From de64b65f7cf2ac58c1abc310ba547637fdbb8557 Mon Sep 17 00:00:00 2001 From: Yanbo Liang Date: Mon, 30 Nov 2015 15:01:08 -0800 Subject: [SPARK-11975][ML] Remove duplicate mllib example (DT/RF/GBT in Java/Python) Remove duplicate mllib example (DT/RF/GBT in Java/Python). Since we have tutorial code for DT/RF/GBT classification/regression in Scala/Java/Python and example applications for DT/RF/GBT in Scala, so we mark these as duplicated and remove them. mengxr Author: Yanbo Liang Closes #9954 from yanboliang/SPARK-11975. --- .../src/main/python/mllib/decision_tree_runner.py | 144 --------------------- .../main/python/mllib/gradient_boosted_trees.py | 77 ----------- .../src/main/python/mllib/random_forest_example.py | 90 ------------- 3 files changed, 311 deletions(-) delete mode 100755 examples/src/main/python/mllib/decision_tree_runner.py delete mode 100644 examples/src/main/python/mllib/gradient_boosted_trees.py delete mode 100755 examples/src/main/python/mllib/random_forest_example.py (limited to 'examples/src/main/python/mllib') diff --git a/examples/src/main/python/mllib/decision_tree_runner.py b/examples/src/main/python/mllib/decision_tree_runner.py deleted file mode 100755 index 513ed8fd51..0000000000 --- a/examples/src/main/python/mllib/decision_tree_runner.py +++ /dev/null @@ -1,144 +0,0 @@ -# -# Licensed to the Apache Software Foundation (ASF) under one or more -# contributor license agreements. See the NOTICE file distributed with -# this work for additional information regarding copyright ownership. -# The ASF licenses this file to You under the Apache License, Version 2.0 -# (the "License"); you may not use this file except in compliance with -# the License. You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# - -""" -Decision tree classification and regression using MLlib. - -This example requires NumPy (http://www.numpy.org/). -""" -from __future__ import print_function - -import numpy -import os -import sys - -from operator import add - -from pyspark import SparkContext -from pyspark.mllib.regression import LabeledPoint -from pyspark.mllib.tree import DecisionTree -from pyspark.mllib.util import MLUtils - - -def getAccuracy(dtModel, data): - """ - Return accuracy of DecisionTreeModel on the given RDD[LabeledPoint]. - """ - seqOp = (lambda acc, x: acc + (x[0] == x[1])) - predictions = dtModel.predict(data.map(lambda x: x.features)) - truth = data.map(lambda p: p.label) - trainCorrect = predictions.zip(truth).aggregate(0, seqOp, add) - if data.count() == 0: - return 0 - return trainCorrect / (0.0 + data.count()) - - -def getMSE(dtModel, data): - """ - Return mean squared error (MSE) of DecisionTreeModel on the given - RDD[LabeledPoint]. - """ - seqOp = (lambda acc, x: acc + numpy.square(x[0] - x[1])) - predictions = dtModel.predict(data.map(lambda x: x.features)) - truth = data.map(lambda p: p.label) - trainMSE = predictions.zip(truth).aggregate(0, seqOp, add) - if data.count() == 0: - return 0 - return trainMSE / (0.0 + data.count()) - - -def reindexClassLabels(data): - """ - Re-index class labels in a dataset to the range {0,...,numClasses-1}. - If all labels in that range already appear at least once, - then the returned RDD is the same one (without a mapping). - Note: If a label simply does not appear in the data, - the index will not include it. - Be aware of this when reindexing subsampled data. - :param data: RDD of LabeledPoint where labels are integer values - denoting labels for a classification problem. - :return: Pair (reindexedData, origToNewLabels) where - reindexedData is an RDD of LabeledPoint with labels in - the range {0,...,numClasses-1}, and - origToNewLabels is a dictionary mapping original labels - to new labels. - """ - # classCounts: class --> # examples in class - classCounts = data.map(lambda x: x.label).countByValue() - numExamples = sum(classCounts.values()) - sortedClasses = sorted(classCounts.keys()) - numClasses = len(classCounts) - # origToNewLabels: class --> index in 0,...,numClasses-1 - if (numClasses < 2): - print("Dataset for classification should have at least 2 classes." - " The given dataset had only %d classes." % numClasses, file=sys.stderr) - exit(1) - origToNewLabels = dict([(sortedClasses[i], i) for i in range(0, numClasses)]) - - print("numClasses = %d" % numClasses) - print("Per-class example fractions, counts:") - print("Class\tFrac\tCount") - for c in sortedClasses: - frac = classCounts[c] / (numExamples + 0.0) - print("%g\t%g\t%d" % (c, frac, classCounts[c])) - - if (sortedClasses[0] == 0 and sortedClasses[-1] == numClasses - 1): - return (data, origToNewLabels) - else: - reindexedData = \ - data.map(lambda x: LabeledPoint(origToNewLabels[x.label], x.features)) - return (reindexedData, origToNewLabels) - - -def usage(): - print("Usage: decision_tree_runner [libsvm format data filepath]", file=sys.stderr) - exit(1) - - -if __name__ == "__main__": - if len(sys.argv) > 2: - usage() - sc = SparkContext(appName="PythonDT") - - # Load data. - dataPath = 'data/mllib/sample_libsvm_data.txt' - if len(sys.argv) == 2: - dataPath = sys.argv[1] - if not os.path.isfile(dataPath): - sc.stop() - usage() - points = MLUtils.loadLibSVMFile(sc, dataPath) - - # Re-index class labels if needed. - (reindexedData, origToNewLabels) = reindexClassLabels(points) - numClasses = len(origToNewLabels) - - # Train a classifier. - categoricalFeaturesInfo = {} # no categorical features - model = DecisionTree.trainClassifier(reindexedData, numClasses=numClasses, - categoricalFeaturesInfo=categoricalFeaturesInfo) - # Print learned tree and stats. - print("Trained DecisionTree for classification:") - print(" Model numNodes: %d" % model.numNodes()) - print(" Model depth: %d" % model.depth()) - print(" Training accuracy: %g" % getAccuracy(model, reindexedData)) - if model.numNodes() < 20: - print(model.toDebugString()) - else: - print(model) - - sc.stop() diff --git a/examples/src/main/python/mllib/gradient_boosted_trees.py b/examples/src/main/python/mllib/gradient_boosted_trees.py deleted file mode 100644 index 781bd61c9d..0000000000 --- a/examples/src/main/python/mllib/gradient_boosted_trees.py +++ /dev/null @@ -1,77 +0,0 @@ -# -# Licensed to the Apache Software Foundation (ASF) under one or more -# contributor license agreements. See the NOTICE file distributed with -# this work for additional information regarding copyright ownership. -# The ASF licenses this file to You under the Apache License, Version 2.0 -# (the "License"); you may not use this file except in compliance with -# the License. You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# - -""" -Gradient boosted Trees classification and regression using MLlib. -""" -from __future__ import print_function - -import sys - -from pyspark.context import SparkContext -from pyspark.mllib.tree import GradientBoostedTrees -from pyspark.mllib.util import MLUtils - - -def testClassification(trainingData, testData): - # Train a GradientBoostedTrees model. - # Empty categoricalFeaturesInfo indicates all features are continuous. - model = GradientBoostedTrees.trainClassifier(trainingData, categoricalFeaturesInfo={}, - numIterations=30, maxDepth=4) - # Evaluate model on test instances and compute test error - predictions = model.predict(testData.map(lambda x: x.features)) - labelsAndPredictions = testData.map(lambda lp: lp.label).zip(predictions) - testErr = labelsAndPredictions.filter(lambda v_p: v_p[0] != v_p[1]).count() \ - / float(testData.count()) - print('Test Error = ' + str(testErr)) - print('Learned classification ensemble model:') - print(model.toDebugString()) - - -def testRegression(trainingData, testData): - # Train a GradientBoostedTrees model. - # Empty categoricalFeaturesInfo indicates all features are continuous. - model = GradientBoostedTrees.trainRegressor(trainingData, categoricalFeaturesInfo={}, - numIterations=30, maxDepth=4) - # Evaluate model on test instances and compute test error - predictions = model.predict(testData.map(lambda x: x.features)) - labelsAndPredictions = testData.map(lambda lp: lp.label).zip(predictions) - testMSE = labelsAndPredictions.map(lambda vp: (vp[0] - vp[1]) * (vp[0] - vp[1])).sum() \ - / float(testData.count()) - print('Test Mean Squared Error = ' + str(testMSE)) - print('Learned regression ensemble model:') - print(model.toDebugString()) - - -if __name__ == "__main__": - if len(sys.argv) > 1: - print("Usage: gradient_boosted_trees", file=sys.stderr) - exit(1) - sc = SparkContext(appName="PythonGradientBoostedTrees") - - # Load and parse the data file into an RDD of LabeledPoint. - data = MLUtils.loadLibSVMFile(sc, 'data/mllib/sample_libsvm_data.txt') - # Split the data into training and test sets (30% held out for testing) - (trainingData, testData) = data.randomSplit([0.7, 0.3]) - - print('\nRunning example of classification using GradientBoostedTrees\n') - testClassification(trainingData, testData) - - print('\nRunning example of regression using GradientBoostedTrees\n') - testRegression(trainingData, testData) - - sc.stop() diff --git a/examples/src/main/python/mllib/random_forest_example.py b/examples/src/main/python/mllib/random_forest_example.py deleted file mode 100755 index 4cfdad868c..0000000000 --- a/examples/src/main/python/mllib/random_forest_example.py +++ /dev/null @@ -1,90 +0,0 @@ -# -# Licensed to the Apache Software Foundation (ASF) under one or more -# contributor license agreements. See the NOTICE file distributed with -# this work for additional information regarding copyright ownership. -# The ASF licenses this file to You under the Apache License, Version 2.0 -# (the "License"); you may not use this file except in compliance with -# the License. You may obtain a copy of the License at -# -# http://www.apache.org/licenses/LICENSE-2.0 -# -# Unless required by applicable law or agreed to in writing, software -# distributed under the License is distributed on an "AS IS" BASIS, -# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. -# See the License for the specific language governing permissions and -# limitations under the License. -# - -""" -Random Forest classification and regression using MLlib. - -Note: This example illustrates binary classification. - For information on multiclass classification, please refer to the decision_tree_runner.py - example. -""" -from __future__ import print_function - -import sys - -from pyspark.context import SparkContext -from pyspark.mllib.tree import RandomForest -from pyspark.mllib.util import MLUtils - - -def testClassification(trainingData, testData): - # Train a RandomForest model. - # Empty categoricalFeaturesInfo indicates all features are continuous. - # Note: Use larger numTrees in practice. - # Setting featureSubsetStrategy="auto" lets the algorithm choose. - model = RandomForest.trainClassifier(trainingData, numClasses=2, - categoricalFeaturesInfo={}, - numTrees=3, featureSubsetStrategy="auto", - impurity='gini', maxDepth=4, maxBins=32) - - # Evaluate model on test instances and compute test error - predictions = model.predict(testData.map(lambda x: x.features)) - labelsAndPredictions = testData.map(lambda lp: lp.label).zip(predictions) - testErr = labelsAndPredictions.filter(lambda v_p: v_p[0] != v_p[1]).count()\ - / float(testData.count()) - print('Test Error = ' + str(testErr)) - print('Learned classification forest model:') - print(model.toDebugString()) - - -def testRegression(trainingData, testData): - # Train a RandomForest model. - # Empty categoricalFeaturesInfo indicates all features are continuous. - # Note: Use larger numTrees in practice. - # Setting featureSubsetStrategy="auto" lets the algorithm choose. - model = RandomForest.trainRegressor(trainingData, categoricalFeaturesInfo={}, - numTrees=3, featureSubsetStrategy="auto", - impurity='variance', maxDepth=4, maxBins=32) - - # Evaluate model on test instances and compute test error - predictions = model.predict(testData.map(lambda x: x.features)) - labelsAndPredictions = testData.map(lambda lp: lp.label).zip(predictions) - testMSE = labelsAndPredictions.map(lambda v_p1: (v_p1[0] - v_p1[1]) * (v_p1[0] - v_p1[1]))\ - .sum() / float(testData.count()) - print('Test Mean Squared Error = ' + str(testMSE)) - print('Learned regression forest model:') - print(model.toDebugString()) - - -if __name__ == "__main__": - if len(sys.argv) > 1: - print("Usage: random_forest_example", file=sys.stderr) - exit(1) - sc = SparkContext(appName="PythonRandomForestExample") - - # Load and parse the data file into an RDD of LabeledPoint. - data = MLUtils.loadLibSVMFile(sc, 'data/mllib/sample_libsvm_data.txt') - # Split the data into training and test sets (30% held out for testing) - (trainingData, testData) = data.randomSplit([0.7, 0.3]) - - print('\nRunning example of classification using RandomForest\n') - testClassification(trainingData, testData) - - print('\nRunning example of regression using RandomForest\n') - testRegression(trainingData, testData) - - sc.stop() -- cgit v1.2.3