Set up Spark Context and Spark session

  Load the Data set

  Deal with categorical data and Covert the data to dense vectors(Features and Label)

  Transform the dataset to dataframe

  Identify categorical features, and index them

  Split the data into train and test set

  Build and train the random forest model

  Predict using the test set

  Evaluate the metrics

from pyspark.sql import SparkSession
from pyspark.ml import Pipeline
from pyspark.ml.feature import StringIndexer, OneHotEncoder, VectorAssembler
from pyspark.sql.functions import col
from pyspark.ml.feature import VectorIndexer
from pyspark.ml.regression import RandomForestRegressor
#Set up SparkContext and SparkSession
spark=SparkSession \
.builder \
.appName(“Python spark regression example”)\
.config(“spark.some.config.option”,”some-value”)\
.getOrCreate()

#Load the data set
df=spark.read.format(‘com.databricks.spark.csv’).options(header=’True’,inferschema=’True’).load(“/home/…./servo.csv”)
# Automatically identify categorical features, and index them.
def get_dummy(df,categoricalCols,continuousCols,labelCol):
indexers = [ StringIndexer(inputCol=c, outputCol=”{0}_indexed”.format(c))
for c in categoricalCols ]
# default setting: dropLast=True
encoders = [ OneHotEncoder(inputCol=indexer.getOutputCol(),
outputCol=”{0}_encoded”.format(indexer.getOutputCol()))
for indexer in indexers ]
assembler = VectorAssembler(inputCols=[encoder.getOutputCol() for encoder in encoders]
+ continuousCols, outputCol=”features”)
pipeline = Pipeline(stages=indexers + encoders + [assembler])
model=pipeline.fit(df)
data = model.transform(df)
data = data.withColumn(‘label’,col(labelCol))
return data.select(‘features’,’label’)
catcols =[“Motor”,”Screw”]
catcols
num_cols = [“Pgain”,”Vgain”]
labelCol = ‘Class’
data = get_dummy(df,catcols,num_cols,labelCol)
# Index labels, adding metadata to the label column
labelIndexer = StringIndexer(inputCol=’label’,outputCol=’indexedLabel’).fit(data)
labelIndexer.transform(data).show(5, True)
# Set maxCategories so features with > 6 distinct values are treated as continuous.
featureIndexer =VectorIndexer(inputCol=”features”, \
outputCol=”indexedFeatures”, \
maxCategories=6).fit(data)
# Split the data into training and test sets (20% held out for testing)
(trainingData, testData) = data.randomSplit([0.7, 0.3])
rf = RandomForestRegressor(featuresCol=”indexedFeatures”)
# Chain indexer and tree in a Pipeline
pipeline = Pipeline(stages=[featureIndexer, rf])
# Train model. This also runs the indexer.
model = pipeline.fit(trainingData)
# Make predictions.
predictions = model.transform(testData)
# Select example rows to display.
predictions.select(“prediction”, “label”, “features”).show(5)
# Select (prediction, true label) and compute test error
from pyspark.ml.evaluation import RegressionEvaluator
evaluator = RegressionEvaluator(
labelCol=”label”, predictionCol=”prediction”, metricName=”rmse”)
rmse = evaluator.evaluate(predictions)
print(“RMSE : “,rmse)
treeModel = model.stages[1]
print (treeModel) # summary only