asddfl created SPARK-60646:
------------------------------
Summary: pandas-on-Spark Series.fillna rejects a Series fill value
from the same DataFrame
Key: SPARK-60646
URL: https://issues.apache.org/jira/browse/SPARK-60646
Project: Spark
Issue Type: Bug
Components: Pandas API on Spark, PySpark
Affects Versions: 4.2.0
Environment: Platform: Ubuntu 24.04 Linux-6.17.0-29-generic x86_64
Python: 3.11.17
pyspark 4.2.0
findspark 2.0.1
pandas 3.0.6
numpy 2.4.6
Reporter: asddfl
Filling a missing floating-point value with the corresponding value from
another column of the same DataFrame fails in the pandas API on Spark. Pandas
accepts the aligned Series fill value and returns 1.0.
{code:python}
import pandas as pd
import numpy as np
from pyspark.sql import SparkSession
import pyspark.pandas as ps
t0 = pd.DataFrame({'c0': [np.nan], 'c1': [1.0]})
print("Pandas:", flush=True)
result = t0['c0'].fillna(t0['c1'])
print(result, flush=True)
spark = (
SparkSession.builder
.master("local[2]")
.config("spark.sql.ansi.enabled", "false")
.config("spark.sql.shuffle.partitions", "2")
.getOrCreate()
)
t0 = ps.DataFrame({'c0': [np.nan], 'c1': [1.0]})
print("PySpark Pandas:", flush=True)
result = t0['c0'].fillna(t0['c1'])
print(result, flush=True)
spark.stop()
{code}
{code:bash}
Pandas:
0 1.0
Name: c0, dtype: float64
PySpark Pandas:
Traceback (most recent call last):
File "/test.py", line 25, in <module>
result = t0['c0'].fillna(t0['c1'])
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/lib/python3.11/site-packages/pyspark/pandas/series.py",
line 2217, in fillna
return self._fillna_with_method(
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/lib/python3.11/site-packages/pyspark/pandas/series.py",
line 2233, in _fillna_with_method
psser = self._fillna(value=value, method=method, axis=axis, limit=limit)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/lib/python3.11/site-packages/pyspark/pandas/series.py",
line 2277, in _fillna
raise TypeError("Unsupported type %s" % type(value).__name__)
TypeError: Unsupported type Series
{code}
*Expected behavior*
Fill c0's missing value from c1 and return a float Series named c0 containing
1.0, matching Pandas.
*Actual behavior*
t0['c0'].fillna(t0['c1']) raises TypeError: Unsupported type Series.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
