Commit Graph

81 Commits

Author SHA1 Message Date
Ines Montani
c08b3f294c Support env vars and CLI overrides for project.yml 2021-02-10 13:45:27 +11:00
Adriane Boyd
a45d89f09a Add initialize.before_init and after_init callbacks
Add `initialize.before_init` and `initialize.after_init` callbacks to
the config. The `initialize.before_init` callback is a place to
implement one-time tokenizer customizations that are then saved with the
model.
2021-01-12 13:07:44 +01:00
Ines Montani
513c4e332a
Include custom code via spacy package command (#6531) 2020-12-10 20:36:46 +08:00
Adriane Boyd
fa8fa474a3 Add nlp.batch_size setting
Add a default `batch_size` setting for `Language.pipe` and
`Language.evaluate` as `nlp.batch_size`.
2020-12-09 09:13:26 +01:00
Sofie Van Landeghem
f98a04434a
pretrain architectures (#6451)
* define new architectures for the pretraining objective

* add loss function as attr of the omdel

* cleanup

* cleanup

* shorten name

* fix typo

* remove unused error
2020-12-08 14:41:03 +08:00
Jan Margeta
1ad2213349 Fix TokenPatternSchema pattern field validation
Empty pattern field should be considered invalid

This is fixed by replacing minItems with min_items
as described in Pydantic docs:
https://pydantic-docs.helpmanual.io/usage/schema/
2020-10-16 00:41:21 +02:00
Ines Montani
99606e46fe Relax meta.json schema [ci skip] 2020-10-11 12:30:57 +02:00
Sofie Van Landeghem
d093d6343b
TrainablePipe (#6213)
* rename Pipe to TrainablePipe

* split functionality between Pipe and TrainablePipe

* remove unnecessary methods from certain components

* cleanup

* hasattr(component, "pipe") should be sufficient again

* remove serialization and vocab/cfg from Pipe

* unify _ensure_examples and validate_examples

* small fixes

* hasattr checks for self.cfg and self.vocab

* make is_resizable and is_trainable properties

* serialize strings.json instead of vocab

* fix KB IO + tests

* fix typos

* more typos

* _added_strings as a set

* few more tests specifically for _added_strings field

* bump to 3.0.0a36
2020-10-08 21:33:49 +02:00
Ines Montani
568e12215d
Merge pull request #6206 from svlandeg/fix/patterns-init 2020-10-06 10:27:23 +02:00
svlandeg
ff9ac39c88 read entity_ruler patterns with srsly.read_jsonl.v1 2020-10-05 22:50:14 +02:00
Ines Montani
d58fb42707 Add spacy_version option and validation for project.yml 2020-10-05 20:00:42 +02:00
Adriane Boyd
62ccd5c4df
Relax model meta performance schema (#6185)
Allow more embedded per_x in `ModelMetaSchema`
2020-10-02 16:37:21 +02:00
Ines Montani
fa47f87924 Tidy up and auto-format 2020-09-29 21:39:28 +02:00
Ines Montani
2be80379ec Fix small issues, resolve_dot_names and debug model 2020-09-29 20:38:35 +02:00
Ines Montani
dba26186ef Handle None default args in Cython methods 2020-09-29 18:08:02 +02:00
Ines Montani
fd594cfb9b Tighten up format 2020-09-29 16:47:55 +02:00
Ines Montani
f171903139 Clean up sgd and pipeline -> nlp 2020-09-29 12:20:26 +02:00
Ines Montani
42f0e4c946 Clean up 2020-09-29 12:14:08 +02:00
Ines Montani
50410c17ac Update schemas.py 2020-09-29 12:05:38 +02:00
Ines Montani
dec984a9c1 Update Language.initialize and support components/tokenizer settings 2020-09-29 11:52:45 +02:00
Ines Montani
1590de11b1 Update config 2020-09-28 12:05:23 +02:00
Ines Montani
e44a7519cd Update CLI and add [initialize] block 2020-09-28 11:56:14 +02:00
Ines Montani
658fad428a Fix base schema integration 2020-09-27 22:50:36 +02:00
Ines Montani
7e938ed63e Update config resolution to use new Thinc 2020-09-27 22:21:31 +02:00
Adriane Boyd
3c062b3911
Add MORPH handling to Matcher (#6107)
* Add MORPH handling to Matcher

* Add `MORPH` to `Matcher` schema
* Rename `_SetMemberPredicate` to `_SetPredicate`
* Add `ISSUBSET` and `ISSUPERSET` operators to `_SetPredicate`
  * Add special handling for normalization and conversion of morph
    values into sets
  * For other attrs, `ISSUBSET` acts like `IN` and `ISSUPERSET` only
    matches for 0 or 1 values

* Update test

* Rename to IS_SUBSET and IS_SUPERSET
2020-09-24 16:55:09 +02:00
Ines Montani
74e1f192b4
Merge pull request #6134 from explosion/feature/training_before_to_disk 2020-09-24 14:44:11 +02:00
Ines Montani
88e54caa12 accuracy -> performance 2020-09-24 14:32:35 +02:00
Ines Montani
be56c0994b Add [training.before_to_disk] callback 2020-09-24 12:40:25 +02:00
Ines Montani
ae51f580c1 Fix handling of score_weights 2020-09-24 10:27:33 +02:00
Ines Montani
1114219ae3 Tidy up and auto-format 2020-09-21 10:59:07 +02:00
Ines Montani
e863b3dc14
Merge pull request #6092 from adrianeboyd/bugfix/load-vocab-lookups-2 2020-09-19 12:33:38 +02:00
Sofie Van Landeghem
39872de1f6
Introducing the gpu_allocator (#6091)
* rename 'use_pytorch_for_gpu_memory' to 'gpu_allocator'

* --code instead of --code-path

* update documentation

* avoid querying the "system" section directly

* add explanation of gpu_allocator to TF/PyTorch section in docs

* fix typo

* fix typo 2

* use set_gpu_allocator from thinc 8.0.0a34

* default null instead of empty string
2020-09-19 01:17:02 +02:00
Adriane Boyd
eed4b785f5 Load vocab lookups tables at beginning of training
Similar to how vectors are handled, move the vocab lookups to be loaded
at the start of training rather than when the vocab is initialized,
since the vocab doesn't have access to the full config when it's
created.

The option moves from `nlp.load_vocab_data` to `training.lookups`.

Typically these tables will come from `spacy-lookups-data`, but any
`Lookups` object can be provided.

The loading from `spacy-lookups-data` is now strict, so configs for each
language should specify the exact tables required. This also makes it
easier to control whether the larger clusters and probs tables are
included.

To load `lexeme_norm` from `spacy-lookups-data`:

```
[training.lookups]
@misc = "spacy.LoadLookupsData.v1"
lang = ${nlp.lang}
tables = ["lexeme_norm"]
```
2020-09-18 15:59:16 +02:00
svlandeg
427dbecdd6 cleanup and formatting 2020-09-17 11:48:04 +02:00
svlandeg
0c35885751 generalize corpora, dot notation for dev and train corpus 2020-09-17 11:38:59 +02:00
svlandeg
7336657662 corpus is a Dict 2020-09-15 22:07:16 +02:00
svlandeg
51fa929f47 rewrite train_corpus to corpus.train in config 2020-09-15 21:58:04 +02:00
Matthew Honnibal
54c40223a1
Improve v3 pretrain command (#6040)
* Starts to run

* Update pretrain script

* Update corpus

* Update pretrain schema

* Remove outdated test

* Make JsonlTexts produce Example objects.
2020-09-13 14:05:05 +02:00
Ines Montani
62eec33bc4 Fix meta.json validation 2020-09-11 11:38:33 +02:00
Sofie Van Landeghem
8e7557656f
Renaming gold & annotation_setter (#6042)
* version bump to 3.0.0a16

* rename "gold" folder to "training"

* rename 'annotation_setter' to 'set_extra_annotations'

* formatting
2020-09-09 10:31:03 +02:00
Ines Montani
d7cc2ee72d Fix tests 2020-09-04 14:05:55 +02:00
Ines Montani
3ce5be4b76 Allow loaded but disabled components 2020-08-28 15:20:14 +02:00
Sofie Van Landeghem
79d460e3a2
Weights & Biases logger for train CLI (#5971)
* quick test as part of train script

* train_logger in config, default ConsoleLogger in loggers catalogue

* entitiy typo

* add wandb_logger

* cleanup

* Update spacy/cli/train_logger.py

Co-authored-by: Ines Montani <ines@ines.io>

* move loggers to gold.loggers

Co-authored-by: Ines Montani <ines@ines.io>
2020-08-26 15:24:33 +02:00
Ines Montani
f10989e8c4 Add "project document" and more project.yml meta fields 2020-08-25 17:14:27 +02:00
Ines Montani
94705c21c8 Allow reuse on validators to prevent reload error
Otherwise this will cause an error if spaCy is live reloaded, e.g. in Streamlit
2020-08-25 17:13:11 +02:00
Ines Montani
dd84577a98 Update CLI utils, project.yml schema and add test 2020-08-25 11:54:53 +02:00
Ines Montani
0e7f99da58
Fix handling of optional [pretraining] block (#5954)
* Fix handling of optional [pretraining] block

* Remote pretraining from default config

* Fix test

* Add schema option for empty pretrain block
2020-08-24 15:56:03 +02:00
Matthew Honnibal
e559867605
Allow spacy project to push and pull to/from remote storage (#5949)
* Add utils for working with remote storage

* WIP add remote_cache for project

* WIP add push and pull commands

* Use pathy in remote_cache

* Updarte util

* Update remote_cache

* Update util

* Update project assets

* Update pull script

* Update push script

* Fix type annotation in util

* Work on remote storage

* Remove site and env hash

* Fix imports

* Fix type annotation

* Require pathy

* Require pathy

* Fix import

* Add a util to handle project variable substitution

* Import push and pull commands

* Fix pull command

* Fix push command

* Fix tarfile in remote_storage

* Improve printing

* Fiddle with status messages

* Set version to v3.0.0a9

* Draft docs for spacy project remote storages

* Update docs [ci skip]

* Use Thinc config to simplify and unify template variables

* Auto-format

* Don't import Pathy globally for now

Causes slow and annoying Google Cloud warning

* Tidy up test

* Tidy up and update tests

* Update to latest Thinc

* Update docs

* variables -> vars

* Update docs [ci skip]

* Update docs [ci skip]

Co-authored-by: Ines Montani <ines@ines.io>
2020-08-23 18:32:09 +02:00
Ines Montani
e2f2ef3a5a Update init config and recommendations
- As much as I dislike YAML, it seemed like a better format here because it allows us to add comments if we want to explain the different recommendations
- Don't include the generated JS in the repo by default and build it on the fly when running or deploying the site. This ensures it's always up to date.
- Simplify jinja_to_js script and use fewer dependencies
2020-08-19 13:33:15 +02:00
Ines Montani
82f0e20318 Update docs and consistency [ci skip] 2020-08-18 14:39:40 +02:00