A growing post which gathers short pieces of code
Let’s suppose spark to be an opened spark session.
1 2 3 4 | # Open a spark session spark = SparkSession \ .builder.config(conf=conf) \ .appName("code-test").getOrCreate() |
How to merge 2 spark DataFrame
- Python:
1 2 3 4 | # Id is the column on which we want to merge 2 DataFrame dataset1 = spark.read.csv("../input/data1.csv", header="true",inferSchema="true") dataset2 = spark.read.csv("../input/data2.csv", header="true",inferSchema="true") dataset= dataset1.join(dataset2, dataset1.Id ==dataset2.Id).drop(dataset1.Id) |
- Scala:
1 2 3 4 | //Id is the column on which we want to merge 2 DataFrame val dataset1 = spark.read.option("header", "true").option("inferSchema", true).csv(inputTestPath + "/input/dataset1.csv") val dataset2 = spark.read.option("header", "true").option("inferSchema", true).csv(inputTestPath + "/input/dataset2.csv") val dataset = datset1.join(dataset1, dataset1("Id") === dataset2("Id")).drop(dataset1.col("Id")) |
How to perform Cross Validation using spark ML in python
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | gbt = GBTClassifier(labelCol="label", featuresCol="features", maxIter=100, seed=77) paramGrid = ParamGridBuilder().addGrid(gbt.maxDepth, [9,10,11]).addGrid(gbt.stepSize, [0.1,0.01,0.3]).build() # Chain indexers and GBT in a Pipeline pipeline = Pipeline(stages=[gbt]) evaluator = MulticlassClassificationEvaluator( labelCol="label", predictionCol="prediction", metricName="accuracy") # Set up 3-fold cross validation crossval = CrossValidator(estimator=pipeline, estimatorParamMaps=paramGrid, evaluator=evaluator, numFolds=3) model = crossval.fit(trainingData) # Fetch best model best_model = model.bestModel |
Leave a Reply
Be the First to Comment!