You are viewing documentation about an older version (1.42.0). View latest version

snowflake.snowpark.functions.vectorized

snowflake.snowpark.functions.vectorized(input: type, max_batch_size: Optional[int] = None) → Callable[source]

Marks a function or UDTF method as vectorized for pandas input. This decorator is a no-op for local invocation. When combined with udf(), this will make the function behave as a vectorized UDF using pandas_udf().

Parameters:

  • input – The type of the input to the function. Must be either pandas.Series or pandas.DataFrame.

  • max_batch_size – The maximum batch size to use for the function. If not provided, the default batch size will be used.

Returns:

A decorator that marks the function as vectorized.

Example:

>>> import pandas as pd
>>> from snowflake.snowpark.functions import udf, vectorized
>>> from snowflake.snowpark.types import IntegerType
>>> @udf(return_type=IntegerType(), input_types=[IntegerType(), IntegerType()])
... @vectorized(input=pd.DataFrame)
... def add_one_to_inputs(df):
...     return df[0] + df[1] + 1
>>> df = session.create_dataframe([[1, 2], [3, 4]], schema=["a", "b"])
>>> df.select(add_one_to_inputs("a", "b").alias("result")).collect()
[Row(RESULT=4), Row(RESULT=8)]

See also