import pandas as pd

#create a pandas df from some data
df = pd.read_csv("some/data/test_v2.csv")

#if SparkContext is not already initialized, run the next line
#sc = SparkContext(conf=conf)

#create a SQLContext
sqlc = SQLContext(sc)

#pass pandas dataframe into createDataFrame() function of SQLContext
sdf = sqlc.createDataFrame(df)