我正在嘗試在 pyspark (Spark 2.4.5) 中應用非常簡單的 Pandas UDF，但它對我不起作用。例子：pyspark --master local[4] --conf "spark.pyspark.python=/opt/anaconda/envs/bd9/bin/python3" --conf "spark.pyspark.driver.python=/opt/anaconda/envs/bd9/bin/python3" >>> my_df = spark.createDataFrame(...     [...         (1, 0),...         (2, 1),...         (3, 1),...     ],...     ["uid", "partition_id"]... )from pyspark.sql.types import StructType, StructField, StringTypeschema = StructType([StructField("uid", StringType())])from pyspark.sql.functions import pandas_udf, PandasUDFTypeimport pandas>>> @pandas_udf(schema, PandasUDFType.GROUPED_MAP)... def apply_model(sample_df):...     print(sample_df)...     return pandas.DataFrame({"uid": sample_df["uid"]})...>>> result = my_df.groupBy("partition_id").apply(apply_model)>>> result.show()   uid  partition_id0    1             0[Stage 13:==================================================>    (92 + 4) / 100]   uid  partition_id0    2             11    3             1+---+|uid|+---+|  ||  ||  |+---+不知何故 uid 沒有反映在結果中。你能說我在這里缺少什么嗎？