For creating Spark DataFrame, we can pass RDD data Or Pandas Dataframe.
I was doing experimentation with both,
Spark: Standalone Mode using pyspark.sql