Issues: Current version of Spacy allows the use of packaged transformer pipeline. KeyphraseVectorizers doesn't seem to accommodate as such in setting up the instance.
Error Message: On calling .fit() function, spacy's ([spacy/tokens/underscore.py in getattr(self, name) ) reports the following error:
AttributeError: [E046] Can't retrieve unregistered extension attribute 'trf_data'. Did you forget to call the set_extension method?
Precondition: already loaded en_core_web_trf, spacy_transformers
Wondering if we can make use of transformer pipeline in KeyphraseVectorizers at all.
Issues: Current version of Spacy allows the use of packaged transformer pipeline. KeyphraseVectorizers doesn't seem to accommodate as such in setting up the instance.
Error Message: On calling .fit() function, spacy's ([spacy/tokens/underscore.py in getattr(self, name) ) reports the following error:
AttributeError: [E046] Can't retrieve unregistered extension attribute 'trf_data'. Did you forget to call the
set_extensionmethod?Precondition: already loaded en_core_web_trf, spacy_transformers
Wondering if we can make use of transformer pipeline in KeyphraseVectorizers at all.