Matthew Honnibal b2ef6100af Only run backprop once when shared tok2vec weights (#5331) ... Previously, pipelines with shared tok2vec weights would call the tok2vec backprop callback multiple times, once for each pipeline component. This caused errors for PyTorch, and was inefficient. Instead, accumulate the gradient for all but one component, and just call the callback once.		2020-04-21 19:30:41 +02:00
..
__init__.py	Update spaCy for thinc 8.0.0 (#4920)	2020-01-29 17:06:46 +01:00
entityruler.py	Tidy up and auto-format	2020-03-25 12:28:12 +01:00
functions.py	Drop Python 2.7 and 3.5 (#4828)	2019-12-22 01:53:56 +01:00
hooks.py	Default settings to configurations (#4995)	2020-02-27 18:42:27 +01:00
morphologizer.pyx	Update morphologizer (#5108)	2020-04-02 14:46:32 +02:00
pipes.pyx	fix NEL overfitting test for GPU (#5236)	2020-04-02 10:32:52 +02:00
tok2vec.py	Only run backprop once when shared tok2vec weights (#5331)	2020-04-21 19:30:41 +02:00