python/mllib/linear_regression_with_sgd_example.py

0001 #
0002 # Licensed to the Apache Software Foundation (ASF) under one or more
0003 # contributor license agreements.  See the NOTICE file distributed with
0004 # this work for additional information regarding copyright ownership.
0005 # The ASF licenses this file to You under the Apache License, Version 2.0
0006 # (the "License"); you may not use this file except in compliance with
0007 # the License.  You may obtain a copy of the License at
0008 #
0009 #    http://www.apache.org/licenses/LICENSE-2.0
0010 #
0011 # Unless required by applicable law or agreed to in writing, software
0012 # distributed under the License is distributed on an "AS IS" BASIS,
0013 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
0014 # See the License for the specific language governing permissions and
0015 # limitations under the License.
0016 #
0017
0018 """
0019 Linear Regression With SGD Example.
0020 """
0021 from __future__ import print_function
0022
0023 from pyspark import SparkContext
0024 # $example on$
0025 from pyspark.mllib.regression import LabeledPoint, LinearRegressionWithSGD, LinearRegressionModel
0026 # $example off$
0027
0028 if __name__ == "__main__":
0029
0030     sc = SparkContext(appName="PythonLinearRegressionWithSGDExample")
0031
0032     # $example on$
0033     # Load and parse the data
0034     def parsePoint(line):
0035         values = [float(x) for x in line.replace(',', ' ').split(' ')]
0036         return LabeledPoint(values[0], values[1:])
0037
0038     data = sc.textFile("data/mllib/ridge-data/lpsa.data")
0039     parsedData = data.map(parsePoint)
0040
0041     # Build the model
0042     model = LinearRegressionWithSGD.train(parsedData, iterations=100, step=0.00000001)
0043
0044     # Evaluate the model on training data
0045     valuesAndPreds = parsedData.map(lambda p: (p.label, model.predict(p.features)))
0046     MSE = valuesAndPreds \
0047         .map(lambda vp: (vp[0] - vp[1])**2) \
0048         .reduce(lambda x, y: x + y) / valuesAndPreds.count()
0049     print("Mean Squared Error = " + str(MSE))
0050
0051     # Save and load model
0052     model.save(sc, "target/tmp/pythonLinearRegressionWithSGDModel")
0053     sameModel = LinearRegressionModel.load(sc, "target/tmp/pythonLinearRegressionWithSGDModel")
0054     # $example off$