explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-08-04 04:10:20 +03:00
Code Issues Packages Projects Releases Wiki Activity

Website updates for v3-5 draft

Browse Source
This commit is contained in:
Adriane Boyd 2023-01-16 16:58:20 +01:00
parent 283067ef35
commit 3a300b0962
1 changed files with 10 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
website/docs/usage/v3-5.md → website/docs/usage/v3-5.mdx
View File

@ -6,13 +6,13 @@ menu:
- ['Upgrading Notes', 'upgrading'] - ['Upgrading Notes', 'upgrading']
--- ---
## New features {#features hidden="true"} ## New features {id="features",hidden="true"}
spaCy v3.5 introduces three new CLI commands, `apply`, `benchmark` and spaCy v3.5 introduces three new CLI commands, `apply`, `benchmark` and
`find-threshold`, provides improvements and extensions to our entity linking `find-threshold`, provides improvements and extensions to our entity linking
functionality, XXX functionality, XXX
### New CLI commands {#cli} ### New CLI commands {id="cli"}
TODO `apply` TODO `apply`
@ -20,16 +20,16 @@ TODO `benchmark`
TODO `find-threshold` TODO `find-threshold`
### Fuzzy matching {#fuzzy} ### Fuzzy matching {id="fuzzy"}
TODO TODO
### Entity linking generalization {#el} ### Entity linking generalization {id="el"}
The knowledge base used for entity linking is now easier to customize and has a The knowledge base used for entity linking is now easier to customize and has a
new default implementation [`InMemoryLookupKB`](/api/kb_in_memory). new default implementation [`InMemoryLookupKB`](/api/kb_in_memory).
### Additional features and improvements {#additional-features-and-improvements} ### Additional features and improvements {id="additional-features-and-improvements"}
- Language updates: - Language updates:
- Extended support for Slovenian. - Extended support for Slovenian.
@ -61,7 +61,7 @@ new default implementation [`InMemoryLookupKB`](/api/kb_in_memory).
`vectors`. `vectors`.
- Correctly handle missing annotations in the edit tree lemmatizer. - Correctly handle missing annotations in the edit tree lemmatizer.
### Trained pipeline updates {#pipelines} ### Trained pipeline updates {id="pipelines"}
- The CNN pipelines add `IS_SPACE` as a `tok2vec` feature for `tagger` and - The CNN pipelines add `IS_SPACE` as a `tok2vec` feature for `tagger` and
`morphologizer` components to improve tagging of non-whitespace vs. whitespace `morphologizer` components to improve tagging of non-whitespace vs. whitespace
@ -74,15 +74,15 @@ new default implementation [`InMemoryLookupKB`](/api/kb_in_memory).
in the in the
[v1.2.0 release notes](https://github.com/explosion/spacy-transformers/releases/tag/v1.2.0). [v1.2.0 release notes](https://github.com/explosion/spacy-transformers/releases/tag/v1.2.0).
## Notes about upgrading from v3.4 {#upgrading} ## Notes about upgrading from v3.4 {id="upgrading"}
### Validation of textcat values {#textcat-validation} ### Validation of textcat values {id="textcat-validation"}
An error is now raised when unsupported values are given as input to train a An error is now raised when unsupported values are given as input to train a
`textcat` or `textcat_multilabel` model - ensure that values are `0.0` or `1.0` `textcat` or `textcat_multilabel` model - ensure that values are `0.0` or `1.0`
as explained in the [docs](/api/textcategorizer#assigned-attributes). as explained in the [docs](/api/textcategorizer#assigned-attributes).
### Updated default scores for tokenization and textcat {#scores} ### Updated default scores for tokenization and textcat {id="scores"}
We fixed a bug that inflated the `token_acc` scores in v3.0-v3.4. The reported We fixed a bug that inflated the `token_acc` scores in v3.0-v3.4. The reported
`token_acc` will drop from v3.4 to v3.5, but if `token_p/r/f` stay the same, `token_acc` will drop from v3.4 to v3.5, but if `token_p/r/f` stay the same,
@ -97,7 +97,7 @@ For new `textcat` or `textcat_multilabel` configs, the new default `v2` scorers:
- custom scorers can be used to score multiple `textcat` and - custom scorers can be used to score multiple `textcat` and
`textcat_multilabel` components with the built-in `Scorer.score_cats` scorer `textcat_multilabel` components with the built-in `Scorer.score_cats` scorer
### Pipeline package version compatibility {#version-compat} ### Pipeline package version compatibility {id="version-compat"}
> #### Using legacy implementations > #### Using legacy implementations
> >