Created
January 24, 2018 22:30
-
-
Save icexelloss/88f6c6fdaf04aac39d68d74cd0942c07 to your computer and use it in GitHub Desktop.
Groupby apply group key benchmark
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from pyspark.sql.functions import pandas_udf | |
df = spark.range(0, 10 * 1000 * 1000).withColumn('id', (col('id') / 1000).cast(IntegerType())) | |
df.cache() | |
df.count() | |
from pyspark.sql.functions import PandasUDFType | |
@pandas_udf(df.schema, PandasUDFType.GROUP_MAP) | |
# Input/output are both a pandas.DataFrame | |
def foo_udf(pdf): | |
return pdf | |
# 8.61 s ± 333 ms per loop (mean ± std. dev. of 7 runs, 1 loop each) | |
%timeit df.groupby('id').apply(foo_udf).agg(count(col('id'))).show() | |
# 8.51 s ± 477 ms per loop (mean ± std. dev. of 7 runs, 1 loop each) | |
%timeit df.groupby(df.id + 1).apply(foo_udf).agg(count(col('id'))).show() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment