Matthew Honnibal
a023cf3ecc
Add (untested) resolve_dot_names util
2020-09-28 03:06:12 +02:00
Matthew Honnibal
a976da168c
Support data augmentation in Corpus ( #6155 )
...
* Support data augmentation in Corpus
* Note initial docs for data augmentation
* Add augmenter to quickstart
* Fix flake8
* Format
* Fix test
* Update spacy/tests/training/test_training.py
* Improve data augmentation arguments
* Update templates
* Move randomization out into caller
* Refactor
* Update spacy/training/augment.py
* Update spacy/tests/training/test_training.py
* Fix augment
* Fix test
2020-09-28 03:03:27 +02:00
Matthew Honnibal
13b1605ee6
Add init script
2020-09-28 01:08:49 +02:00
Matthew Honnibal
a3e1791c9c
Upd train
2020-09-28 01:08:30 +02:00
Matthew Honnibal
b5556093e2
Start updating train script
2020-09-27 23:59:44 +02:00
Ines Montani
cad4dbddaa
Merge pull request #6156 from explosion/feature/new-thinc-config-resolution
2020-09-27 23:57:52 +02:00
Ines Montani
9016d23cc5
Fix exclude and add test
2020-09-27 23:34:03 +02:00
Ines Montani
c0c842ae5b
Update Thinc version
2020-09-27 23:24:40 +02:00
Ines Montani
658fad428a
Fix base schema integration
2020-09-27 22:50:36 +02:00
Ines Montani
47c6a461e5
Revert except all in CLI error handling [ci skip]
2020-09-27 22:41:00 +02:00
Ines Montani
5c53a76021
Improve CLI error handling [ci skip]
2020-09-27 22:39:04 +02:00
Ines Montani
e04bd16f7f
Merge branch 'develop' into feature/new-thinc-config-resolution
2020-09-27 22:34:46 +02:00
Ines Montani
d7ad65a9bb
Fix handling of error description [ci skip]
2020-09-27 22:31:57 +02:00
Ines Montani
7e938ed63e
Update config resolution to use new Thinc
2020-09-27 22:21:31 +02:00
Adriane Boyd
013b66de05
Add tokenizer scoring to ja / ko / zh ( #6152 )
2020-09-27 22:20:45 +02:00
Adriane Boyd
a6548ead17
Add _ as a symbol ( #6153 )
...
* Add _ to StringStore in Morphology
* Add _ as a symbol
Add `_` as a symbol instead of adding to the `StringStore`.
2020-09-27 22:20:14 +02:00
Matthew Honnibal
39b178999c
Tmp notes
2020-09-27 20:13:38 +02:00
Ines Montani
f29d5b9b89
Update docs [ci skip]
2020-09-27 18:39:38 +02:00
Adriane Boyd
8393dbedad
Minor fixes
...
* Put `cfg` back in serialization
* Add `pickle5` to pytest conf
2020-09-27 15:15:53 +02:00
Adriane Boyd
54fe871935
Fix formatting, refactor pickle5 exceptions
2020-09-27 14:37:28 +02:00
Adriane Boyd
11e195d3ed
Update ChineseTokenizer
...
* Allow `pkuseg_model` to be set to `None` on initialization
* Don't save config within tokenizer
* Force convert pkuseg_model to use pickle protocol 4 by reencoding with
`pickle5` on serialization
* Update pkuseg serialization test
2020-09-27 14:00:18 +02:00
Ines Montani
3838b14148
Merge pull request #6151 from explosion/fix/train-config-interpolation
2020-09-26 15:56:45 +02:00
Ines Montani
b4486d747d
Merge branch 'develop' into fix/train-config-interpolation
2020-09-26 15:32:14 +02:00
Ines Montani
8fea06d55e
Merge pull request #6149 from adrianeboyd/feature/attributeruler-match-ids
...
Simplify string match IDs for AttributeRuler
2020-09-26 15:31:30 +02:00
Ines Montani
b78a60ef73
Merge pull request #6150 from explosion/feature/cli-config-validation
...
Improve CLI config validation with latest Thinc
2020-09-26 15:30:51 +02:00
Ines Montani
b2d07de786
Construct nlp from uninterpolated config before training
2020-09-26 15:16:59 +02:00
Ines Montani
e06ff8b71d
Update docs [ci skip]
2020-09-26 13:18:08 +02:00
Ines Montani
ca3c997062
Improve CLI config validation with latest Thinc
2020-09-26 13:13:57 +02:00
Adriane Boyd
6c25e60089
Simplify string match IDs for AttributeRuler
2020-09-26 11:12:39 +02:00
Matthew Honnibal
702edf52a0
Fix attributeruler
2020-09-26 00:30:48 +02:00
Matthew Honnibal
821f37254c
Fix attributeruler
2020-09-26 00:19:53 +02:00
Matthew Honnibal
98327f66a9
Fix attributeruler key
2020-09-25 23:20:50 +02:00
Matthew Honnibal
092ce4648e
Make DocBin output stable data (set iteration)
2020-09-25 22:20:44 +02:00
Matthew Honnibal
26afd3bd90
Fix iteration order
2020-09-25 21:47:22 +02:00
Matthew Honnibal
3d8388969e
Sort paths for cache consistency
2020-09-25 19:07:26 +02:00
Adriane Boyd
c3b5a3cfff
Clean up MorphAnalysisC struct ( #6146 )
2020-09-25 15:56:48 +02:00
Sofie Van Landeghem
009ba14aaf
Fix pretraining in train script ( #6143 )
...
* update pretraining API in train CLI
* bump thinc to 8.0.0a35
* bump to 3.0.0a26
* doc fixes
* small doc fix
2020-09-25 15:47:10 +02:00
Ines Montani
02a1b6ab83
Update links [ci skip]
2020-09-25 13:21:43 +02:00
Ines Montani
2cfe9340a1
Link model components [ci skip]
2020-09-25 13:21:20 +02:00
Ines Montani
35f707aa59
Merge pull request #6145 from adrianeboyd/bugfix/revert-score-spans-edits
...
Revert changes to Scorer.score_spans
2020-09-25 09:28:00 +02:00
Ines Montani
c7956a4047
Update models.js [ci skip]
2020-09-25 09:25:46 +02:00
Ines Montani
1aef484985
Fix version check in models directory [ci skip]
2020-09-25 09:24:11 +02:00
Ines Montani
27c5795ea5
Fix version check in models directory [ci skip]
2020-09-25 09:23:29 +02:00
Adriane Boyd
50f20cf722
Revert changes to Scorer.score_spans
2020-09-25 08:21:47 +02:00
Matthew Honnibal
93d7ff309f
Remove print
2020-09-24 21:05:27 +02:00
Ines Montani
2aa4d65734
Update docs [ci skip]
2020-09-24 20:41:09 +02:00
Matthew Honnibal
16475528f7
Fix skipped documents in entity scorer ( #6137 )
...
* Fix skipped documents in entity scorer
* Add back the skipping of unannotated entities
* Update spacy/scorer.py
* Use more specific NER scorer
* Fix import
* Fix get_ner_prf
* Add scorer
* Fix scorer
Co-authored-by: Ines Montani <ines@ines.io>
2020-09-24 20:38:57 +02:00
Matthew Honnibal
2abb4ba9db
Make a pre-check to speed up alignment cache ( #6139 )
...
* Dirty trick to fast-track alignment cache
* Improve alignment cache check
* Fix header
* Fix align cache
* Fix align logic
2020-09-24 18:13:39 +02:00
Ines Montani
26e28ed413
Fix combined scores if multiple components report it
2020-09-24 17:11:13 +02:00
Ines Montani
0b52b6904c
Update entity_linker.py
2020-09-24 17:10:35 +02:00
