Daniël de Kok e5debc68e4 Tagger: use unnormalized probabilities for inference (#10197) ... * Tagger: use unnormalized probabilities for inference Using unnormalized softmax avoids use of the relatively expensive exp function, which can significantly speed up non-transformer models (e.g. I got a speedup of 27% on a German tagging + parsing pipeline). * Add spacy.Tagger.v2 with configurable normalization Normalization of probabilities is disabled by default to improve performance. * Update documentation, models, and tests to spacy.Tagger.v2 * Move Tagger.v1 to spacy-legacy * docs/architectures: run prettier * Unnormalized softmax is now a Softmax_v2 option * Require thinc 8.0.14 and spacy-legacy 3.0.9		2022-03-15 14:15:31 +01:00
..
__init__.py	move tests to correct subdir	2020-09-15 21:40:38 +02:00
test_augmenters.py	Add whitespace and combined augmenters (#10170)	2022-02-17 15:54:09 +01:00
test_new_example.py	Allow Example to align whitespace annotation (#10189)	2022-02-03 17:01:53 +01:00
test_pretraining.py	Tagger: use unnormalized probabilities for inference (#10197)	2022-03-15 14:15:31 +01:00
test_readers.py	Address random results in slow readers tests (#9544)	2021-10-26 16:53:10 +02:00
test_rehearse.py	Auto-format code with black (#10377)	2022-02-25 10:00:21 +01:00
test_training.py	Tagger: use unnormalized probabilities for inference (#10197)	2022-03-15 14:15:31 +01:00