src/transformers/pipelines/token_classification.py · d447c460b16626c656e4d7a9425f648fe69517b3 · chenpangpang / transformers

Speed up TF token classification postprocessing by converting complete tensors to numpy (#19976) · d447c460

Patrick Deutschmann authored Nov 03, 2022



* Speed up TF postprocessing by converting to numpy before

* Fix bug that was triggered when offset_mapping was None
Co-authored-by: Patrick Deutschmann <patrick.deutschmann@dedalus.com>

d447c460

token_classification.py 20.6 KB

Replace token_classification.py