ML функционирует как pyspark UDF - PullRequest
Новая
       7

ML функционирует как pyspark UDF

1 голос
/ 16 июня 2020

Я немного новичок в pyspark и python. Я пытаюсь запустить функцию ML как pyspark UDF.

Вот пример: 

from pyspark.sql.functions import col, pandas_udf
from pyspark.sql.types import StringType

df = spark.createDataFrame(['Bob has a dog. He loves him'], StringType())

def parse(text):
    import spacy
    import neuralcoref
    nlp = spacy.load('en_core_web_sm')
    # Let's try before using the conversion dictionary:
    neuralcoref.add_to_pipe(nlp)
    doc = nlp(text)
    return doc._.coref_resolved

 pd_udf = pandas_udf(parse, returnType=StringType())

 df.select(pd_udf(col("value"))).show()

Получение этой ошибки: 

org.apache.spark.api.python.PythonException: Traceback (most recent call last):
  File "/home/user/tools/spark-2.4.3-bin-hadoop2.7/python/lib/pyspark.zip/pyspark/worker.py", line 377, in main
    process()
  File "/home/user/tools/spark-2.4.3-bin-hadoop2.7/python/lib/pyspark.zip/pyspark/worker.py", line 372, in process
    serializer.dump_stream(func(split_index, iterator), outfile)
  File "/home/user/tools/spark-2.4.3-bin-hadoop2.7/python/lib/pyspark.zip/pyspark/serializers.py", line 286, in dump_stream
    for series in iterator:
  File "<string>", line 1, in <lambda>
  File "/home/user/tools/spark-2.4.3-bin-hadoop2.7/python/lib/pyspark.zip/pyspark/worker.py", line 101, in <lambda>
    return lambda *a: (verify_result_length(*a), arrow_return_type)
  File "/home/user/tools/spark-2.4.3-bin-hadoop2.7/python/lib/pyspark.zip/pyspark/worker.py", line 92, in verify_result_length
    result = f(*a)
  File "/home/user/tools/spark-2.4.3-bin-hadoop2.7/python/lib/pyspark.zip/pyspark/util.py", line 99, in wrapper
    return f(*args, **kwargs)
  File "<stdin>", line 7, in parse
  File "/home/user/anaconda3/lib/python3.7/site-packages/spacy/language.py", line 377, in __call__
    doc = self.make_doc(text)
  File "/home/user/anaconda3/lib/python3.7/site-packages/spacy/language.py", line 401, in make_doc
    return self.tokenizer(text)
TypeError: Argument 'string' has incorrect type (expected str, got Series)

Можно ли запустить этот код в Pyspark?

...