeBuilder)
                .numTrees(50);
            RandomForest randomForest = randomForestBuilder.buildPredictiveModel(instances);

            attributes = AttributesMap.newHashMap() ;
            attributes.put("height",62);
            attributes.put("weight", 201);
            attributes.put("gender", "female");
            Serializable classification = randomForest.getClassificationByMaxProb(attributes);
            System.out.println("Assigned class: " + classification);
       
        }
       
    }
View Full Code Here
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
        final long startTime = System.currentTimeMillis();
        final DownsamplingClassifier downsamplingClassifier = wb.buildPredictiveModel(instances);

        TreeBuilderTestUtils.serializeDeserialize(downsamplingClassifier);

        RandomForest randomForest = (RandomForest) downsamplingClassifier.wrappedClassifier;
        final List<Tree> trees = randomForest.trees;
        final int treeSize = trees.size();
        final int firstTreeNodeSize = trees.get(0).node.size();
        org.testng.Assert.assertTrue(treeSize < 400, "Forest size should be less than 400");
        org.testng.Assert.assertTrue((System.currentTimeMillis() - startTime) < 20000, "Building this node should take far less than 20 seconds");

        final List<Instance<AttributesMap>> newInstances = TreeBuilderTestUtils.getIntegerInstances(1000);
        final DownsamplingClassifier downsamplingClassifier1 = wb.buildPredictiveModel(newInstances);
        final RandomForest newRandomForest = (RandomForest) downsamplingClassifier1.wrappedClassifier;
        org.testng.Assert.assertTrue(downsamplingClassifier == downsamplingClassifier1, "Expect same tree to be updated");
        org.testng.Assert.assertEquals(treeSize, newRandomForest.trees.size(), "Expected same number of trees");
        org.testng.Assert.assertEquals(firstTreeNodeSize, newRandomForest.trees.get(0).node.size(), "Expected same nodes");
    }
View Full Code Here
33
34
35
36
37
38
39
40
41
42
43
        final TreeBuilder tb = new TreeBuilder(new SplitDiffScorer()).splitPredictiveModel("gender", whiteList);
        final RandomForestBuilder rfb = new RandomForestBuilder(tb);
        final SplitOnAttributeClassifierBuilder cpmb = new SplitOnAttributeClassifierBuilder("gender", rfb, 10, 0.1, whiteList, 1);
        final long startTime = System.currentTimeMillis();
        final SplitOnAttributeClassifier splitOnAttributeClassifier = cpmb.buildPredictiveModel(instances);
        final RandomForest randomForest = (RandomForest) splitOnAttributeClassifier.getDefaultPM();

        TreeBuilderTestUtils.serializeDeserialize(randomForest);
        final List<Tree> trees = randomForest.trees;
        final int treeSize = trees.size();
        Assert.assertTrue(treeSize < 400, "Forest size should be less than 400");
View Full Code Here
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
        final List<Instance<AttributesMap>> instances = TreeBuilderTestUtils.getInstances(1000);
        final PredictiveModelWithDataBuilder<AttributesMap ,SplitOnAttributeClassifier> wb = getWrappedUpdatablePredictiveModelBuilder();
        wb.splitNodeThreshold(1);
        final long startTime = System.currentTimeMillis();
        final SplitOnAttributeClassifier splitOnAttributeClassifier = wb.buildPredictiveModel(instances);
        final RandomForest randomForest = (RandomForest) splitOnAttributeClassifier.getDefaultPM();

        TreeBuilderTestUtils.serializeDeserialize(randomForest);

        final List<Tree> trees = randomForest.trees;
        final int treeSize = trees.size();
        final int firstTreeNodeSize = trees.get(0).node.size();
        Assert.assertTrue(treeSize < 400, "Forest size should be less than 400");
        Assert.assertTrue((System.currentTimeMillis() - startTime) < 20000,"Building this node should take far less than 20 seconds");

        final List<Instance<AttributesMap>> newInstances = TreeBuilderTestUtils.getInstances(1000);
        final SplitOnAttributeClassifier splitOnAttributeClassifier1 = wb.buildPredictiveModel(newInstances);
        final RandomForest newRandomForest = (RandomForest) splitOnAttributeClassifier1.getDefaultPM();
        Assert.assertTrue(splitOnAttributeClassifier == splitOnAttributeClassifier1, "Expect same tree to be updated");
        Assert.assertEquals(treeSize, newRandomForest.trees.size(), "Expected same number of trees");
        Assert.assertNotEquals(firstTreeNodeSize, newRandomForest.trees.get(0).node.size(), "Expected new nodes");
    }
View Full Code Here
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
    public void simpleBmiTestNoSplit() throws Exception {
        final List<Instance<AttributesMap>> instances = TreeBuilderTestUtils.getInstances(1000);
        final PredictiveModelWithDataBuilder<AttributesMap ,SplitOnAttributeClassifier> wb = getWrappedUpdatablePredictiveModelBuilder();
        final long startTime = System.currentTimeMillis();
        final SplitOnAttributeClassifier splitOnAttributeClassifier = wb.buildPredictiveModel(instances);
        final RandomForest randomForest = (RandomForest) splitOnAttributeClassifier.getDefaultPM();

        TreeBuilderTestUtils.serializeDeserialize(randomForest);

        final List<Tree> trees = randomForest.trees;
        final int treeSize = trees.size();
        final int firstTreeNodeSize = trees.get(0).node.size();
        Assert.assertTrue(treeSize < 400, "Forest size should be less than 400");
        Assert.assertTrue((System.currentTimeMillis() - startTime) < 20000,"Building this node should take far less than 20 seconds");

        final List<Instance<AttributesMap>> newInstances = TreeBuilderTestUtils.getInstances(1000);
        final SplitOnAttributeClassifier splitOnAttributeClassifier1 = wb.buildPredictiveModel(newInstances);
        final RandomForest newRandomForest = (RandomForest) splitOnAttributeClassifier1.getDefaultPM();
        Assert.assertTrue(splitOnAttributeClassifier == splitOnAttributeClassifier1, "Expect same tree to be updated");
        Assert.assertEquals(treeSize, newRandomForest.trees.size(), "Expected same number of trees");
        Assert.assertEquals(firstTreeNodeSize, newRandomForest.trees.get(0).node.size(), "Expected same nodes");
    }
View Full Code Here
TOP

Related Classes of quickml.supervised.classifier.randomForest.RandomForest

  • com.google.common.util.concurrent.AtomicDouble
  • quickml.data.PredictionMap
  • quickml.ReadMeExample
  • quickml.supervised.classifier.decisionTree.tree.Leaf
  • quickml.supervised.classifier.downsampling.DownsamplingClassifierBuilderTest
  • quickml.supervised.classifier.splitOnAttribute.SplitOnAttributeClassifierBuilderTest
  • java.io.Serializable
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.