# in python: explicitly define the schema, read in CSV data using the schema and a defined timestamp format [and an extra column to be used for partitioning; this part is optional] csvSchema = StructType([ StructField("Timestamp",TimestampType(),True), StructField("Name",StringType(),True), StructField("Value",DoubleType(),True) ])

df = spark.read \ .csv(file_path, header = True, multiLine = True, escape = "\"", schema = csvSchema, timestampFormat = "MM/dd/yyyy HH:mm:ss.SSSZZ" ) \ .withColumn("year", date_format(col("Timestamp"), "yyyy").cast(IntegerType())) \ .withColumn("month", date_format(col("Timestamp"), "MM").cast(IntegerType()))

display(df)