I can't fit the pipeline for RoBerta For Sequence Classification
kareemgamalmahmoud opened this issue · 1 comments
Description
When i try to run (HuggingFace in Spark NLP - RoBertaForSequenceClassification) notebook with colab i always face issues with the last fitting cell with pipline
Error 1 :
NameError: name 'Pipeline' is not defined
And after importing the pipline ( from pyspark.ml import Pipeline )
Then Error 2 :
IllegalArgumentException: requirement failed: Wrong or missing inputCols annotators in REGEX_TOKENIZER_cfae21e0e52f.
Current inputCols: doc ument. Dataset's columns:
(column_name=text,is_nlp_annotator=false)
(column_name=document,is_nlp_annotator=true,type=document).
Make sure such annotators exist in your pipeline, with the right output names and that they have following annotator types: document
I don't know hot to solve !!!
Hi @kareemgamalmahmoud,
There was a typo in this notebook a missing import line as you mentioned. You can use it now without any issues. Thanks for reporting.
