CODEWITHSUNDEEP
pythonmachine-learningpysparkapache-spark-ml
Project_Prkt
Project_Prkt

Reputation: 91

PySpark - How to get precision / recall / ROC from TrainValidationSplit?

My current approach to evaluate different parameters for LinearSVC and get the best one:

tokenizer = Tokenizer(inputCol="Text", outputCol="words")
wordsData = tokenizer.transform(df)

hashingTF = HashingTF(inputCol="words", outputCol="rawFeatures")
featurizedData = hashingTF.transform(wordsData)

idf = IDF(inputCol="rawFeatures", outputCol="features")
idfModel = idf.fit(featurizedData)

LSVC = LinearSVC()

rescaledData = idfModel.transform(featurizedData)

paramGrid = ParamGridBuilder()\
                            .addGrid(LSVC.maxIter, [1])\
                            .addGrid(LSVC.regParam, [0.001, 10.0])\
                            .build()

crossval = TrainValidationSplit(estimator=LSVC,
                                estimatorParamMaps=paramGrid,
                                evaluator=MulticlassClassificationEvaluator(metricName="weightedPrecision"),
                                testRatio=0.01)

cvModel = crossval.fit(rescaledData.select("KA", "features").selectExpr("KA as label", "features as features"))

bestModel = cvModel.bestModel

Now I would like to get the basic parameters of ML (like precision, recall etc.), how do I get those?

Upvotes: 1

Views: 9214

Answers (1)

Bhaskar
Bhaskar

Reputation: 343

You can try this

from pyspark.mllib.evaluation import MulticlassMetrics

# Instantiate metrics object
metrics = MulticlassMetrics(predictionAndLabels)

# Overall statistics
precision = metrics.precision()
recall = metrics.recall()
f1Score = metrics.fMeasure()
print("Summary Stats")
print("Precision = %s" % precision)
print("Recall = %s" % recall)
print("F1 Score = %s" % f1Score)

You can check this link for further info

https://spark.apache.org/docs/2.1.0/mllib-evaluation-metrics.html

Upvotes: 1

Related Questions