From e6a7deb7cc564ed72b9c4e94148446befa20c7fb Mon Sep 17 00:00:00 2001 From: Matthew Honnibal Date: Sun, 26 Jul 2020 13:42:08 +0200 Subject: [PATCH 01/10] Edits to the training 101 section --- website/docs/usage/101/_training.md | 30 +++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) diff --git a/website/docs/usage/101/_training.md b/website/docs/usage/101/_training.md index baf3a1891..ca7971db7 100644 --- a/website/docs/usage/101/_training.md +++ b/website/docs/usage/101/_training.md @@ -1,26 +1,30 @@ -spaCy's models are **statistical** and every "decision" they make – for example, +spaCy's tagger, parser, text categorizer and many other components are powered +by **statistical models**. Every "decision" these components make – for example, which part-of-speech tag to assign, or whether a word is a named entity – is a -**prediction**. This prediction is based on the examples the model has seen +**prediction** based on the model's current **weight values**. The weight +values are estimated based on examples the model has seen during **training**. To train a model, you first need training data – examples of text, and the labels you want the model to predict. This could be a part-of-speech tag, a named entity or any other information. -The model is then shown the unlabelled text and will make a prediction. Because -we know the correct answer, we can give the model feedback on its prediction in -the form of an **error gradient** of the **loss function** that calculates the -difference between the training example and the expected output. The greater the -difference, the more significant the gradient and the updates to our model. +Training is an iterative process in which the model's predictions are compared +against the reference annotations in order to estimate the **gradient of the +loss**. The gradient of the loss is then used to calculate the gradient of the +weights through [backpropagation](https://thinc.ai/backprop101). The gradients +indicate how the weight values should be changed so that the model's +predictions become more similar to the reference labels over time. > - **Training data:** Examples and their annotations. > - **Text:** The input text the model should predict a label for. > - **Label:** The label the model should predict. -> - **Gradient:** Gradient of the loss function calculating the difference -> between input and expected output. +> - **Gradient:** The direction and rate of change for a numeric value. +> Minimising the gradient of the weights should result in predictions that +> are closer to the reference labels on the training data. ![The training process](../../images/training.svg) When training a model, we don't just want it to memorize our examples – we want -it to come up with a theory that can be **generalized across other examples**. +it to come up with a theory that can be **generalized across unseen data**. After all, we don't just want the model to learn that this one instance of "Amazon" right here is a company – we want it to learn that "Amazon", in contexts _like this_, is most likely a company. That's why the training data @@ -34,5 +38,7 @@ it's learning the right things, you don't only need **training data** – you'll also need **evaluation data**. If you only test the model with the data it was trained on, you'll have no idea how well it's generalizing. If you want to train a model from scratch, you usually need at least a few hundred examples for both -training and evaluation. To update an existing model, you can already achieve -decent results with very few examples – as long as they're representative. +training and evaluation. A good rule of thumb is that you should have 10 +samples for each significant figure of accuracy you report. +If you only have 100 samples and your model predicts 92 of them correctly, you +would report accuracy of 0.9 rather than 0.92. From fb5dbe30b5cb662113d77903cb64552c57aa6ef9 Mon Sep 17 00:00:00 2001 From: Matthew Honnibal Date: Sun, 26 Jul 2020 13:43:22 +0200 Subject: [PATCH 02/10] Trim training 101 --- website/docs/usage/101/_training.md | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/website/docs/usage/101/_training.md b/website/docs/usage/101/_training.md index ca7971db7..4573f5ea3 100644 --- a/website/docs/usage/101/_training.md +++ b/website/docs/usage/101/_training.md @@ -38,7 +38,4 @@ it's learning the right things, you don't only need **training data** – you'll also need **evaluation data**. If you only test the model with the data it was trained on, you'll have no idea how well it's generalizing. If you want to train a model from scratch, you usually need at least a few hundred examples for both -training and evaluation. A good rule of thumb is that you should have 10 -samples for each significant figure of accuracy you report. -If you only have 100 samples and your model predicts 92 of them correctly, you -would report accuracy of 0.9 rather than 0.92. +training and evaluation. From 53d37da29a5103300a561be7ca8547a5ddcbedc3 Mon Sep 17 00:00:00 2001 From: Ines Montani Date: Sun, 26 Jul 2020 15:11:24 +0200 Subject: [PATCH 03/10] Make sure @factories is removed from config --- spacy/language.py | 1 + 1 file changed, 1 insertion(+) diff --git a/spacy/language.py b/spacy/language.py index 93d239356..4567fe57b 100644 --- a/spacy/language.py +++ b/spacy/language.py @@ -532,6 +532,7 @@ class Language: resolved, filled = registry.resolve(cfg, validate=validate, overrides=overrides) filled = filled[factory_name] filled["factory"] = factory_name + filled.pop("@factories", None) self._pipe_configs[name] = filled return resolved[factory_name] From ed61fb10fcf7f95d1e9532eea6bf7da391a75e78 Mon Sep 17 00:00:00 2001 From: Ines Montani Date: Sun, 26 Jul 2020 15:11:43 +0200 Subject: [PATCH 04/10] Rename default textcat arch to TextCatEnsemble --- spacy/ml/models/textcat.py | 2 +- spacy/pipeline/textcat.py | 2 +- spacy/tests/pipeline/test_pipe_factories.py | 2 +- spacy/tests/pipeline/test_textcat.py | 6 +++--- 4 files changed, 6 insertions(+), 6 deletions(-) diff --git a/spacy/ml/models/textcat.py b/spacy/ml/models/textcat.py index 94e4f72b5..e5f4af2fb 100644 --- a/spacy/ml/models/textcat.py +++ b/spacy/ml/models/textcat.py @@ -53,7 +53,7 @@ def build_bow_text_classifier(exclusive_classes, ngram_size, no_output_layer, nO return model -@registry.architectures.register("spacy.TextCat.v1") +@registry.architectures.register("spacy.TextCatEnsemble.v1") def build_text_classifier( width, embed_size, diff --git a/spacy/pipeline/textcat.py b/spacy/pipeline/textcat.py index 3c0808342..628ff250b 100644 --- a/spacy/pipeline/textcat.py +++ b/spacy/pipeline/textcat.py @@ -14,7 +14,7 @@ from ..vocab import Vocab default_model_config = """ [model] -@architectures = "spacy.TextCat.v1" +@architectures = "spacy.TextCatEnsemble.v1" exclusive_classes = false pretrained_vectors = null width = 64 diff --git a/spacy/tests/pipeline/test_pipe_factories.py b/spacy/tests/pipeline/test_pipe_factories.py index db090fdf2..990509b02 100644 --- a/spacy/tests/pipeline/test_pipe_factories.py +++ b/spacy/tests/pipeline/test_pipe_factories.py @@ -155,7 +155,7 @@ def test_pipe_class_component_model(): name = "test_class_component_model" default_config = { "model": { - "@architectures": "spacy.TextCat.v1", + "@architectures": "spacy.TextCatEnsemble.v1", "exclusive_classes": False, "pretrained_vectors": None, "width": 64, diff --git a/spacy/tests/pipeline/test_textcat.py b/spacy/tests/pipeline/test_textcat.py index 5e8dab0bd..4ae6d63ac 100644 --- a/spacy/tests/pipeline/test_textcat.py +++ b/spacy/tests/pipeline/test_textcat.py @@ -131,9 +131,9 @@ def test_overfitting_IO(): {"@architectures": "spacy.TextCatBOW.v1", "exclusive_classes": True, "ngram_size": 4, "no_output_layer": False}, {"@architectures": "spacy.TextCatBOW.v1", "exclusive_classes": False, "ngram_size": 3, "no_output_layer": True}, {"@architectures": "spacy.TextCatBOW.v1", "exclusive_classes": True, "ngram_size": 2, "no_output_layer": True}, - {"@architectures": "spacy.TextCat.v1", "exclusive_classes": False, "ngram_size": 1, "pretrained_vectors": False, "width": 64, "conv_depth": 2, "embed_size": 2000, "window_size": 2, "dropout": None}, - {"@architectures": "spacy.TextCat.v1", "exclusive_classes": True, "ngram_size": 5, "pretrained_vectors": False, "width": 128, "conv_depth": 2, "embed_size": 2000, "window_size": 1, "dropout": None}, - {"@architectures": "spacy.TextCat.v1", "exclusive_classes": True, "ngram_size": 2, "pretrained_vectors": False, "width": 32, "conv_depth": 3, "embed_size": 500, "window_size": 3, "dropout": None}, + {"@architectures": "spacy.TextCatEnsemble.v1", "exclusive_classes": False, "ngram_size": 1, "pretrained_vectors": False, "width": 64, "conv_depth": 2, "embed_size": 2000, "window_size": 2, "dropout": None}, + {"@architectures": "spacy.TextCatEnsemble.v1", "exclusive_classes": True, "ngram_size": 5, "pretrained_vectors": False, "width": 128, "conv_depth": 2, "embed_size": 2000, "window_size": 1, "dropout": None}, + {"@architectures": "spacy.TextCatEnsemble.v1", "exclusive_classes": True, "ngram_size": 2, "pretrained_vectors": False, "width": 32, "conv_depth": 3, "embed_size": 500, "window_size": 3, "dropout": None}, {"@architectures": "spacy.TextCatCNN.v1", "tok2vec": DEFAULT_TOK2VEC_MODEL, "exclusive_classes": True}, {"@architectures": "spacy.TextCatCNN.v1", "tok2vec": DEFAULT_TOK2VEC_MODEL, "exclusive_classes": False}, ], From 80271ac0baed8596260c6f44598d736fe55b5f75 Mon Sep 17 00:00:00 2001 From: Matthew Honnibal Date: Sun, 26 Jul 2020 15:27:39 +0200 Subject: [PATCH 05/10] Update default config --- spacy/default_config.cfg | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/spacy/default_config.cfg b/spacy/default_config.cfg index 7ef4eadfc..43881bf84 100644 --- a/spacy/default_config.cfg +++ b/spacy/default_config.cfg @@ -39,18 +39,13 @@ score_weights = {"tag_acc": 0.2, "dep_las": 0.4, "ents_f": 0.4} # These settings are invalid for the transformer models. init_tok2vec = null discard_oversize = false -batch_by = "sequences" raw_text = null tag_map = null morph_rules = null base_model = null vectors = null - -[training.batch_size] -@schedules = "compounding.v1" -start = 1000 -stop = 1000 -compound = 1.001 +batch_by = "words" +batch_size = 1000 [training.optimizer] @optimizers = "Adam.v1" From 3d56a3f286485a9f1741286476f8872d8e3be1c8 Mon Sep 17 00:00:00 2001 From: Ines Montani Date: Mon, 27 Jul 2020 00:27:53 +0200 Subject: [PATCH 06/10] Make more args keyword-only --- spacy/language.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/spacy/language.py b/spacy/language.py index 4567fe57b..ba5b499ec 100644 --- a/spacy/language.py +++ b/spacy/language.py @@ -110,6 +110,7 @@ class Language: def __init__( self, vocab: Union[Vocab, bool] = True, + *, max_length: int = 10 ** 6, meta: Dict[str, Any] = {}, create_tokenizer: Optional[Callable[["Language"], Callable[[str], Doc]]] = None, @@ -1268,6 +1269,7 @@ class Language: def from_config( cls, config: Union[Dict[str, Any], Config] = {}, + *, disable: Iterable[str] = tuple(), overrides: Dict[str, Any] = {}, auto_fill: bool = True, From 7adbaf9a5b5b76b10271cbc66677016bb7170202 Mon Sep 17 00:00:00 2001 From: Ines Montani Date: Mon, 27 Jul 2020 00:29:45 +0200 Subject: [PATCH 07/10] Update docs [ci skip] --- website/README.md | 2 +- website/docs/api/dependencyparser.md | 30 +- website/docs/api/entitylinker.md | 33 +- website/docs/api/entityrecognizer.md | 30 +- website/docs/api/entityruler.md | 17 +- website/docs/api/example.md | 2 +- website/docs/api/language.md | 296 +++++++-- website/docs/api/morphologizer.md | 120 ++-- website/docs/api/pipeline-functions.md | 26 +- website/docs/api/sentencerecognizer.md | 26 +- website/docs/api/sentencizer.md | 14 +- website/docs/api/tagger.md | 15 +- website/docs/api/textcategorizer.md | 32 +- website/docs/api/tok2vec.md | 15 +- website/docs/api/tokenizer.md | 2 +- website/docs/api/top-level.md | 4 +- website/docs/images/pipeline.svg | 61 +- website/docs/images/training.svg | 103 ++-- website/docs/usage/101/_architecture.md | 16 +- website/docs/usage/101/_pipelines.md | 36 +- website/docs/usage/linguistic-features.md | 14 +- website/docs/usage/processing-pipelines.md | 677 +++++++++++++++------ website/docs/usage/projects.md | 8 + website/docs/usage/rule-based-matching.md | 106 ++-- website/docs/usage/saving-loading.md | 318 +++++----- website/docs/usage/training.md | 15 +- website/docs/usage/transformers.md | 6 + website/docs/usage/v3.md | 72 +++ website/docs/usage/vectors-embeddings.md | 2 + website/meta/sidebars.json | 1 + website/src/styles/layout.sass | 2 +- website/src/styles/table.module.sass | 3 +- 32 files changed, 1298 insertions(+), 806 deletions(-) create mode 100644 website/docs/usage/transformers.md diff --git a/website/README.md b/website/README.md index a02d5a151..c1f6e5805 100644 --- a/website/README.md +++ b/website/README.md @@ -384,7 +384,7 @@ original file is shown at the top of the widget. > ``` ```python -https://github.com/explosion/spaCy/tree/master/examples/pipeline/custom_component_countries_api.py +https://github.com/explosion/spaCy/tree/master/spacy/language.py ``` ### Infobox diff --git a/website/docs/api/dependencyparser.md b/website/docs/api/dependencyparser.md index 135caf0c2..52442fe49 100644 --- a/website/docs/api/dependencyparser.md +++ b/website/docs/api/dependencyparser.md @@ -1,23 +1,22 @@ --- title: DependencyParser tag: class -source: spacy/pipeline/pipes.pyx +source: spacy/pipeline/dep_parser.pyx --- This class is a subclass of `Pipe` and follows the same API. The pipeline component is available in the [processing pipeline](/usage/processing-pipelines) via the ID `"parser"`. -## Default config {#config} +## Implementation and defaults {#implementation} -This is the default configuration used to initialize the model powering the -pipeline component. See the [model architectures](/api/architectures) -documentation for details on the architectures and their arguments and -hyperparameters. To learn more about how to customize the config and train -custom models, check out the [training config](/usage/training#config) docs. +See the [model architectures](/api/architectures) documentation for details on +the architectures and their arguments and hyperparameters. To learn more about +how to customize the config and train custom models, check out the +[training config](/usage/training#config) docs. ```python -https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/parser_defaults.cfg +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/dep_parser.pyx ``` ## DependencyParser.\_\_init\_\_ {#init tag="method"} @@ -25,22 +24,17 @@ https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/parser_d > #### Example > > ```python -> # Construction via create_pipe with default model -> parser = nlp.create_pipe("parser") +> # Construction via add_pipe with default model +> parser = nlp.add_pipe("parser") > -> # Construction via create_pipe with custom model +> # Construction via add_pipe with custom model > config = {"model": {"@architectures": "my_parser"}} -> parser = nlp.create_pipe("parser", config) -> -> # Construction from class with custom model from file -> from spacy.pipeline import DependencyParser -> model = util.load_config("model.cfg", create_objects=True)["model"] -> parser = DependencyParser(nlp.vocab, model) +> parser = nlp.add_pipe("parser", config=config) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and -[`nlp.create_pipe`](/api/language#create_pipe). +[`nlp.add_pipe`](/api/language#add_pipe). | Name | Type | Description | | ----------- | ------------------ | ------------------------------------------------------------------------------- | diff --git a/website/docs/api/entitylinker.md b/website/docs/api/entitylinker.md index b77fc059d..bbc6cc632 100644 --- a/website/docs/api/entitylinker.md +++ b/website/docs/api/entitylinker.md @@ -4,7 +4,7 @@ teaser: Functionality to disambiguate a named entity in text to a unique knowledge base identifier. tag: class -source: spacy/pipeline/pipes.pyx +source: spacy/pipeline/entity_linker.py new: 2.2 --- @@ -12,16 +12,15 @@ This class is a subclass of `Pipe` and follows the same API. The pipeline component is available in the [processing pipeline](/usage/processing-pipelines) via the ID `"entity_linker"`. -## Default config {#config} +## Implementation and defaults {#implementation} -This is the default configuration used to initialize the model powering the -pipeline component. See the [model architectures](/api/architectures) -documentation for details on the architectures and their arguments and -hyperparameters. To learn more about how to customize the config and train -custom models, check out the [training config](/usage/training#config) docs. +See the [model architectures](/api/architectures) documentation for details on +the architectures and their arguments and hyperparameters. To learn more about +how to customize the config and train custom models, check out the +[training config](/usage/training#config) docs. ```python -https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/entity_linker_defaults.cfg +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/entity_linker.py ``` ## EntityLinker.\_\_init\_\_ {#init tag="method"} @@ -29,22 +28,17 @@ https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/entity_l > #### Example > > ```python -> # Construction via create_pipe with default model -> entity_linker = nlp.create_pipe("entity_linker") +> # Construction via add_pipe with default model +> entity_linker = nlp.add_pipe("entity_linker") > -> # Construction via create_pipe with custom model +> # Construction via add_pipe with custom model > config = {"model": {"@architectures": "my_el"}} -> entity_linker = nlp.create_pipe("entity_linker", config) -> -> # Construction from class with custom model from file -> from spacy.pipeline import EntityLinker -> model = util.load_config("model.cfg", create_objects=True)["model"] -> entity_linker = EntityLinker(nlp.vocab, model) +> entity_linker = nlp.add_pipe("entity_linker", config=config) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and -[`nlp.create_pipe`](/api/language#create_pipe). +[`nlp.add_pipe`](/api/language#add_pipe). | Name | Type | Description | | ------- | ------- | ------------------------------------------------------------------------------- | @@ -185,9 +179,8 @@ method, a knowledge base should have been defined with > #### Example > > ```python -> entity_linker = EntityLinker(nlp.vocab) +> entity_linker = nlp.add_pipe("entity_linker", last=True) > entity_linker.set_kb(kb) -> nlp.add_pipe(entity_linker, last=True) > optimizer = entity_linker.begin_training(pipeline=nlp.pipeline) > ``` diff --git a/website/docs/api/entityrecognizer.md b/website/docs/api/entityrecognizer.md index 23cc71558..6b5b1f93c 100644 --- a/website/docs/api/entityrecognizer.md +++ b/website/docs/api/entityrecognizer.md @@ -1,23 +1,22 @@ --- title: EntityRecognizer tag: class -source: spacy/pipeline/pipes.pyx +source: spacy/pipeline/ner.pyx --- This class is a subclass of `Pipe` and follows the same API. The pipeline component is available in the [processing pipeline](/usage/processing-pipelines) via the ID `"ner"`. -## Default config {#config} +## Implementation and defaults {#implementation} -This is the default configuration used to initialize the model powering the -pipeline component. See the [model architectures](/api/architectures) -documentation for details on the architectures and their arguments and -hyperparameters. To learn more about how to customize the config and train -custom models, check out the [training config](/usage/training#config) docs. +See the [model architectures](/api/architectures) documentation for details on +the architectures and their arguments and hyperparameters. To learn more about +how to customize the config and train custom models, check out the +[training config](/usage/training#config) docs. ```python -https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/ner_defaults.cfg +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/ner.pyx ``` ## EntityRecognizer.\_\_init\_\_ {#init tag="method"} @@ -25,22 +24,17 @@ https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/ner_defa > #### Example > > ```python -> # Construction via create_pipe -> ner = nlp.create_pipe("ner") +> # Construction via add_pipe with default model +> ner = nlp.add_pipe("ner") > -> # Construction via create_pipe with custom model +> # Construction via add_pipe with custom model > config = {"model": {"@architectures": "my_ner"}} -> parser = nlp.create_pipe("ner", config) -> -> # Construction from class with custom model from file -> from spacy.pipeline import EntityRecognizer -> model = util.load_config("model.cfg", create_objects=True)["model"] -> ner = EntityRecognizer(nlp.vocab, model) +> parser = nlp.add_pipe("ner", config=config) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and -[`nlp.create_pipe`](/api/language#create_pipe). +[`nlp.add_pipe`](/api/language#add_pipe). | Name | Type | Description | | ----------- | ------------------ | ------------------------------------------------------------------------------- | diff --git a/website/docs/api/entityruler.md b/website/docs/api/entityruler.md index 1279a3685..12a6ab748 100644 --- a/website/docs/api/entityruler.md +++ b/website/docs/api/entityruler.md @@ -8,10 +8,10 @@ new: 2.1 The EntityRuler lets you add spans to the [`Doc.ents`](/api/doc#ents) using token-based rules or exact phrase matches. It can be combined with the statistical [`EntityRecognizer`](/api/entityrecognizer) to boost accuracy, or -used on its own to implement a purely rule-based entity recognition system. -After initialization, the component is typically added to the processing -pipeline using [`nlp.add_pipe`](/api/language#add_pipe). For usage examples, see -the docs on +used on its own to implement a purely rule-based entity recognition system. The +pipeline component is available in the +[processing pipeline](/usage/processing-pipelines) via the ID `"entity_ruler"`. +For usage examples, see the docs on [rule-based entity recognition](/usage/rule-based-matching#entityruler). ## EntityRuler.\_\_init\_\_ {#init tag="method"} @@ -19,13 +19,13 @@ the docs on Initialize the entity ruler. If patterns are supplied here, they need to be a list of dictionaries with a `"label"` and `"pattern"` key. A pattern can either be a token pattern (list) or a phrase pattern (string). For example: -`{'label': 'ORG', 'pattern': 'Apple'}`. +`{"label": "ORG", "pattern": "Apple"}`. > #### Example > > ```python -> # Construction via create_pipe -> ruler = nlp.create_pipe("entity_ruler") +> # Construction via add_pipe +> ruler = nlp.add_pipe("entity_ruler") > > # Construction from class > from spacy.pipeline import EntityRuler @@ -90,9 +90,8 @@ is chosen. > #### Example > > ```python -> ruler = EntityRuler(nlp) +> ruler = nlp.add_pipe("entity_ruler") > ruler.add_patterns([{"label": "ORG", "pattern": "Apple"}]) -> nlp.add_pipe(ruler) > > doc = nlp("A text about Apple.") > ents = [(ent.text, ent.label_) for ent in doc.ents] diff --git a/website/docs/api/example.md b/website/docs/api/example.md index 421828f95..0d06c79a1 100644 --- a/website/docs/api/example.md +++ b/website/docs/api/example.md @@ -223,7 +223,7 @@ in `example.predicted`. > #### Example > > ```python -> nlp.add_pipe(my_ner) +> nlp.add_pipe("my_ner") > doc = nlp("Mr and Mrs Smith flew to New York") > tokens_ref = ["Mr and Mrs", "Smith", "flew", "to", "New York"] > example = Example.from_dict(doc, {"words": tokens_ref}) diff --git a/website/docs/api/language.md b/website/docs/api/language.md index 9ab25597d..c41b349e5 100644 --- a/website/docs/api/language.md +++ b/website/docs/api/language.md @@ -15,6 +15,88 @@ the tagger or parser that are called on a document in order. You can also add your own processing pipeline components that take a `Doc` object, modify it and return it. +## Language.component {#component tag="classmethod" new="3"} + +Register a custom pipeline component under a given name. This allows +initializing the component by name using +[`Language.add_pipe`](/api/language#add_pipe) and referring to it in +[config files](/usage/training#config). This classmethod and decorator is +intended for **simple stateless functions** that take a `Doc` and return it. For +more complex stateful components that allow settings and need access to the +shared `nlp` object, use the [`Language.factory`](/api/language#factory) +decorator. For more details and examples, see the +[usage documentation](/usage/processing-pipelines#custom-components). + +> #### Example +> +> ```python +> from spacy.language import Language +> +> # Usage as a decorator +> @Language.component("my_component") +> def my_component(doc): +> # Do something to the doc +> return doc +> +> # Usage as a function +> Language.component("my_component2", func=my_component) +> ``` + +| Name | Type | Description | +| -------------- | -------------------- | --------------------------------------------------------------------------------------------------------------------------------------------- | +| `name` | str | The name of the component factory. | +| _keyword-only_ | | | +| `assigns` | `Iterable[str]` | `Doc` or `Token` attributes assigned by this component, e.g. `["token.ent_id"]`. Used for pipeline analysis. | +| `requires` | `Iterable[str]` | `Doc` or `Token` attributes required by this component, e.g. `["token.ent_id"]`. Used for pipeline analysis. | +| `retokenizes` | bool | Whether the component changes tokenization. Used for pipeline analysis. | +| `func` | `Optional[Callable]` | Optional function if not used a a decorator. | + +## Language.factory {#factory tag="classmethod"} + +Register a custom pipeline component factory under a given name. This allows +initializing the component by name using +[`Language.add_pipe`](/api/language#add_pipe) and referring to it in +[config files](/usage/training#config). The registered factory function needs to +take at least two **named arguments** which spaCy fills in automatically: `nlp` +for the current `nlp` object and `name` for the component instance name. This +can be useful to distinguish multiple instances of the same component and allows +trainable components to add custom losses using the component instance name. The +`default_config` defines the default values of the remaining factory arguments. +It's merged into the [`nlp.config`](/api/language#config). For more details and +examples, see the +[usage documentation](/usage/processing-pipelines#custom-components). + +> #### Example +> +> ```python +> from spacy.language import Language +> +> # Usage as a decorator +> @Language.factory( +> "my_component", +> default_config={"some_setting": True}, +> ) +> def create_my_component(nlp, name, some_setting): +> return MyComponent(some_setting) +> +> # Usage as function +> Language.factory( +> "my_component", +> default_config={"some_setting": True}, +> func=create_my_component +> ) +> ``` + +| Name | Type | Description | +| ---------------- | -------------------- | --------------------------------------------------------------------------------------------------------------------------------------------- | +| `name` | str | The name of the component factory. | +| _keyword-only_ | | | +| `default_config` | `Dict[str, any]` | The default config, describing the default values of the factory arguments. | +| `assigns` | `Iterable[str]` | `Doc` or `Token` attributes assigned by this component, e.g. `["token.ent_id"]`. Used for pipeline analysis. | +| `requires` | `Iterable[str]` | `Doc` or `Token` attributes required by this component, e.g. `["token.ent_id"]`. Used for pipeline analysis. | +| `retokenizes` | bool | Whether the component changes tokenization. Used for pipeline analysis. | +| `func` | `Optional[Callable]` | Optional function if not used a a decorator. | + ## Language.\_\_init\_\_ {#init tag="method"} Initialize a `Language` object. @@ -30,12 +112,41 @@ Initialize a `Language` object. > nlp = English() > ``` -| Name | Type | Description | -| ----------- | ---------- | ------------------------------------------------------------------------------------------ | -| `vocab` | `Vocab` | A `Vocab` object. If `True`, a vocab is created via `Language.Defaults.create_vocab`. | -| `make_doc` | callable | A function that takes text and returns a `Doc` object. Usually a `Tokenizer`. | -| `meta` | dict | Custom meta data for the `Language` class. Is written to by models to add model meta data. | -| **RETURNS** | `Language` | The newly constructed object. | +| Name | Type | Description | +| ------------------ | ----------- | ------------------------------------------------------------------------------------------ | +| `vocab` | `Vocab` | A `Vocab` object. If `True`, a vocab is created using the default language data settings. | +| _keyword-only_ | | | +| `max_length` | int | Maximum number of characters allowed in a single text. Defaults to `10 ** 6`. | +| `meta` | dict | Custom meta data for the `Language` class. Is written to by models to add model meta data. | +| `create_tokenizer` |  `Callable` | Optional function that receives the `nlp` object and returns a tokenizer. | +| **RETURNS** | `Language` | The newly constructed object. | + +## Language.from_config {#from_config tag="classmethod"} + +Create a `Language` object from a loaded config. Will set up the tokenizer and +language data, add pipeline components based on the pipeline and components +define in the config and validate the results. If no config is provided, the +default config of the given language is used. This is also how spaCy loads a +model under the hood based on its [`config.cfg`](/api/data-formats#config). + +> #### Example +> +> ```python +> from thinc.api import Config +> from spacy.language import Language +> +> config = Config().from_disk("./config.cfg") +> nlp = Language.from_config(config) +> ``` + +| Name | Type | Description | +| -------------- | ---------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------- | +| `config` | `Dict[str, Any]` / [`Config`](https://thinc.ai/docs/api-config#config) | The loaded config. | +| _keyword-only_ | | +| `disable` | `Iterable[str]` | List of pipeline component names to disable. | +| `auto_fill` | bool | Whether to automatically fill in missing values in the config, based on defaults and function argument annotations. Defaults to `True`. | +| `validate` | bool | Whether to validate the component config and arguments against the types expected by the factory. Defaults to `True`. | +| **RETURNS** | `Language` | The initialized object. | ## Language.\_\_call\_\_ {#call tag="method"} @@ -162,43 +273,99 @@ their original weights after the block. Create a pipeline component from a factory. + + +As of v3.0, the [`Language.add_pipe`](/api/language#add_pipe) method also takes +the string name of the factory, creates the component, adds it to the pipeline +and returns it. The `Language.create_pipe` method is now mostly used internally. +To create a component and add it to the pipeline, you should always use +`Language.add_pipe`. + + + > #### Example > > ```python > parser = nlp.create_pipe("parser") -> nlp.add_pipe(parser) > ``` -| Name | Type | Description | -| ----------- | -------- | ---------------------------------------------------------------------------------- | -| `name` | str | Factory name to look up in [`Language.factories`](/api/language#class-attributes). | -| `config` | dict | Configuration parameters to initialize component. | -| **RETURNS** | callable | The pipeline component. | +| Name | Type | Description | +| ------------------------------------- | ---------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `factory_name` | str | Name of the registered component factory. | +| `name` | str | Optional unique name of pipeline component instance. If not set, the factory name is used. An error is raised if the name already exists in the pipeline. | +| `config` 3 | `Dict[str, Any]` | Optional config parameters to use for this component. Will be merged with the `default_config` specified by the component factory. | +| `validate` 3 | bool | Whether to validate the component config and arguments against the types expected by the factory. Defaults to `True`. | +| **RETURNS** | callable | The pipeline component. | ## Language.add_pipe {#add_pipe tag="method" new="2"} -Add a component to the processing pipeline. Valid components are callables that -take a `Doc` object, modify it and return it. Only one of `before`, `after`, -`first` or `last` can be set. Default behavior is `last=True`. +Add a component to the processing pipeline. Expects a name that maps to a +component factory registered using +[`@Language.component`](/api/language#component) or +[`@Language.factory`](/api/language#factory). Components should be callables +that take a `Doc` object, modify it and return it. Only one of `before`, +`after`, `first` or `last` can be set. Default behavior is `last=True`. + + + +As of v3.0, the [`Language.add_pipe`](/api/language#add_pipe) method doesn't +take callables anymore and instead expects the name of a component factory +registered using [`@Language.component`](/api/language#component) or +[`@Language.factory`](/api/language#factory). It now takes care of creating the +component, adds it to the pipeline and returns it. + + > #### Example > > ```python -> def component(doc): +> @Language.component("component") +> def component_func(doc): > # modify Doc and return it return doc > -> nlp.add_pipe(component, before="ner") -> nlp.add_pipe(component, name="custom_name", last=True) +> nlp.add_pipe("component", before="ner") +> component = nlp.add_pipe("component", name="custom_name", last=True) > ``` -| Name | Type | Description | -| ----------- | -------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -| `component` | callable | The pipeline component. | -| `name` | str | Name of pipeline component. Overwrites existing `component.name` attribute if available. If no `name` is set and the component exposes no name attribute, `component.__name__` is used. An error is raised if the name already exists in the pipeline. | -| `before` | str | Component name to insert component directly before. | -| `after` | str | Component name to insert component directly after: | -| `first` | bool | Insert component first / not first in the pipeline. | -| `last` | bool | Insert component last / not last in the pipeline. | +| Name | Type | Description | +| -------------------------------------- | ---------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `factory_name` | str | Name of the registered component factory. | +| `name` | str | Optional unique name of pipeline component instance. If not set, the factory name is used. An error is raised if the name already exists in the pipeline. | +| _keyword-only_ | | | +| `before` | str / int | Component name or index to insert component directly before. | +| `after` | str / int | Component name or index to insert component directly after: | +| `first` | bool | Insert component first / not first in the pipeline. | +| `last` | bool | Insert component last / not last in the pipeline. | +| `config` 3 | `Dict[str, Any]` | Optional config parameters to use for this component. Will be merged with the `default_config` specified by the component factory. | +| `validate` 3 | bool | Whether to validate the component config and arguments against the types expected by the factory. Defaults to `True`. | +| **RETURNS** 3 | callable | The pipeline component. | + +## Language.has_factory {#has_factory tag="classmethod" new="3"} + +Check whether a factory name is registered on the `Language` class or subclass. +Will check for +[language-specific factories](/usage/processing-pipelines#factories-language) +registered on the subclass, as well as general-purpose factories registered on +the `Language` base class, available to all subclasses. + +> #### Example +> +> ```python +> from spacy.language import Language +> from spacy.lang.en import English +> +> @English.component("component") +> def component(doc): +> return doc +> +> assert English.has_factory("component") +> assert not Language.has_factory("component") +> ``` + +| Name | Type | Description | +| ----------- | ---- | ---------------------------------------------------------- | +| `name` | str | Name of the pipeline factory to check. | +| **RETURNS** | bool | Whether a factory of that name is registered on the class. | ## Language.has_pipe {#has_pipe tag="method" new="2"} @@ -208,9 +375,13 @@ Check whether a component is present in the pipeline. Equivalent to > #### Example > > ```python -> nlp.add_pipe(lambda doc: doc, name="component") -> assert "component" in nlp.pipe_names -> assert nlp.has_pipe("component") +> @Language.component("component") +> def component(doc): +> return doc +> +> nlp.add_pipe("component", name="my_component") +> assert "my_component" in nlp.pipe_names +> assert nlp.has_pipe("my_component") > ``` | Name | Type | Description | @@ -324,6 +495,43 @@ As of spaCy v3.0, the `disable_pipes` method has been renamed to `select_pipes`: | `enable` | str / list | Names(s) of pipeline components that will not be disabled. | | **RETURNS** | `DisabledPipes` | The disabled pipes that can be restored by calling the object's `.restore()` method. | +## Language.meta {#meta tag="property"} + +Custom meta data for the Language class. If a model is loaded, contains meta +data of the model. The `Language.meta` is also what's serialized as the +`meta.json` when you save an `nlp` object to disk. + +> #### Example +> +> ```python +> print(nlp.meta) +> ``` + +| Name | Type | Description | +| ----------- | ---- | -------------- | +| **RETURNS** | dict | The meta data. | + +## Language.config {#config tag="property" new="3"} + +Export a trainable [`config.cfg`](/api/data-formats#config) for the current +`nlp` object. Includes the current pipeline, all configs used to create the +currently active pipeline components, as well as the default training config +that can be used with [`spacy train`](/api/cli#train). `Language.config` returns +a [Thinc `Config` object](https://thinc.ai/docs/api-config#config), which is a +subclass of the built-in `dict`. It supports the additional methods `to_disk` +(serialize the config to a file) and `to_str` (output the config as a string). + +> #### Example +> +> ```python +> nlp.config.to_disk("./config.cfg") +> print(nlp.config.to_str()) +> ``` + +| Name | Type | Description | +| ----------- | --------------------------------------------------- | ----------- | +| **RETURNS** | [`Config`](https://thinc.ai/docs/api-config#config) | The config. | + ## Language.to_disk {#to_disk tag="method" new="2"} Save the current state to a directory. If a model is loaded, this will **include @@ -405,23 +613,25 @@ available to the loaded object. ## Attributes {#attributes} -| Name | Type | Description | -| ------------------------------------------ | ----------- | ----------------------------------------------------------------------------------------------- | -| `vocab` | `Vocab` | A container for the lexical types. | -| `tokenizer` | `Tokenizer` | The tokenizer. | -| `make_doc` | `callable` | Callable that takes a string and returns a `Doc`. | -| `pipeline` | list | List of `(name, component)` tuples describing the current processing pipeline, in order. | -| `pipe_names` 2 | list | List of pipeline component names, in order. | -| `pipe_labels` 2.2 | dict | List of labels set by the pipeline components, if available, keyed by component name. | -| `meta` | dict | Custom meta data for the Language class. If a model is loaded, contains meta data of the model. | -| `path` 2 | `Path` | Path to the model data directory, if a model is loaded. Otherwise `None`. | +| Name | Type | Description | +| --------------------------------------------- | ---------------------- | ---------------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | A container for the lexical types. | +| `tokenizer` | `Tokenizer` | The tokenizer. | +| `make_doc` | `Callable` | Callable that takes a string and returns a `Doc`. | +| `pipeline` | `List[str, Callable]` | List of `(name, component)` tuples describing the current processing pipeline, in order. | +| `pipe_names` 2 | `List[str]` | List of pipeline component names, in order. | +| `pipe_labels` 2.2 | `Dict[str, List[str]]` | List of labels set by the pipeline components, if available, keyed by component name. | +| `pipe_factories` 2.2 | `Dict[str, str]` | Dictionary of pipeline component names, mapped to their factory names. | +| `factory_names` 3 | `List[str]` | List of all available factory names. | +| `path` 2 | `Path` | Path to the model data directory, if a model is loaded. Otherwise `None`. | ## Class attributes {#class-attributes} -| Name | Type | Description | -| ---------- | ----- | ----------------------------------------------------------------------------------------------- | -| `Defaults` | class | Settings, data and factory methods for creating the `nlp` object and processing pipeline. | -| `lang` | str | Two-letter language ID, i.e. [ISO code](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes). | +| Name | Type | Description | +| ---------------- | ----- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `Defaults` | class | Settings, data and factory methods for creating the `nlp` object and processing pipeline. | +| `lang` | str | Two-letter language ID, i.e. [ISO code](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes). | +| `default_config` | dict | Base [config](/usage/training#config) to use for [Language.config](/api/language#config). Defaults to [`default_config.cfg`](https://github.com/explosion/spaCy/tree/develop/spacy/default_config.cfg). | ## Defaults {#defaults} diff --git a/website/docs/api/morphologizer.md b/website/docs/api/morphologizer.md index ab2b1df73..3585c76f9 100644 --- a/website/docs/api/morphologizer.md +++ b/website/docs/api/morphologizer.md @@ -10,20 +10,18 @@ coarse-grained POS tags following the Universal Dependencies [UPOS](https://universaldependencies.org/u/pos/index.html) and [FEATS](https://universaldependencies.org/format.html#morphological-annotation) annotation guidelines. This class is a subclass of `Pipe` and follows the same -API. The component is also available via the string name `"morphologizer"`. -After initialization, it is typically added to the processing pipeline using -[`nlp.add_pipe`](/api/language#add_pipe). +API. The pipeline component is available in the +[processing pipeline](/usage/processing-pipelines) via the ID `"morphologizer"`. -## Default config {#config} +## Implementation and defaults {#implementation} -This is the default configuration used to initialize the model powering the -pipeline component. See the [model architectures](/api/architectures) -documentation for details on the architectures and their arguments and -hyperparameters. To learn more about how to customize the config and train -custom models, check out the [training config](/usage/training#config) docs. +See the [model architectures](/api/architectures) documentation for details on +the architectures and their arguments and hyperparameters. To learn more about +how to customize the config and train custom models, check out the +[training config](/usage/training#config) docs. ```python -https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/morphologizer_defaults.cfg +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/morphologizer.pyx ``` ## Morphologizer.\_\_init\_\_ {#init tag="method"} @@ -33,24 +31,19 @@ Initialize the morphologizer. > #### Example > > ```python -> # Construction via create_pipe -> morphologizer = nlp.create_pipe("morphologizer") -> -> # Construction from class -> from spacy.pipeline import Morphologizer -> morphologizer = Morphologizer() +> # Construction via add_pipe +> morphologizer = nlp.add_pipe("morphologizer") > ``` - Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and -[`nlp.create_pipe`](/api/language#create_pipe). +[`nlp.add_pipe`](/api/language#add_pipe). -| Name | Type | Description | -| ----------- | -------- | ------------------------------------------------------------------------------- | -| `vocab` | `Vocab` | The shared vocabulary. | -| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | -| `**cfg` | - | Configuration parameters. | +| Name | Type | Description | +| ----------- | --------------- | ------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | The shared vocabulary. | +| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | +| `**cfg` | - | Configuration parameters. | | **RETURNS** | `Morphologizer` | The newly constructed object. | ## Morphologizer.\_\_call\_\_ {#call tag="method"} @@ -58,8 +51,8 @@ shortcut for this and instantiate the component using its string name and Apply the pipe to one document. The document is modified in place, and returned. This usually happens under the hood when the `nlp` object is called on a text and all pipeline components are applied to the `Doc` in order. Both -[`__call__`](/api/morphologizer#call) and [`pipe`](/api/morphologizer#pipe) delegate to the -[`predict`](/api/morphologizer#predict) and +[`__call__`](/api/morphologizer#call) and [`pipe`](/api/morphologizer#pipe) +delegate to the [`predict`](/api/morphologizer#predict) and [`set_annotations`](/api/morphologizer#set_annotations) methods. > #### Example @@ -81,7 +74,8 @@ and all pipeline components are applied to the `Doc` in order. Both Apply the pipe to a stream of documents. This usually happens under the hood when the `nlp` object is called on a text and all pipeline components are applied to the `Doc` in order. Both [`__call__`](/api/morphologizer#call) and -[`pipe`](/api/morphologizer#pipe) delegate to the [`predict`](/api/morphologizer#predict) and +[`pipe`](/api/morphologizer#pipe) delegate to the +[`predict`](/api/morphologizer#predict) and [`set_annotations`](/api/morphologizer#set_annotations) methods. > #### Example @@ -126,9 +120,9 @@ Modify a batch of documents, using pre-computed scores. > morphologizer.set_annotations([doc1, doc2], scores) > ``` -| Name | Type | Description | -| -------- | --------------- | ------------------------------------------------ | -| `docs` | `Iterable[Doc]` | The documents to modify. | +| Name | Type | Description | +| -------- | --------------- | ------------------------------------------------------- | +| `docs` | `Iterable[Doc]` | The documents to modify. | | `scores` | - | The scores to set, produced by `Morphologizer.predict`. | ## Morphologizer.update {#update tag="method"} @@ -145,15 +139,15 @@ pipe's model. Delegates to [`predict`](/api/morphologizer#predict) and > losses = morphologizer.update(examples, sgd=optimizer) > ``` -| Name | Type | Description | -| ----------------- | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------ | -| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | -| _keyword-only_ | | | -| `drop` | float | The dropout rate. | +| Name | Type | Description | +| ----------------- | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------- | +| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | +| _keyword-only_ | | | +| `drop` | float | The dropout rate. | | `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/morphologizer#set_annotations). | -| `sgd` | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | -| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | -| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | +| `sgd` | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | +| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | ## Morphologizer.get_loss {#get_loss tag="method"} @@ -187,12 +181,12 @@ Initialize the pipe for training, using data examples if available. Return an > optimizer = morphologizer.begin_training(pipeline=nlp.pipeline) > ``` -| Name | Type | Description | -| -------------- | ----------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- | -| `get_examples` | `Iterable[Example]` | Optional gold-standard annotations in the form of [`Example`](/api/example) objects. | -| `pipeline` | `List[(str, callable)]` | Optional list of pipeline components that this component is part of. | +| Name | Type | Description | +| -------------- | ----------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `get_examples` | `Iterable[Example]` | Optional gold-standard annotations in the form of [`Example`](/api/example) objects. | +| `pipeline` | `List[(str, callable)]` | Optional list of pipeline components that this component is part of. | | `sgd` | `Optimizer` | An optional [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. Will be created via [`create_optimizer`](/api/morphologizer#create_optimizer) if not set. | -| **RETURNS** | `Optimizer` | An optimizer. | +| **RETURNS** | `Optimizer` | An optimizer. | ## Morphologizer.create_optimizer {#create_optimizer tag="method"} @@ -237,9 +231,9 @@ both `pos` and `morph`, the label should include the UPOS as the feature `POS`. > morphologizer.add_label("Mood=Ind|POS=VERB|Tense=Past|VerbForm=Fin") > ``` -| Name | Type | Description | -| -------- | ---- | --------------------------------------------------------------- | -| `label` | str | The label to add. | +| Name | Type | Description | +| ------- | ---- | ----------------- | +| `label` | str | The label to add. | ## Morphologizer.to_disk {#to_disk tag="method"} @@ -268,11 +262,11 @@ Load the pipe from disk. Modifies the object in place and returns it. > morphologizer.from_disk("/path/to/morphologizer") > ``` -| Name | Type | Description | -| ----------- | ------------ | -------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | `Morphologizer` | The modified `Morphologizer` object. | +| Name | Type | Description | +| ----------- | --------------- | -------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | +| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | `Morphologizer` | The modified `Morphologizer` object. | ## Morphologizer.to_bytes {#to_bytes tag="method"} @@ -288,7 +282,7 @@ Serialize the pipe to a bytestring. | Name | Type | Description | | ----------- | ----- | ------------------------------------------------------------------------- | | `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | bytes | The serialized form of the `Morphologizer` object. | +| **RETURNS** | bytes | The serialized form of the `Morphologizer` object. | ## Morphologizer.from_bytes {#from_bytes tag="method"} @@ -302,16 +296,16 @@ Load the pipe from a bytestring. Modifies the object in place and returns it. > morphologizer.from_bytes(morphologizer_bytes) > ``` -| Name | Type | Description | -| ------------ | -------- | ------------------------------------------------------------------------- | -| `bytes_data` | bytes | The data to load from. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | `Morphologizer` | The `Morphologizer` object. | +| Name | Type | Description | +| ------------ | --------------- | ------------------------------------------------------------------------- | +| `bytes_data` | bytes | The data to load from. | +| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | `Morphologizer` | The `Morphologizer` object. | ## Morphologizer.labels {#labels tag="property"} -The labels currently added to the component in Universal Dependencies [FEATS -format](https://universaldependencies.org/format.html#morphological-annotation). +The labels currently added to the component in Universal Dependencies +[FEATS format](https://universaldependencies.org/format.html#morphological-annotation). Note that even for a blank component, this will always include the internal empty label `_`. If POS features are used, the labels will include the coarse-grained POS as the feature `POS`. @@ -339,8 +333,8 @@ serialization by passing in the string names via the `exclude` argument. > data = morphologizer.to_disk("/path", exclude=["vocab"]) > ``` -| Name | Description | -| --------- | ------------------------------------------------------------------------------------------ | -| `vocab` | The shared [`Vocab`](/api/vocab). | -| `cfg` | The config file. You usually don't want to exclude this. | -| `model` | The binary model data. You usually don't want to exclude this. | +| Name | Description | +| ------- | -------------------------------------------------------------- | +| `vocab` | The shared [`Vocab`](/api/vocab). | +| `cfg` | The config file. You usually don't want to exclude this. | +| `model` | The binary model data. You usually don't want to exclude this. | diff --git a/website/docs/api/pipeline-functions.md b/website/docs/api/pipeline-functions.md index fc417845c..5c2eb2b97 100644 --- a/website/docs/api/pipeline-functions.md +++ b/website/docs/api/pipeline-functions.md @@ -11,8 +11,7 @@ menu: ## merge_noun_chunks {#merge_noun_chunks tag="function"} Merge noun chunks into a single token. Also available via the string name -`"merge_noun_chunks"`. After initialization, the component is typically added to -the processing pipeline using [`nlp.add_pipe`](/api/language#add_pipe). +`"merge_noun_chunks"`. > #### Example > @@ -20,9 +19,7 @@ the processing pipeline using [`nlp.add_pipe`](/api/language#add_pipe). > texts = [t.text for t in nlp("I have a blue car")] > assert texts == ["I", "have", "a", "blue", "car"] > -> merge_nps = nlp.create_pipe("merge_noun_chunks") -> nlp.add_pipe(merge_nps) -> +> nlp.add_pipe("merge_noun_chunks") > texts = [t.text for t in nlp("I have a blue car")] > assert texts == ["I", "have", "a blue car"] > ``` @@ -44,8 +41,7 @@ all other components. ## merge_entities {#merge_entities tag="function"} Merge named entities into a single token. Also available via the string name -`"merge_entities"`. After initialization, the component is typically added to -the processing pipeline using [`nlp.add_pipe`](/api/language#add_pipe). +`"merge_entities"`. > #### Example > @@ -53,8 +49,7 @@ the processing pipeline using [`nlp.add_pipe`](/api/language#add_pipe). > texts = [t.text for t in nlp("I like David Bowie")] > assert texts == ["I", "like", "David", "Bowie"] > -> merge_ents = nlp.create_pipe("merge_entities") -> nlp.add_pipe(merge_ents) +> nlp.add_pipe("merge_entities") > > texts = [t.text for t in nlp("I like David Bowie")] > assert texts == ["I", "like", "David Bowie"] @@ -76,12 +71,9 @@ components to the end of the pipeline and after all other components. ## merge_subtokens {#merge_subtokens tag="function" new="2.1"} Merge subtokens into a single token. Also available via the string name -`"merge_subtokens"`. After initialization, the component is typically added to -the processing pipeline using [`nlp.add_pipe`](/api/language#add_pipe). - -As of v2.1, the parser is able to predict "subtokens" that should be merged into -one single token later on. This is especially relevant for languages like -Chinese, Japanese or Korean, where a "word" isn't defined as a +`"merge_subtokens"`. As of v2.1, the parser is able to predict "subtokens" that +should be merged into one single token later on. This is especially relevant for +languages like Chinese, Japanese or Korean, where a "word" isn't defined as a whitespace-delimited sequence of characters. Under the hood, this component uses the [`Matcher`](/api/matcher) to find sequences of tokens with the dependency label `"subtok"` and then merges them into a single token. @@ -96,9 +88,7 @@ label `"subtok"` and then merges them into a single token. > print([(token.text, token.dep_) for token in doc]) > # [('拜', 'subtok'), ('托', 'subtok')] > -> merge_subtok = nlp.create_pipe("merge_subtokens") -> nlp.add_pipe(merge_subtok) -> +> nlp.add_pipe("merge_subtokens") > doc = nlp("拜托") > print([token.text for token in doc]) > # ['拜托'] diff --git a/website/docs/api/sentencerecognizer.md b/website/docs/api/sentencerecognizer.md index 458e42975..2cea97aaf 100644 --- a/website/docs/api/sentencerecognizer.md +++ b/website/docs/api/sentencerecognizer.md @@ -1,26 +1,24 @@ --- title: SentenceRecognizer tag: class -source: spacy/pipeline/pipes.pyx +source: spacy/pipeline/senter.pyx new: 3 --- A trainable pipeline component for sentence segmentation. For a simpler, ruse-based strategy, see the [`Sentencizer`](/api/sentencizer). This class is a subclass of `Pipe` and follows the same API. The component is also available via -the string name `"senter"`. After initialization, it is typically added to the -processing pipeline using [`nlp.add_pipe`](/api/language#add_pipe). +the string name `"senter"`. -## Default config {#config} +## Implementation and defaults {#implementation} -This is the default configuration used to initialize the model powering the -pipeline component. See the [model architectures](/api/architectures) -documentation for details on the architectures and their arguments and -hyperparameters. To learn more about how to customize the config and train -custom models, check out the [training config](/usage/training#config) docs. +See the [model architectures](/api/architectures) documentation for details on +the architectures and their arguments and hyperparameters. To learn more about +how to customize the config and train custom models, check out the +[training config](/usage/training#config) docs. ```python -https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/senter_defaults.cfg +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/senter.pyx ``` ## SentenceRecognizer.\_\_init\_\_ {#init tag="method"} @@ -30,12 +28,8 @@ Initialize the sentence recognizer. > #### Example > > ```python -> # Construction via create_pipe -> senter = nlp.create_pipe("senter") -> -> # Construction from class -> from spacy.pipeline import SentenceRecognizer -> senter = SentenceRecognizer() +> # Construction via add_pipe +> senter = nlp.add_pipe("senter") > ``` diff --git a/website/docs/api/sentencizer.md b/website/docs/api/sentencizer.md index 9c6e2d58c..181babe06 100644 --- a/website/docs/api/sentencizer.md +++ b/website/docs/api/sentencizer.md @@ -9,8 +9,7 @@ that doesn't require the dependency parse. By default, sentence segmentation is performed by the [`DependencyParser`](/api/dependencyparser), so the `Sentencizer` lets you implement a simpler, rule-based strategy that doesn't require a statistical model to be loaded. The component is also available via -the string name `"sentencizer"`. After initialization, it is typically added to -the processing pipeline using [`nlp.add_pipe`](/api/language#add_pipe). +the string name `"sentencizer"`. ## Sentencizer.\_\_init\_\_ {#init tag="method"} @@ -19,12 +18,8 @@ Initialize the sentencizer. > #### Example > > ```python -> # Construction via create_pipe -> sentencizer = nlp.create_pipe("sentencizer") -> -> # Construction from class -> from spacy.pipeline import Sentencizer -> sentencizer = Sentencizer() +> # Construction via add_pipe +> sentencizer = nlp.add_pipe("sentencizer") > ``` | Name | Type | Description | @@ -58,8 +53,7 @@ the component has been added to the pipeline using > from spacy.lang.en import English > > nlp = English() -> sentencizer = nlp.create_pipe("sentencizer") -> nlp.add_pipe(sentencizer) +> nlp.add_pipe("sentencizer") > doc = nlp("This is a sentence. This is another sentence.") > assert len(list(doc.sents)) == 2 > ``` diff --git a/website/docs/api/tagger.md b/website/docs/api/tagger.md index 9ef0843cf..82658b8e0 100644 --- a/website/docs/api/tagger.md +++ b/website/docs/api/tagger.md @@ -1,7 +1,7 @@ --- title: Tagger tag: class -source: spacy/pipeline/pipes.pyx +source: spacy/pipeline/tagger.pyx --- This class is a subclass of `Pipe` and follows the same API. The pipeline @@ -13,22 +13,17 @@ via the ID `"tagger"`. > #### Example > > ```python -> # Construction via create_pipe -> tagger = nlp.create_pipe("tagger") +> # Construction via add_pipe with default model +> tagger = nlp.add_pipe("tagger") > > # Construction via create_pipe with custom model > config = {"model": {"@architectures": "my_tagger"}} -> parser = nlp.create_pipe("tagger", config) -> -> # Construction from class with custom model from file -> from spacy.pipeline import Tagger -> model = util.load_config("model.cfg", create_objects=True)["model"] -> tagger = Tagger(nlp.vocab, model) +> parser = nlp.add_pipe("tagger", config) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and -[`nlp.create_pipe`](/api/language#create_pipe). +[`nlp.add_pipe`](/api/language#add_pipe). | Name | Type | Description | | ----------- | -------- | ------------------------------------------------------------------------------- | diff --git a/website/docs/api/textcategorizer.md b/website/docs/api/textcategorizer.md index 431ee683b..52b7e399c 100644 --- a/website/docs/api/textcategorizer.md +++ b/website/docs/api/textcategorizer.md @@ -1,7 +1,7 @@ --- title: TextCategorizer tag: class -source: spacy/pipeline/pipes.pyx +source: spacy/pipeline/textcat.py new: 2 --- @@ -9,41 +9,33 @@ This class is a subclass of `Pipe` and follows the same API. The pipeline component is available in the [processing pipeline](/usage/processing-pipelines) via the ID `"textcat"`. -## Default config {#config} +## Implementation and defaults {#implementation} -This is the default configuration used to initialize the model powering the -pipeline component. See the [model architectures](/api/architectures) -documentation for details on the architectures and their arguments and -hyperparameters. To learn more about how to customize the config and train -custom models, check out the [training config](/usage/training#config) docs. +See the [model architectures](/api/architectures) documentation for details on +the architectures and their arguments and hyperparameters. To learn more about +how to customize the config and train custom models, check out the +[training config](/usage/training#config) docs. ```python -https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/textcat_defaults.cfg +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/textcat.py ``` - - ## TextCategorizer.\_\_init\_\_ {#init tag="method"} > #### Example > > ```python -> # Construction via create_pipe -> textcat = nlp.create_pipe("textcat") +> # Construction via add_pipe with default model +> textcat = nlp.add_pipe("textcat") > -> # Construction via create_pipe with custom model +> # Construction via add_pipe with custom model > config = {"model": {"@architectures": "my_textcat"}} -> parser = nlp.create_pipe("textcat", config) -> -> # Construction from class with custom model from file -> from spacy.pipeline import TextCategorizer -> model = util.load_config("model.cfg", create_objects=True)["model"] -> textcat = TextCategorizer(nlp.vocab, model) +> parser = nlp.add_pipe("textcat", config=config) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and -[`nlp.create_pipe`](/api/language#create_pipe). +[`nlp.add_pipe`](/api/language#create_pipe). | Name | Type | Description | | ----------- | ----------------- | ------------------------------------------------------------------------------- | diff --git a/website/docs/api/tok2vec.md b/website/docs/api/tok2vec.md index 3667ed8ad..34e6e267a 100644 --- a/website/docs/api/tok2vec.md +++ b/website/docs/api/tok2vec.md @@ -4,16 +4,15 @@ source: spacy/pipeline/tok2vec.py new: 3 --- -TODO: document + -## Default config {#config} +## Implementation and defaults {#implementation} -This is the default configuration used to initialize the model powering the -pipeline component. See the [model architectures](/api/architectures) -documentation for details on the architectures and their arguments and -hyperparameters. To learn more about how to customize the config and train -custom models, check out the [training config](/usage/training#config) docs. +See the [model architectures](/api/architectures) documentation for details on +the architectures and their arguments and hyperparameters. To learn more about +how to customize the config and train custom models, check out the +[training config](/usage/training#config) docs. ```python -https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/defaults/tok2vec_defaults.cfg +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/tok2vec.py ``` diff --git a/website/docs/api/tokenizer.md b/website/docs/api/tokenizer.md index 3281275e1..47e5aa9b3 100644 --- a/website/docs/api/tokenizer.md +++ b/website/docs/api/tokenizer.md @@ -31,7 +31,7 @@ the > nlp = English() > # Create a Tokenizer with the default settings for English > # including punctuation rules and exceptions -> tokenizer = nlp.Defaults.create_tokenizer(nlp) +> tokenizer = nlp.tokenizer > ``` | Name | Type | Description | diff --git a/website/docs/api/top-level.md b/website/docs/api/top-level.md index 8e9fff6aa..9c4380626 100644 --- a/website/docs/api/top-level.md +++ b/website/docs/api/top-level.md @@ -45,7 +45,8 @@ class, loads in the model data and returns it. ### Abstract example cls = util.get_lang_class(lang) # get language for ID, e.g. 'en' nlp = cls() # initialise the language -for name in pipeline: component = nlp.create_pipe(name) # create each pipeline component nlp.add_pipe(component) # add component to pipeline +for name in pipeline: + nlp.add_pipe(name) # add component to pipeline nlp.from_disk(model_data_path) # load in model data ``` @@ -479,7 +480,6 @@ you can use the [`set_lang_class`](/api/top-level#util.set_lang_class) helper. > for lang_id in ["en", "de"]: > lang_class = util.get_lang_class(lang_id) > lang = lang_class() -> tokenizer = lang.Defaults.create_tokenizer() > ``` | Name | Type | Description | diff --git a/website/docs/images/pipeline.svg b/website/docs/images/pipeline.svg index 022219c5f..9ece70e6f 100644 --- a/website/docs/images/pipeline.svg +++ b/website/docs/images/pipeline.svg @@ -1,30 +1,33 @@ - - - - - Doc - - - - Text - - - - nlp - - tokenizer - - tagger - - - - parser - - ner - - ... + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + diff --git a/website/docs/images/training.svg b/website/docs/images/training.svg index 0a4649c48..4d6bcc0cc 100644 --- a/website/docs/images/training.svg +++ b/website/docs/images/training.svg @@ -1,47 +1,60 @@ - - - - - - + + + + + - - - - - - PREDICT - - - - SAVE - - Model - - - - - - Training data - - - - label - - - - label - - Updated - Model - - - text - - - - GRADIENT + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + diff --git a/website/docs/usage/101/_architecture.md b/website/docs/usage/101/_architecture.md index d2e1aee39..d321b7cb9 100644 --- a/website/docs/usage/101/_architecture.md +++ b/website/docs/usage/101/_architecture.md @@ -18,13 +18,13 @@ an **annotated document**. It also orchestrates training and serialization. ### Container objects {#architecture-containers} -| Name | Description | -| ----------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- | -| [`Doc`](/api/doc) | A container for accessing linguistic annotations. | -| [`Span`](/api/span) | A slice from a `Doc` object. | -| [`Token`](/api/token) | An individual token — i.e. a word, punctuation symbol, whitespace, etc. | -| [`Lexeme`](/api/lexeme) | An entry in the vocabulary. It's a word type with no context, as opposed to a word token. It therefore has no part-of-speech tag, dependency parse etc. | -| [`MorphAnalysis`](/api/morphanalysis) | A morphological analysis. | +| Name | Description | +| ------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- | +| [`Doc`](/api/doc) | A container for accessing linguistic annotations. | +| [`Span`](/api/span) | A slice from a `Doc` object. | +| [`Token`](/api/token) | An individual token — i.e. a word, punctuation symbol, whitespace, etc. | +| [`Lexeme`](/api/lexeme) | An entry in the vocabulary. It's a word type with no context, as opposed to a word token. It therefore has no part-of-speech tag, dependency parse etc. | +| [`MorphAnalysis`](/api/morphanalysis) | A morphological analysis. | ### Processing pipeline {#architecture-pipeline} @@ -52,5 +52,3 @@ an **annotated document**. It also orchestrates training and serialization. | [`StringStore`](/api/stringstore) | Map strings to and from hash values. | | [`Vectors`](/api/vectors) | Container class for vector data keyed by string. | | [`Example`](/api/example) | Collection for training annotations. | - -| diff --git a/website/docs/usage/101/_pipelines.md b/website/docs/usage/101/_pipelines.md index d33ea45fd..1ea165515 100644 --- a/website/docs/usage/101/_pipelines.md +++ b/website/docs/usage/101/_pipelines.md @@ -12,29 +12,32 @@ passed on to the next component. > - **Creates:** Objects, attributes and properties modified and set by the > component. -| Name | Component | Creates | Description | -| ----------------- | ------------------------------------------------------------------ | ----------------------------------------------------------- | ------------------------------------------------ | -| **tokenizer** | [`Tokenizer`](/api/tokenizer) | `Doc` | Segment text into tokens. | -| **tagger** | [`Tagger`](/api/tagger) | `Doc[i].tag` | Assign part-of-speech tags. | -| **parser** | [`DependencyParser`](/api/dependencyparser) | `Doc[i].head`, `Doc[i].dep`, `Doc.sents`, `Doc.noun_chunks` | Assign dependency labels. | -| **ner** | [`EntityRecognizer`](/api/entityrecognizer) | `Doc.ents`, `Doc[i].ent_iob`, `Doc[i].ent_type` | Detect and label named entities. | -| **textcat** | [`TextCategorizer`](/api/textcategorizer) | `Doc.cats` | Assign document labels. | -| ... | [custom components](/usage/processing-pipelines#custom-components) | `Doc._.xxx`, `Token._.xxx`, `Span._.xxx` | Assign custom attributes, methods or properties. | +| Name | Component | Creates | Description | +| ------------- | ------------------------------------------------------------------ | ----------------------------------------------------------- | ------------------------------------------------ | +| **tokenizer** | [`Tokenizer`](/api/tokenizer) | `Doc` | Segment text into tokens. | +| **tagger** | [`Tagger`](/api/tagger) | `Doc[i].tag` | Assign part-of-speech tags. | +| **parser** | [`DependencyParser`](/api/dependencyparser) | `Doc[i].head`, `Doc[i].dep`, `Doc.sents`, `Doc.noun_chunks` | Assign dependency labels. | +| **ner** | [`EntityRecognizer`](/api/entityrecognizer) | `Doc.ents`, `Doc[i].ent_iob`, `Doc[i].ent_type` | Detect and label named entities. | +| **textcat** | [`TextCategorizer`](/api/textcategorizer) | `Doc.cats` | Assign document labels. | +| ... | [custom components](/usage/processing-pipelines#custom-components) | `Doc._.xxx`, `Token._.xxx`, `Span._.xxx` | Assign custom attributes, methods or properties. | The processing pipeline always **depends on the statistical model** and its capabilities. For example, a pipeline can only include an entity recognizer component if the model includes data to make predictions of entity labels. This -is why each model will specify the pipeline to use in its meta data, as a simple -list containing the component names: +is why each model will specify the pipeline to use in its meta data and +[config](/usage/training#config), as a simple list containing the component +names: -```json -"pipeline": ["tagger", "parser", "ner"] +```ini +pipeline = ["tagger", "parser", "ner"] ``` import Accordion from 'components/accordion.js' + + In spaCy v2.x, the statistical components like the tagger or parser are independent and don't share any data between themselves. For example, the named entity recognizer doesn't use any features set by the tagger and parser, and so @@ -48,11 +51,10 @@ pre-defined sentence boundaries, so if a previous component in the pipeline sets them, its dependency predictions may be different. Similarly, it matters if you add the [`EntityRuler`](/api/entityruler) before or after the statistical entity recognizer: if it's added before, the entity recognizer will take the existing -entities into account when making predictions. -The [`EntityLinker`](/api/entitylinker), which resolves named entities to -knowledge base IDs, should be preceded by -a pipeline component that recognizes entities such as the -[`EntityRecognizer`](/api/entityrecognizer). +entities into account when making predictions. The +[`EntityLinker`](/api/entitylinker), which resolves named entities to knowledge +base IDs, should be preceded by a pipeline component that recognizes entities +such as the [`EntityRecognizer`](/api/entityrecognizer). diff --git a/website/docs/usage/linguistic-features.md b/website/docs/usage/linguistic-features.md index 18c33c7bb..881a0e333 100644 --- a/website/docs/usage/linguistic-features.md +++ b/website/docs/usage/linguistic-features.md @@ -909,9 +909,8 @@ If you're using a statistical model, writing to the `nlp.Defaults` or `English.Defaults` directly won't work, since the regular expressions are read from the model and will be compiled when you load it. If you modify `nlp.Defaults`, you'll only see the effect if you call -[`spacy.blank`](/api/top-level#spacy.blank) or `Defaults.create_tokenizer()`. If -you want to modify the tokenizer loaded from a statistical model, you should -modify `nlp.tokenizer` directly. +[`spacy.blank`](/api/top-level#spacy.blank). If you want to modify the tokenizer +loaded from a statistical model, you should modify `nlp.tokenizer` directly. @@ -1386,8 +1385,7 @@ import spacy from spacy.lang.en import English nlp = English() # just the language with no model -sentencizer = nlp.create_pipe("sentencizer") -nlp.add_pipe(sentencizer) +nlp.add_pipe("sentencizer") doc = nlp("This is a sentence. This is another sentence.") for sent in doc.sents: print(sent.text) @@ -1422,6 +1420,7 @@ take advantage of dependency-based sentence segmentation. ```python ### {executable="true"} +from spacy.language import Language import spacy text = "this is a sentence...hello...and another sentence." @@ -1430,13 +1429,14 @@ nlp = spacy.load("en_core_web_sm") doc = nlp(text) print("Before:", [sent.text for sent in doc.sents]) +@Language.component("set_custom_coundaries") def set_custom_boundaries(doc): for token in doc[:-1]: if token.text == "...": - doc[token.i+1].is_sent_start = True + doc[token.i + 1].is_sent_start = True return doc -nlp.add_pipe(set_custom_boundaries, before="parser") +nlp.add_pipe("set_custom_boundaries", before="parser") doc = nlp(text) print("After:", [sent.text for sent in doc.sents]) ``` diff --git a/website/docs/usage/processing-pipelines.md b/website/docs/usage/processing-pipelines.md index 5679c318e..08e8e964f 100644 --- a/website/docs/usage/processing-pipelines.md +++ b/website/docs/usage/processing-pipelines.md @@ -97,32 +97,40 @@ but also your own custom processing functions. A pipeline component can be added to an already existing `nlp` object, specified when initializing a `Language` class, or defined within a [model package](/usage/saving-loading#models). -When you load a model, spaCy first consults the model's -[`meta.json`](/usage/saving-loading#models). The meta typically includes the -model details, the ID of a language class, and an optional list of pipeline -components. spaCy then does the following: - -> #### meta.json (excerpt) +> #### config.cfg (excerpt) > -> ```json -> { -> "lang": "en", -> "name": "core_web_sm", -> "description": "Example model for spaCy", -> "pipeline": ["tagger", "parser", "ner"] -> } +> ```ini +> [nlp] +> lang = "en" +> pipeline = ["tagger", "parser"] +> +> [components] +> +> [components.tagger] +> factory = "tagger" +> # settings for the tagger component +> +> [components.parser] +> factory = "parser" +> # settings for the parser component > ``` +When you load a model, spaCy first consults the model's +[`meta.json`](/usage/saving-loading#models) and +[`config.cfg`](/usage/training#config). The config tells spaCy what language +class to use, which components are in the pipeline, and how those components +should be created. spaCy will then do the following: + 1. Load the **language class and data** for the given ID via [`get_lang_class`](/api/top-level#util.get_lang_class) and initialize it. The `Language` class contains the shared vocabulary, tokenization rules and the - language-specific annotation scheme. -2. Iterate over the **pipeline names** and create each component using - [`create_pipe`](/api/language#create_pipe), which looks them up in - `Language.factories`. -3. Add each pipeline component to the pipeline in order, using - [`add_pipe`](/api/language#add_pipe). -4. Make the **model data** available to the `Language` class by calling + language-specific settings. +2. Iterate over the **pipeline names** and look up each component name in the + `[components]` block. The `factory` tells spaCy which + [component factory](#custom-components-factories) to use for adding the + component with with [`add_pipe`](/api/language#add_pipe). The settings are + passed into the factory. +3. Make the **model data** available to the `Language` class by calling [`from_disk`](/api/language#from_disk) with the path to the model data directory. @@ -132,17 +140,25 @@ So when you call this... nlp = spacy.load("en_core_web_sm") ``` -... the model's `meta.json` tells spaCy to use the language `"en"` and the +... the model's `config.cfg` tells spaCy to use the language `"en"` and the pipeline `["tagger", "parser", "ner"]`. spaCy will then initialize `spacy.lang.en.English`, and create each pipeline component and add it to the processing pipeline. It'll then load in the model's data from its data directory and return the modified `Language` class for you to use as the `nlp` object. + + +spaCy v3.0 introduces a `config.cfg`, which includes more detailed settings for +the model pipeline, its components and the +[training process](/usage/training#config). You can export the config of your +current `nlp` object by calling [`nlp.config.to_disk`](/api/language#config). + + + Fundamentally, a [spaCy model](/models) consists of three components: **the weights**, i.e. binary data loaded in from a directory, a **pipeline** of functions called in order, and **language data** like the tokenization rules and -annotation scheme. All of this is specific to each model, and defined in the -model's `meta.json` – for example, a Spanish NER model requires different +language-specific settings. For example, a Spanish NER model requires different weights, language data and pipeline components than an English parsing and tagging model. This is also why the pipeline state is always held by the `Language` class. [`spacy.load`](/api/top-level#spacy.load) puts this all @@ -158,9 +174,8 @@ data_path = "path/to/en_core_web_sm/en_core_web_sm-2.0.0" cls = spacy.util.get_lang_class(lang) # 1. Get Language instance, e.g. English() nlp = cls() # 2. Initialize it for name in pipeline: - component = nlp.create_pipe(name) # 3. Create the pipeline components - nlp.add_pipe(component) # 4. Add the component to the pipeline -nlp.from_disk(model_data_path) # 5. Load in the binary data + nlp.add_pipe(name) # 3. Add the component to the pipeline +nlp.from_disk(model_data_path) # 4. Load in the binary data ``` When you call `nlp` on a text, spaCy will **tokenize** it and then **call each @@ -190,36 +205,34 @@ print(nlp.pipe_names) ### Built-in pipeline components {#built-in} -spaCy ships with several built-in pipeline components that are also available in -the `Language.factories`. This means that you can initialize them by calling -[`nlp.create_pipe`](/api/language#create_pipe) with their string names and -require them in the pipeline settings in your model's `meta.json`. +spaCy ships with several built-in pipeline components that are registered with +string names. This means that you can initialize them by calling +[`nlp.add_pipe`](/api/language#add_pipe) with their names and spaCy will know +how to create them. See the [API documentation](/api) for a full list of +available pipeline components and component functions. > #### Usage > > ```python -> # Option 1: Import and initialize -> from spacy.pipeline import EntityRuler -> ruler = EntityRuler(nlp) -> nlp.add_pipe(ruler) -> -> # Option 2: Using nlp.create_pipe -> sentencizer = nlp.create_pipe("sentencizer") -> nlp.add_pipe(sentencizer) +> nlp = spacy.blank("en") +> nlp.add_pipe("sentencizer") +> # add_pipe returns the added component +> ruler = nlp.add_pipe("entity_ruler") > ``` -| String name | Component | Description | -| ------------------- | ---------------------------------------------------------------- | --------------------------------------------------------------------------------------------- | -| `tagger` | [`Tagger`](/api/tagger) | Assign part-of-speech-tags. | -| `parser` | [`DependencyParser`](/api/dependencyparser) | Assign dependency labels. | -| `ner` | [`EntityRecognizer`](/api/entityrecognizer) | Assign named entities. | -| `entity_linker` | [`EntityLinker`](/api/entitylinker) | Assign knowledge base IDs to named entities. Should be added after the entity recognizer. | -| `textcat` | [`TextCategorizer`](/api/textcategorizer) | Assign text categories. | -| `entity_ruler` | [`EntityRuler`](/api/entityruler) | Assign named entities based on pattern rules. | -| `sentencizer` | [`Sentencizer`](/api/sentencizer) | Add rule-based sentence segmentation without the dependency parse. | -| `merge_noun_chunks` | [`merge_noun_chunks`](/api/pipeline-functions#merge_noun_chunks) | Merge all noun chunks into a single token. Should be added after the tagger and parser. | -| `merge_entities` | [`merge_entities`](/api/pipeline-functions#merge_entities) | Merge all entities into a single token. Should be added after the entity recognizer. | -| `merge_subtokens` | [`merge_subtokens`](/api/pipeline-functions#merge_subtokens) | Merge subtokens predicted by the parser into single tokens. Should be added after the parser. | +| String name | Component | Description | +| --------------- | ------------------------------------------- | ----------------------------------------------------------------------------------------- | +| `tagger` | [`Tagger`](/api/tagger) | Assign part-of-speech-tags. | +| `parser` | [`DependencyParser`](/api/dependencyparser) | Assign dependency labels. | +| `ner` | [`EntityRecognizer`](/api/entityrecognizer) | Assign named entities. | +| `entity_linker` | [`EntityLinker`](/api/entitylinker) | Assign knowledge base IDs to named entities. Should be added after the entity recognizer. | +| `textcat` | [`TextCategorizer`](/api/textcategorizer) | Assign text categories. | +| `entity_ruler` | [`EntityRuler`](/api/entityruler) | Assign named entities based on pattern rules. | +| `sentencizer` | [`Sentencizer`](/api/sentencizer) | Add rule-based sentence segmentation without the dependency parse. | + + + + ### Disabling and modifying pipeline components {#disabling} @@ -233,7 +246,6 @@ list: ```python ### Disable loading nlp = spacy.load("en_core_web_sm", disable=["tagger", "parser"]) -nlp = English().from_disk("/model", disable=["ner"]) ``` In some cases, you do want to load all pipeline components and their weights, @@ -297,15 +309,18 @@ nlp.replace_pipe("tagger", my_custom_tagger) ## Creating custom pipeline components {#custom-components} -A component receives a `Doc` object and can modify it – for example, by using -the current weights to make a prediction and set some annotation on the -document. By adding a component to the pipeline, you'll get access to the `Doc` -at any point **during processing** – instead of only being able to modify it -afterwards. +A pipeline component is a function that receives a `Doc` object, modifies it and +returns it – – for example, by using the current weights to make a prediction +and set some annotation on the document. By adding a component to the pipeline, +you'll get access to the `Doc` at any point **during processing** – instead of +only being able to modify it afterwards. > #### Example > > ```python +> from spacy.language import Language +> +> @Language.component("my_component") > def my_component(doc): > # do something to the doc here > return doc @@ -316,6 +331,12 @@ afterwards. | `doc` | `Doc` | The `Doc` object processed by the previous component. | | **RETURNS** | `Doc` | The `Doc` object processed by this pipeline component. | +The [`@Language.component`](/api/language#component) decorator lets you turn a +simple function into a pipeline component. It takes at least one argument, the +**name** of the component factory. You can use this name to add an instance of +your component to the pipeline. It can also be listed in your model config, so +you can save, load and train models using your component. + Custom components can be added to the pipeline using the [`add_pipe`](/api/language#add_pipe) method. Optionally, you can either specify a component to add it **before or after**, tell spaCy to add it **first or @@ -325,23 +346,43 @@ last** in the pipeline, or define a **custom name**. If no name is set and no > #### Example > > ```python -> nlp.add_pipe(my_component) -> nlp.add_pipe(my_component, first=True) -> nlp.add_pipe(my_component, before="parser") +> nlp.add_pipe("my_component") +> nlp.add_pipe("my_component", first=True) +> nlp.add_pipe("my_component", before="parser") > ``` -| Argument | Type | Description | -| -------- | ---- | ------------------------------------------------------------------------ | -| `last` | bool | If set to `True`, component is added **last** in the pipeline (default). | -| `first` | bool | If set to `True`, component is added **first** in the pipeline. | -| `before` | str | String name of component to add the new component **before**. | -| `after` | str | String name of component to add the new component **after**. | +| Argument | Type | Description | +| -------- | --------- | ------------------------------------------------------------------------ | +| `last` | bool | If set to `True`, component is added **last** in the pipeline (default). | +| `first` | bool | If set to `True`, component is added **first** in the pipeline. | +| `before` | str / int | String name or index to add the new component **before**. | +| `after` | str / int | String name or index to add the new component **after**. | -### Example: A simple pipeline component {#custom-components-simple} + + +As of v3.0, components need to be registered using the +[`@Language.component`](/api/language#component) or +[`@Language.factory`](/api/language#factory) decorator so spaCy knows that a +function is a component. [`nlp.add_pipe`](/api/language#add_pipe) now takes the +**string name** of the component factory instead of the component function. This +doesn't only save you lines of code, it also allows spaCy to validate and track +your custom components, and make sure they can be saved and loaded. + +```diff +- ruler = nlp.create_pipe("entity_ruler") +- nlp.add_pipe(ruler) ++ ruler = nlp.add_pipe("entity_ruler") +``` + + + +### Examples: Simple stateless pipeline components {#custom-components-simple} The following component receives the `Doc` in the pipeline and prints some information about it: the number of tokens, the part-of-speech tags of the -tokens and a conditional message based on the document length. +tokens and a conditional message based on the document length. The +[`@Language.component`](/api/language#component) decorator lets you register the +component under the name `"info_component"`. > #### ✏️ Things to try > @@ -352,11 +393,16 @@ tokens and a conditional message based on the document length. > this change reflected in `nlp.pipe_names`. > 3. Print `nlp.pipeline`. You'll see a list of tuples describing the component > name and the function that's called on the `Doc` object in the pipeline. +> 4. Change the first argument to `@Language.component`, the name, to something +> else. spaCy should now complain that it doesn't know a component of the +> name `"info_component"`. ```python ### {executable="true"} import spacy +from spacy.language import Language +@Language.component("info_component") def my_component(doc): print(f"After tokenization, this doc has {len(doc)} tokens.") print("The part-of-speech tags are:", [token.pos_ for token in doc]) @@ -365,76 +411,16 @@ def my_component(doc): return doc nlp = spacy.load("en_core_web_sm") -nlp.add_pipe(my_component, name="print_info", last=True) +nlp.add_pipe("info_component", name="print_info", last=True) print(nlp.pipe_names) # ['tagger', 'parser', 'ner', 'print_info'] doc = nlp("This is a sentence.") - ``` -Of course, you can also wrap your component as a class to allow initializing it -with custom settings and hold state within the component. This is useful for -**stateful components**, especially ones which **depend on shared data**. In the -following example, the custom component `EntityMatcher` can be initialized with -`nlp` object, a terminology list and an entity label. Using the -[`PhraseMatcher`](/api/phrasematcher), it then matches the terms in the `Doc` -and adds them to the existing entities. - - - -As of v2.1.0, spaCy ships with the [`EntityRuler`](/api/entityruler), a pipeline -component for easy, rule-based named entity recognition. Its implementation is -similar to the `EntityMatcher` code shown below, but it includes some additional -features like support for phrase patterns and token patterns, handling overlaps -with existing entities and pattern export as JSONL. - -We'll still keep the pipeline component example below, as it works well to -illustrate complex components. But if you're planning on using this type of -component in your application, you might find the `EntityRuler` more convenient. -[See here](/usage/rule-based-matching#entityruler) for more details and -examples. - - - -```python -### {executable="true"} -import spacy -from spacy.matcher import PhraseMatcher -from spacy.tokens import Span - -class EntityMatcher: - name = "entity_matcher" - - def __init__(self, nlp, terms, label): - patterns = [nlp.make_doc(text) for text in terms] - self.matcher = PhraseMatcher(nlp.vocab) - self.matcher.add(label, patterns) - - def __call__(self, doc): - matches = self.matcher(doc) - for match_id, start, end in matches: - span = Span(doc, start, end, label=match_id) - doc.ents = list(doc.ents) + [span] - return doc - -nlp = spacy.load("en_core_web_sm") -terms = ("cat", "dog", "tree kangaroo", "giant sea spider") -entity_matcher = EntityMatcher(nlp, terms, "ANIMAL") - -nlp.add_pipe(entity_matcher, after="ner") - -print(nlp.pipe_names) # The components in the pipeline - -doc = nlp("This is a text about Barack Obama and a tree kangaroo") -print([(ent.text, ent.label_) for ent in doc.ents]) -``` - -### Example: Custom sentence segmentation logic {#component-example1} - -Let's say you want to implement custom logic to improve spaCy's sentence -boundary detection. Currently, sentence segmentation is based on the dependency -parse, which doesn't always produce ideal results. The custom logic should -therefore be applied **after** tokenization, but _before_ the dependency parsing -– this way, the parser can also take advantage of the sentence boundaries. +Here's another example of a pipeline component that implements custom logic to +improve the sentence boundaries set by the dependency parser. The custom logic +should therefore be applied **after** tokenization, but _before_ the dependency +parsing – this way, the parser can also take advantage of the sentence +boundaries. > #### ✏️ Things to try > @@ -448,90 +434,318 @@ therefore be applied **after** tokenization, but _before_ the dependency parsing ```python ### {executable="true"} import spacy +from spacy.language import Language +@Language.component("custom_sentencizer") def custom_sentencizer(doc): for i, token in enumerate(doc[:-2]): # Define sentence start if pipe + titlecase token - if token.text == "|" and doc[i+1].is_title: - doc[i+1].is_sent_start = True + if token.text == "|" and doc[i + 1].is_title: + doc[i + 1].is_sent_start = True else: # Explicitly set sentence start to False otherwise, to tell # the parser to leave those tokens alone - doc[i+1].is_sent_start = False + doc[i + 1].is_sent_start = False return doc nlp = spacy.load("en_core_web_sm") -nlp.add_pipe(custom_sentencizer, before="parser") # Insert before the parser +nlp.add_pipe("custom_sentencizer", before="parser") # Insert before the parser doc = nlp("This is. A sentence. | This is. Another sentence.") for sent in doc.sents: print(sent.text) ``` -### Example: Pipeline component for entity matching and tagging with custom attributes {#component-example2} +### Component factories and stateful components {#custom-components-factories} -This example shows how to create a spaCy extension that takes a terminology list -(in this case, single- and multi-word company names), matches the occurrences in -a document, labels them as `ORG` entities, merges the tokens and sets custom -`is_tech_org` and `has_tech_org` attributes. For efficient matching, the example -uses the [`PhraseMatcher`](/api/phrasematcher) which accepts `Doc` objects as -match patterns and works well for large terminology lists. It also ensures your -patterns will always match, even when you customize spaCy's tokenization rules. -When you call `nlp` on a text, the custom pipeline component is applied to the -`Doc`. - -```python -https://github.com/explosion/spaCy/tree/master/examples/pipeline/custom_component_entities.py -``` - -Wrapping this functionality in a pipeline component allows you to reuse the -module with different settings, and have all pre-processing taken care of when -you call `nlp` on your text and receive a `Doc` object. - -### Adding factories {#custom-components-factories} - -When spaCy loads a model via its `meta.json`, it will iterate over the -`"pipeline"` setting, look up every component name in the internal factories and -call [`nlp.create_pipe`](/api/language#create_pipe) to initialize the individual -components, like the tagger, parser or entity recognizer. If your model uses -custom components, this won't work – so you'll have to tell spaCy **where to -find your component**. You can do this by writing to the `Language.factories`: +Component factories are callables that take settings and return a **pipeline +component function**. This is useful if your component is stateful and if you +need to customize their creation, or if you need access to the current `nlp` +object or the shared vocab. Component factories can be registered using the +[`@Language.factory`](/api/language#factory) decorator and they need at least +**two named arguments** that are filled in automatically when the component is +added to the pipeline: + +> #### Example +> +> ```python +> from spacy.language import Language +> +> @Language.factory("my_component") +> def my_component(nlp, name): +> return MyComponent() +> ``` + +| Argument | Type | Description | +| -------- | --------------------------- | ------------------------------------------------------------------------------------------------------------------------- | +| `nlp` | [`Language`](/api/language) | The current `nlp` object. Can be used to access the | +| `name` | str | The **instance name** of the component in the pipeline. This lets you identify different instances of the same component. | + +All other settings can be passed in by the user via the `config` argument on +[`nlp.add_pipe`](/api/language). The +[`@Language.factory`](/api/language#factory) decorator also lets you define a +`default_config` that's used as a fallback. ```python +### With config {highlight="4,9"} +import spacy from spacy.language import Language -Language.factories["entity_matcher"] = lambda nlp, **cfg: EntityMatcher(nlp, **cfg) + +@Language.factory("my_component", default_config={"some_setting": True}) +def my_component(nlp, name, some_setting: bool): + return MyComponent(some_setting=some_setting) + +nlp = spacy.blank("en") +nlp.add_pipe("my_component", config={"some_setting": False}) ``` -You can also ship the above code and your custom component in your packaged -model's `__init__.py`, so it's executed when you load your model. The `**cfg` -config parameters are passed all the way down from -[`spacy.load`](/api/top-level#spacy.load), so you can load the model and its -components with custom settings: + + +The [`@Language.component`](/api/language#component) decorator is essentially a +**shortcut** for stateless pipeline component that don't need any settings. This +means you don't have to always write a function that returns your function if +there's no state to be passed through – spaCy can just take care of this for +you. The following two code examples are equivalent: ```python -nlp = spacy.load("your_custom_model", terms=["tree kangaroo"], label="ANIMAL") +# Statless component with @Language.factory +@Language.factory("my_component") +def create_my_component(): + def my_component(doc): + # Do something to the doc + return doc + + return my_component + +# Stateless component with @Language.component +@Language.component("my_component") +def my_component(doc): + # Do something to the doc + return doc ``` - + -When you load a model via its package name, like `en_core_web_sm`, spaCy will -import the package and then call its `load()` method. This means that custom -code in the model's `__init__.py` will be executed, too. This is **not the -case** if you're loading a model from a path containing the model data. Here, -spaCy will only read in the `meta.json`. If you want to use custom factories -with a model loaded from a path, you need to add them to `Language.factories` -_before_ you load the model. + + +Yes, the [`@Language.factory`](/api/language#factory) decorator can be added to +a function or a class. If it's added to a class, it expects the `__init__` +method to take the arguments `nlp` and `name`, and will populate all other +arguments from the config. That said, it's often cleaner and more intuitive to +make your factory a separate function. That's also how spaCy does it internally. + + + +### Example: Stateful component with settings + +This example shows a **stateful** pipeline component for handling acronyms: +based on a dictionary, it will detect acronyms and their expanded forms in both +directions and add them to a list as the custom `doc._.acronyms` +[extension attribute](#custom-components-attributes). Under the hood, it uses +the [`PhraseMatcher`](/api/phrasematcher) to find instances of the phrases. + +The factory function takes three arguments: the shared `nlp` object and +component instance `name`, which are passed in automatically by spaCy, and a +`case_sensitive` config setting that makes the matching and acronym detection +case-sensitive. + +> #### ✏️ Things to try +> +> 1. Change the `config` passed to `nlp.add_pipe` and set `"case_sensitive"` to +> `True`. You should see that the expanded acronym for "LOL" isn't detected +> anymore. +> 2. Add some more terms to the `DICTIONARY` and update the processed text so +> they're detected. +> 3. Add a `name` argument to `nlp.add_pipe` to change the component name. Print +> `nlp.pipe_names` to see the change reflected in the pipeline. +> 4. Print the config of the current `nlp` object with +> `print(nlp.config.to_str())` and inspect the `[components]` block. You +> should see an entry for the acronyms component, referencing the factory +> `acronyms` and the config settings. + +```python +### {executable="true"} +from spacy.language import Language +from spacy.tokens import Doc +from spacy.matcher import PhraseMatcher +import spacy + +DICTIONARY = {"lol": "laughing out loud", "brb": "be right back"} +DICTIONARY.update({value: key for key, value in DICTIONARY.items()}) + +@Language.factory("acronyms", default_config={"case_sensitive": False}) +def create_acronym_component(nlp: Language, name: str, case_sensitive: bool): + return AcronymComponent(nlp, case_sensitive) + +class AcronymComponent: + def __init__(self, nlp: Language, case_sensitive: bool): + # Create the matcher and match on Token.lower if case-insensitive + matcher_attr = "TEXT" if case_sensitive else "LOWER" + self.matcher = PhraseMatcher(nlp.vocab, attr=matcher_attr) + self.matcher.add("ACRONYMS", [nlp.make_doc(term) for term in DICTIONARY]) + self.case_sensitive = case_sensitive + # Register custom extension on the Doc + if not Doc.has_extension("acronyms"): + Doc.set_extension("acronyms", default=[]) + + def __call__(self, doc: Doc) -> Doc: + # Add the matched spans when doc is processed + for _, start, end in self.matcher(doc): + span = doc[start:end] + acronym = DICTIONARY.get(span.text if self.case_sensitive else span.text.lower()) + doc._.acronyms.append((span, acronym)) + return doc + +# Add the component to the pipeline and configure it +nlp = spacy.blank("en") +nlp.add_pipe("acronyms", config={"case_sensitive": False}) + +# Process a doc and see the results +doc = nlp("LOL, be right back") +print(doc._.acronyms) +``` + +### Python type hints and pydantic validation {#type-hints new="3"} + +spaCy's configs are powered by our machine learning library Thinc's +[configuration system](https://thinc.ai/docs/usage-config), which supports +[type hints](https://docs.python.org/3/library/typing.html) and even +[advanced type annotations](https://thinc.ai/docs/usage-config#advanced-types) +using [`pydantic`](https://github.com/samuelcolvin/pydantic). If your component +factory provides type hints, the values that are passed in will be **checked +against the expected types**. If the value can't be cast to an integer, spaCy +will raise an error. `pydantic` also provides strict types like `StrictFloat`, +which will force the value to be an integer and raise an error if it's not – for +instance, if your config defines a float. + + + +If you're not using +[strict types](https://pydantic-docs.helpmanual.io/usage/types/#strict-types), +values that can be **cast to** the given type will still be accepted. For +example, `1` can be cast to a `float` or a `bool` type, but not to a +`List[str]`. However, if the type is +[`StrictFloat`](https://pydantic-docs.helpmanual.io/usage/types/#strict-types), +only a float will be accepted. +The following example shows a custom pipeline component for debugging. It can be +added anywhere in the pipeline and logs information about the `nlp` object and +the `Doc` that passes through. The `log_level` config setting lets the user +customize what log statements are shown – for instance, `"INFO"` will show info +logs and more critical logging statements, whereas `"DEBUG"` will show +everything. The value is annotated as a `StrictStr`, so it will only accept a +string value. + +> #### ✏️ Things to try +> +> 1. Change the `config` passed to `nlp.add_pipe` to use the log level `"INFO"`. +> You should see that only the statement logged with `logger.info` is shown. +> 2. Change the `config` passed to `nlp.add_pipe` so that it contains unexpected +> values – for example, a boolean instead of a string: `"log_level": False`. +> You should see a validation error. +> 3. Check out the docs on `pydantic`'s +> [constrained types](https://pydantic-docs.helpmanual.io/usage/types/#constrained-types) +> and write a type hint for `log_level` that only accepts the exact string +> values `"DEBUG"`, `"INFO"` or `"CRITICAL"`. + +```python +### {executable="true"} +import spacy +from spacy.language import Language +from spacy.tokens import Doc +from pydantic import StrictStr +import logging + +@Language.factory("debug", default_config={"log_level": "DEBUG"}) +class DebugComponent: + def __init__(self, nlp: Language, name: str, log_level: StrictStr): + self.logger = logging.getLogger(f"spacy.{name}") + self.logger.setLevel(log_level) + self.logger.info(f"Pipeline: {nlp.pipe_names}") + + def __call__(self, doc: Doc) -> Doc: + self.logger.debug(f"Doc: {len(doc)} tokens, is_tagged: {doc.is_tagged}") + return doc + +nlp = spacy.load("en_core_web_sm") +nlp.add_pipe("debug", config={"log_level": "DEBUG"}) +doc = nlp("This is a text...") +``` + +### Language-specific factories {#factories-language new="3"} + +There are many use case where you might want your pipeline components to be +language-specific. Sometimes this requires entirely different implementation per +language, sometimes the only difference is in the settings or data. spaCy allows +you to register factories of the **same name** on both the `Language` base +class, as well as its **subclasses** like `English` or `German`. Factories are +resolved starting with the specific subclass. If the subclass doesn't define a +component of that name, spaCy will check the `Language` base class. + +Here's an example of a pipeline component that overwrites the normalized form of +a token, the `Token.norm_` with an entry from a language-specific lookup table. +It's registered twice under the name `"token_normalizer"` – once using +`@English.factory` and once using `@German.factory`: + +```python +### {executable="true"} +from spacy.lang.en import English +from spacy.lang.de import German + +class TokenNormalizer: + def __init__(self, norm_table): + self.norm_table = norm_table + + def __call__(self, doc): + for token in doc: + # Overwrite the token.norm_ if there's an entry in the data + token.norm_ = self.norm_table.get(token.text, token.norm_) + return doc + +@English.factory("token_normalizer") +def create_en_normalizer(nlp, name): + return TokenNormalizer({"realise": "realize", "colour": "color"}) + +@German.factory("token_normalizer") +def create_de_normalizer(nlp, name): + return TokenNormalizer({"daß": "dass", "wußte": "wusste"}) + +nlp_en = English() +nlp_en.add_pipe("token_normalizer") # uses the English factory +print([token.norm_ for token in nlp_en("realise colour daß wußte")]) + +nlp_de = German() +nlp_de.add_pipe("token_normalizer") # uses the German factory +print([token.norm_ for token in nlp_de("realise colour daß wußte")]) +``` + + + +Under the hood, language-specific factories are added to the +[`factories` registry](/api/top-level#registry) prefixed with the language code, +e.g. `"en.token_normalizer"`. When resolving the factory in +[`nlp.add_pipe`](/api/language#add_pipe), spaCy first checks for a +language-specific version of the factory using `nlp.lang` and if none is +available, falls back to looking up the regular factory name. + + + + + ## Extension attributes {#custom-components-attributes new="2"} -As of v2.0, spaCy allows you to set any custom attributes and methods on the -`Doc`, `Span` and `Token`, which become available as `Doc._`, `Span._` and -`Token._` – for example, `Token._.my_attr`. This lets you store additional -information relevant to your application, add new features and functionality to -spaCy, and implement your own models trained with other machine learning -libraries. It also lets you take advantage of spaCy's data structures and the -`Doc` object as the "single source of truth". +spaCy allows you to set any custom attributes and methods on the `Doc`, `Span` +and `Token`, which become available as `Doc._`, `Span._` and `Token._` – for +example, `Token._.my_attr`. This lets you store additional information relevant +to your application, add new features and functionality to spaCy, and implement +your own models trained with other machine learning libraries. It also lets you +take advantage of spaCy's data structures and the `Doc` object as the "single +source of truth". @@ -641,7 +855,73 @@ attributes on the `Doc`, `Span` and `Token` – for example, the capital, latitude/longitude coordinates and even the country flag. ```python -https://github.com/explosion/spaCy/tree/master/examples/pipeline/custom_component_countries_api.py +### {executable="true"} +import requests +from spacy.lang.en import English +from spacy.language import Language +from spacy.matcher import PhraseMatcher +from spacy.tokens import Doc, Span, Token + +@Language.factory("rest_countries") +class RESTCountriesComponent: + def __init__(self, nlp, name, label="GPE"): + r = requests.get("https://restcountries.eu/rest/v2/all") + r.raise_for_status() # make sure requests raises an error if it fails + countries = r.json() + # Convert API response to dict keyed by country name for easy lookup + self.countries = {c["name"]: c for c in countries} + self.label = label + # Set up the PhraseMatcher with Doc patterns for each country name + self.matcher = PhraseMatcher(nlp.vocab) + self.matcher.add("COUNTRIES", [nlp.make_doc(c) for c in self.countries.keys()]) + # Register attribute on the Token. We'll be overwriting this based on + # the matches, so we're only setting a default value, not a getter. + Token.set_extension("is_country", default=False) + Token.set_extension("country_capital", default=False) + Token.set_extension("country_latlng", default=False) + Token.set_extension("country_flag", default=False) + # Register attributes on Doc and Span via a getter that checks if one of + # the contained tokens is set to is_country == True. + Doc.set_extension("has_country", getter=self.has_country) + Span.set_extension("has_country", getter=self.has_country) + + def __call__(self, doc): + spans = [] # keep the spans for later so we can merge them afterwards + for _, start, end in self.matcher(doc): + # Generate Span representing the entity & set label + entity = Span(doc, start, end, label=self.label) + spans.append(entity) + # Set custom attribute on each token of the entity + # Can be extended with other data returned by the API, like + # currencies, country code, flag, calling code etc. + for token in entity: + token._.set("is_country", True) + token._.set("country_capital", self.countries[entity.text]["capital"]) + token._.set("country_latlng", self.countries[entity.text]["latlng"]) + token._.set("country_flag", self.countries[entity.text]["flag"]) + # Iterate over all spans and merge them into one token + with doc.retokenize() as retokenizer: + for span in spans: + retokenizer.merge(span) + # Overwrite doc.ents and add entity – be careful not to replace! + doc.ents = list(doc.ents) + spans + return doc # don't forget to return the Doc! + + def has_country(self, tokens): + """Getter for Doc and Span attributes. Since the getter is only called + when we access the attribute, we can refer to the Token's 'is_country' + attribute here, which is already set in the processing step.""" + return any([t._.get("is_country") for t in tokens]) + +nlp = English() +nlp.add_pipe("rest_countries", config={"label": "GPE"}) +doc = nlp("Some text about Colombia and the Czech Republic") +print("Pipeline", nlp.pipe_names) # pipeline contains component name +print("Doc has countries", doc._.has_country) # Doc contains countries +for token in doc: + if token._.is_country: + print(token.text, token._.country_capital, token._.country_latlng, token._.country_flag) +print("Entities", [(e.text, e.label_) for e in doc.ents]) ``` In this case, all data can be fetched on initialization in one request. However, @@ -800,11 +1080,6 @@ function that takes a `Doc`, modifies it and returns it. [`load_model_from_path`](/api/top-level#util.load_model_from_path) utility functions. - ```diff - + nlp.add_pipe(my_custom_component) - + return nlp.from_disk(model_path) - ``` - - Once you're ready to share your extension with others, make sure to **add docs and installation instructions** (you can always link to this page for more info). Make it easy for others to install and use your extension, for example @@ -838,10 +1113,12 @@ wrapper has to do is compute the entity spans and overwrite the `doc.ents`. > overlapping entity spans are not allowed. ```python -### {highlight="1,6-7"} +### {highlight="1,8-9"} import your_custom_entity_recognizer from spacy.gold import offsets_from_biluo_tags +from spacy.language import Language +@Language.component("custom_ner_wrapper") def custom_ner_wrapper(doc): words = [token.text for token in doc] custom_entities = your_custom_entity_recognizer(words) @@ -865,22 +1142,24 @@ because it returns the integer ID of the string _and_ makes sure it's added to the vocab. This is especially important if the custom model uses a different label scheme than spaCy's default models. -> #### Example: spacy-stanfordnlp +> #### Example: spacy-stanza > > For an example of an end-to-end wrapper for statistical tokenization, tagging > and parsing, check out -> [`spacy-stanfordnlp`](https://github.com/explosion/spacy-stanfordnlp). It uses -> a very similar approach to the example in this section – the only difference -> is that it fully replaces the `nlp` object instead of providing a pipeline -> component, since it also needs to handle tokenization. +> [`spacy-stanza`](https://github.com/explosion/spacy-stanza). It uses a very +> similar approach to the example in this section – the only difference is that +> it fully replaces the `nlp` object instead of providing a pipeline component, +> since it also needs to handle tokenization. ```python -### {highlight="1,9,15-17"} +### {highlight="1,11,17-19"} import your_custom_model +from spacy.language import Language from spacy.symbols import POS, TAG, DEP, HEAD from spacy.tokens import Doc import numpy +@Language.component("custom_model_wrapper") def custom_model_wrapper(doc): words = [token.text for token in doc] spaces = [token.whitespace for token in doc] diff --git a/website/docs/usage/projects.md b/website/docs/usage/projects.md index ff8f91683..d923766b0 100644 --- a/website/docs/usage/projects.md +++ b/website/docs/usage/projects.md @@ -450,6 +450,14 @@ git init # Initialize a Git repo dvc init # Initialize a DVC project ``` + + +DVC enables usage analytics by default, so if you're working in a +privacy-sensitive environment, make sure to +[**opt-out manually**](https://dvc.org/doc/user-guide/analytics#opting-out). + + + The [`spacy project dvc`](/api/cli#project-dvc) command creates a `dvc.yaml` config file based on a workflow defined in your `project.yml`. Whenever you update your project, you can re-run the command to update your DVC config. You diff --git a/website/docs/usage/rule-based-matching.md b/website/docs/usage/rule-based-matching.md index 14d394651..2e07eff48 100644 --- a/website/docs/usage/rule-based-matching.md +++ b/website/docs/usage/rule-based-matching.md @@ -506,11 +506,16 @@ attribute `bad_html` on the token. ```python ### {executable="true"} import spacy +from spacy.language import Language from spacy.matcher import Matcher from spacy.tokens import Token -# We're using a class because the component needs to be initialized with -# the shared vocab via the nlp object +# We're using a component factory because the component needs to be initialized +# with the shared vocab via the nlp object +@Language.factory("html_merger") +def create_bad_html_merger(nlp, name): + return BadHTMLMerger(nlp) + class BadHTMLMerger: def __init__(self, nlp): patterns = [ @@ -536,8 +541,7 @@ class BadHTMLMerger: return doc nlp = spacy.load("en_core_web_sm") -html_merger = BadHTMLMerger(nlp) -nlp.add_pipe(html_merger, last=True) # Add component to the pipeline +nlp.add_pipe("html_merger", last=True) # Add component to the pipeline doc = nlp("Hello
world!
This is a test.") for token in doc: print(token.text, token._.bad_html) @@ -546,10 +550,16 @@ for token in doc: Instead of hard-coding the patterns into the component, you could also make it take a path to a JSON file containing the patterns. This lets you reuse the -component with different patterns, depending on your application: +component with different patterns, depending on your application. When adding +the component to the pipeline with [`nlp.add_pipe`](/api/language#add_pipe), you +can pass in the argument via the `config`: ```python -html_merger = BadHTMLMerger(nlp, path="/path/to/patterns.json") +@Language.factory("html_merger", default_config={"path": None}) +def create_bad_html_merger(nlp, name, path): + return BadHTMLMerger(nlp, path=path) + +nlp.add_pipe("html_merger", config={"path": "/path/to/patterns.json"}) ``` @@ -835,7 +845,7 @@ patterns can contain single or multiple tokens. import spacy from spacy.matcher import PhraseMatcher -nlp = spacy.load('en_core_web_sm') +nlp = spacy.load("en_core_web_sm") matcher = PhraseMatcher(nlp.vocab) terms = ["Barack Obama", "Angela Merkel", "Washington, D.C."] # Only run nlp.make_doc to speed things up @@ -975,14 +985,12 @@ chosen. ```python ### {executable="true"} from spacy.lang.en import English -from spacy.pipeline import EntityRuler nlp = English() -ruler = EntityRuler(nlp) +ruler = nlp.add_pipe("entity_ruler") patterns = [{"label": "ORG", "pattern": "Apple"}, {"label": "GPE", "pattern": [{"LOWER": "san"}, {"LOWER": "francisco"}]}] ruler.add_patterns(patterns) -nlp.add_pipe(ruler) doc = nlp("Apple is opening its first big office in San Francisco.") print([(ent.text, ent.label_) for ent in doc.ents]) @@ -1000,13 +1008,11 @@ can set `overwrite_ents=True` on initialization. ```python ### {executable="true"} import spacy -from spacy.pipeline import EntityRuler nlp = spacy.load("en_core_web_sm") -ruler = EntityRuler(nlp) +ruler = nlp.add_pipe("entity_ruler") patterns = [{"label": "ORG", "pattern": "MyCorp Inc."}] ruler.add_patterns(patterns) -nlp.add_pipe(ruler) doc = nlp("MyCorp Inc. is a company in the U.S.") print([(ent.text, ent.label_) for ent in doc.ents]) @@ -1014,12 +1020,12 @@ print([(ent.text, ent.label_) for ent in doc.ents]) #### Validating and debugging EntityRuler patterns {#entityruler-pattern-validation new="2.1.8"} -The `EntityRuler` can validate patterns against a JSON schema with the option -`validate=True`. See details under +The entity ruler can validate patterns against a JSON schema with the config +setting `"validate"`. See details under [Validating and debugging patterns](#pattern-validation). ```python -ruler = EntityRuler(nlp, validate=True) +ruler = nlp.add_pipe("entity_ruler", config={"validate": True}) ``` ### Adding IDs to patterns {#entityruler-ent-ids new="2.2.2"} @@ -1031,15 +1037,13 @@ the same entity. ```python ### {executable="true"} from spacy.lang.en import English -from spacy.pipeline import EntityRuler nlp = English() -ruler = EntityRuler(nlp) +ruler = nlp.add_pipe("entity_ruler") patterns = [{"label": "ORG", "pattern": "Apple", "id": "apple"}, {"label": "GPE", "pattern": [{"LOWER": "san"}, {"LOWER": "francisco"}], "id": "san-francisco"}, {"label": "GPE", "pattern": [{"LOWER": "san"}, {"LOWER": "fran"}], "id": "san-francisco"}] ruler.add_patterns(patterns) -nlp.add_pipe(ruler) doc1 = nlp("Apple is opening its first big office in San Francisco.") print([(ent.text, ent.label_, ent.ent_id_) for ent in doc1.ents]) @@ -1068,7 +1072,7 @@ line. ```python ruler.to_disk("./patterns.jsonl") -new_ruler = EntityRuler(nlp).from_disk("./patterns.jsonl") +new_ruler = nlp.add_pipe("entity_ruler").from_disk("./patterns.jsonl") ``` @@ -1086,9 +1090,8 @@ pipeline, its patterns are automatically exported to the model directory: ```python nlp = spacy.load("en_core_web_sm") -ruler = EntityRuler(nlp) +ruler = nlp.add_pipe("entity_ruler") ruler.add_patterns([{"label": "ORG", "pattern": "Apple"}]) -nlp.add_pipe(ruler) nlp.to_disk("/path/to/model") ``` @@ -1100,35 +1103,30 @@ powerful model packages with binary weights _and_ rules included! ### Using a large number of phrase patterns {#entityruler-large-phrase-patterns new="2.2.4"} + + When using a large amount of **phrase patterns** (roughly > 10000) it's useful -to understand how the `add_patterns` function of the EntityRuler works. For each -**phrase pattern**, the EntityRuler calls the nlp object to construct a doc +to understand how the `add_patterns` function of the entity ruler works. For +each **phrase pattern**, the EntityRuler calls the nlp object to construct a doc object. This happens in case you try to add the EntityRuler at the end of an existing pipeline with, for example, a POS tagger and want to extract matches -based on the pattern's POS signature. - -In this case you would pass a config value of `phrase_matcher_attr="POS"` for -the EntityRuler. +based on the pattern's POS signature. In this case you would pass a config value +of `"phrase_matcher_attr": "POS"` for the entity ruler. Running the full language pipeline across every pattern in a large list scales linearly and can therefore take a long time on large amounts of phrase patterns. - As of spaCy 2.2.4 the `add_patterns` function has been refactored to use nlp.pipe on all phrase patterns resulting in about a 10x-20x speed up with -5,000-100,000 phrase patterns respectively. - -Even with this speedup (but especially if you're using an older version) the -`add_patterns` function can still take a long time. - -An easy workaround to make this function run faster is disabling the other -language pipes while adding the phrase patterns. +5,000-100,000 phrase patterns respectively. Even with this speedup (but +especially if you're using an older version) the `add_patterns` function can +still take a long time. An easy workaround to make this function run faster is +disabling the other language pipes while adding the phrase patterns. ```python -entityruler = EntityRuler(nlp) +ruler = nlp.add_pipe("entity_ruler") patterns = [{"label": "TEST", "pattern": str(i)} for i in range(100000)] - with nlp.select_pipes(enable="tagger"): - entityruler.add_patterns(patterns) + ruler.add_patterns(patterns) ``` ## Combining models and rules {#models-rules} @@ -1189,9 +1187,11 @@ have in common is that _if_ they occur, they occur in the **previous token** right before the person entity. ```python -### {highlight="7-11"} +### {highlight="9-13"} +from spacy.language import Language from spacy.tokens import Span +@Language.component("expand_person_entities") def expand_person_entities(doc): new_ents = [] for ent in doc.ents: @@ -1210,18 +1210,20 @@ def expand_person_entities(doc): ``` The above function takes a `Doc` object, modifies its `doc.ents` and returns it. -This is exactly what a [pipeline component](/usage/processing-pipelines) does, -so in order to let it run automatically when processing a text with the `nlp` -object, we can use [`nlp.add_pipe`](/api/language#add_pipe) to add it to the -current pipeline. +Using the [`@Language.component`](/api/language#component) decorator, we can +register it as a [pipeline component](/usage/processing-pipelines) so it can run +automatically when processing a text. We can use +[`nlp.add_pipe`](/api/language#add_pipe) to add it to the current pipeline. ```python ### {executable="true"} import spacy +from spacy.language import Language from spacy.tokens import Span nlp = spacy.load("en_core_web_sm") +@Language.component("expand_person_entities") def expand_person_entities(doc): new_ents = [] for ent in doc.ents: @@ -1236,7 +1238,7 @@ def expand_person_entities(doc): return doc # Add the component after the named entity recognizer -nlp.add_pipe(expand_person_entities, after='ner') +nlp.add_pipe("expand_person_entities", after="ner") doc = nlp("Dr. Alex Smith chaired first board meeting of Acme Corp Inc.") print([(ent.text, ent.label_) for ent in doc.ents]) @@ -1347,7 +1349,7 @@ for ent in person_entities: # children, e.g. at -> Acme Corp Inc. orgs = [token for token in prep.children if token.ent_type_ == "ORG"] # If the verb is in past tense, the company was a previous company - print({'person': ent, 'orgs': orgs, 'past': head.tag_ == "VBD"}) + print({"person": ent, "orgs": orgs, "past": head.tag_ == "VBD"}) ``` To apply this logic automatically when we process a text, we can add it to the @@ -1374,11 +1376,12 @@ the entity `Span` – for example `._.orgs` or `._.prev_orgs` and ```python ### {executable="true"} import spacy -from spacy.pipeline import merge_entities +from spacy.language import Language from spacy import displacy nlp = spacy.load("en_core_web_sm") +@Language.component("extract_person_orgs") def extract_person_orgs(doc): person_entities = [ent for ent in doc.ents if ent.label_ == "PERSON"] for ent in person_entities: @@ -1391,12 +1394,12 @@ def extract_person_orgs(doc): return doc # To make the entities easier to work with, we'll merge them into single tokens -nlp.add_pipe(merge_entities) -nlp.add_pipe(extract_person_orgs) +nlp.add_pipe("merge_entities") +nlp.add_pipe("extract_person_orgs") doc = nlp("Alex Smith worked at Acme Corp Inc.") # If you're not in a Jupyter / IPython environment, use displacy.serve -displacy.render(doc, options={'fine_grained': True}) +displacy.render(doc, options={"fine_grained": True}) ``` If you change the sentence structure above, for example to "was working", you'll @@ -1409,7 +1412,8 @@ information is in the attached auxiliary "was": To solve this, we can adjust the rules to also check for the above construction: ```python -### {highlight="9-11"} +### {highlight="10-12"} +@Language.component("extract_person_orgs") def extract_person_orgs(doc): person_entities = [ent for ent in doc.ents if ent.label_ == "PERSON"] for ent in person_entities: diff --git a/website/docs/usage/saving-loading.md b/website/docs/usage/saving-loading.md index 6d17ee7e3..cdd7d1c49 100644 --- a/website/docs/usage/saving-loading.md +++ b/website/docs/usage/saving-loading.md @@ -15,6 +15,8 @@ import Serialization101 from 'usage/101/\_serialization.md' ### Serializing the pipeline {#pipeline} + + When serializing the pipeline, keep in mind that this will only save out the **binary data for the individual components** to allow spaCy to restore them – not the entire objects. This is a good thing, because it makes serialization @@ -22,32 +24,35 @@ safe. But it also means that you have to take care of storing the language name and pipeline component names as well, and restoring them separately before you can load in the data. -> #### Saving the model meta +> #### Saving the meta and config > -> The `nlp.meta` attribute is a JSON-serializable dictionary and contains all -> model meta information, like the language and pipeline, but also author and -> license information. +> The [`nlp.meta`](/api/language#meta) attribute is a JSON-serializable +> dictionary and contains all model meta information like the author and license +> information. The [`nlp.config`](/api/language#config) attribute is a +> dictionary containing the training configuration, pipeline component factories +> and other settings. It is saved out with a model as the `config.cfg`. ```python ### Serialize bytes_data = nlp.to_bytes() -lang = nlp.meta["lang"] # "en" -pipeline = nlp.meta["pipeline"] # ["tagger", "parser", "ner"] +lang = nlp.config["nlp"]["lang"] # "en" +pipeline = nlp.config["nlp"]["pipeline"] # ["tagger", "parser", "ner"] ``` ```python ### Deserialize nlp = spacy.blank(lang) for pipe_name in pipeline: - pipe = nlp.create_pipe(pipe_name) - nlp.add_pipe(pipe) + nlp.add_pipe(pipe_name) nlp.from_bytes(bytes_data) ``` This is also how spaCy does it under the hood when loading a model: it loads the -model's `meta.json` containing the language and pipeline information, -initializes the language class, creates and adds the pipeline components and -_then_ loads in the binary data. You can read more about this process +model's `config.cfg` containing the language and pipeline information, +initializes the language class, creates and adds the pipeline components based +on the defined +[factories](/usage/processing-pipeline#custom-components-factories) and _then_ +loads in the binary data. You can read more about this process [here](/usage/processing-pipelines#pipelines). ### Serializing Doc objects efficiently {#docs new="2.2"} @@ -192,10 +197,9 @@ add to that data and saves and loads the data to and from a JSON file. > recognizer and including all rules _with_ the model data. ```python -### {highlight="15-19,21-26"} +### {highlight="14-18,20-25"} +@Language.factory("my_component") class CustomComponent: - name = "my_component" - def __init__(self): self.data = [] @@ -228,9 +232,8 @@ component's `to_disk` method. ```python ### {highlight="2-4"} nlp = spacy.load("en_core_web_sm") -my_component = CustomComponent() +my_component = nlp.add_pipe("my_component") my_component.add({"hello": "world"}) -nlp.add_pipe(my_component) nlp.to_disk("/path/to/model") ``` @@ -247,7 +250,8 @@ file `data.json` in its subdirectory: ├── parser # data for "parser" component ├── tagger # data for "tagger" component ├── vocab # model vocabulary - ├── meta.json # model meta.json with name, language and pipeline + ├── meta.json # model meta.json + ├── config.cfg # model config └── tokenizer # tokenization rules ``` @@ -260,19 +264,14 @@ instance, you could add a trained with a different library like TensorFlow or PyTorch and make spaCy load its weights automatically when you load the model package. - + -When you load a model from disk, spaCy will check the `"pipeline"` in the -model's `meta.json` and look up the component name in the internal factories. To -make sure spaCy knows how to initialize `"my_component"`, you'll need to add it -to the factories: - -```python -from spacy.language import Language -Language.factories["my_component"] = lambda nlp, **cfg: CustomComponent() -``` - -For more details, see the documentation on +When you load back a model with custom components, make sure that the components +are **available** and that the [`@Language.component`](/api/language#component) +or [`@Language.factory`](/api/language#factory) decorators are executed _before_ +your model is loaded back. Otherwise, spaCy won't know how to resolve the string +name of a component factory like `"my_component"` back to a function. For more +details, see the documentation on [adding factories](/usage/processing-pipelines#custom-components-factories) or use [entry points](#entry-points) to make your extension package expose your custom components to spaCy automatically. @@ -293,40 +292,31 @@ installed in the same environment – that's it. | Entry point | Description | | ------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -| [`spacy_factories`](#entry-points-components) | Group of entry points for pipeline component factories to add to [`Language.factories`](/usage/processing-pipelines#custom-components-factories), keyed by component name. | +| [`spacy_factories`](#entry-points-components) | Group of entry points for pipeline component factories, keyed by component name. Can be used to expose custom components defined by another package. | | [`spacy_languages`](#entry-points-languages) | Group of entry points for custom [`Language` subclasses](/usage/adding-languages), keyed by language shortcut. | | `spacy_lookups` 2.2 | Group of entry points for custom [`Lookups`](/api/lookups), including lemmatizer data. Used by spaCy's [`spacy-lookups-data`](https://github.com/explosion/spacy-lookups-data) package. | | [`spacy_displacy_colors`](#entry-points-displacy) 2.2 | Group of entry points of custom label colors for the [displaCy visualizer](/usage/visualizers#ent). The key name doesn't matter, but it should point to a dict of labels and color values. Useful for custom models that predict different entity types. | ### Custom components via entry points {#entry-points-components} -When you load a model, spaCy will generally use the model's `meta.json` to set +When you load a model, spaCy will generally use the model's `config.cfg` to set up the language class and construct the pipeline. The pipeline is specified as a -list of strings, e.g. `"pipeline": ["tagger", "paser", "ner"]`. For each of -those strings, spaCy will call `nlp.create_pipe` and look up the name in the -[built-in factories](/usage/processing-pipelines#custom-components-factories). -If your model wanted to specify its own custom components, you usually have to -write to `Language.factories` _before_ loading the model. +list of strings, e.g. `pipeline = ["tagger", "paser", "ner"]`. For each of those +strings, spaCy will call `nlp.add_pipe` and look up the name in all factories +defined by the decorators [`@Language.component`](/api/language#component) and +[`@Language.factory`](/api/language#factory). This means that you have to import +your custom components _before_ loading the model. -```python -pipe = nlp.create_pipe("custom_component") # fails 👎 - -Language.factories["custom_component"] = CustomComponentFactory -pipe = nlp.create_pipe("custom_component") # works 👍 -``` - -This is inconvenient and usually required shipping a bunch of component -initialization code with the model. Using entry points, model packages and -extension packages can now define their own `"spacy_factories"`, which will be -added to the built-in factories when the `Language` class is initialized. If a -package in the same environment exposes spaCy entry points, all of this happens -automatically and no further user action is required. +Using entry points, model packages and extension packages can define their own +`"spacy_factories"`, which will be loaded automatically in the background when +the `Language` class is initialized. So if a user has your package installed, +they'll be able to use your components – even if they **don't import them**! To stick with the theme of [this entry points blog post](https://amir.rachum.com/blog/2017/07/28/python-entry-points/), -consider the following custom spaCy extension which is initialized with the -shared `nlp` object and will print a snake when it's called as a pipeline -component. +consider the following custom spaCy +[pipeline component](/usage/processing-pipelines#custom-coponents) that prints a +snake when it's called: > #### Package directory structure > @@ -337,32 +327,38 @@ component. ```python ### snek.py +from spacy.language import Language + snek = """ --..,_ _,.--. - `'.'. .'`__ o `;__. + `'.'. .'`__ o `;__. {text} '.'. .'.'` '---'` ` '.`'--....--'`.' `'--....--'` """ -class SnekFactory: - def __init__(self, nlp, **cfg): - self.nlp = nlp - - def __call__(self, doc): - print(snek) - return doc +@Language.component("snek") +def snek_component(doc): + print(snek.format(text=doc.text)) + return doc ``` Since it's a very complex and sophisticated module, you want to split it off into its own package so you can version it and upload it to PyPi. You also want -your custom model to be able to define `"pipeline": ["snek"]` in its -`meta.json`. For that, you need to be able to tell spaCy where to find the -factory for `"snek"`. If you don't do this, spaCy will raise an error when you -try to load the model because there's no built-in `"snek"` factory. To add an +your custom model to be able to define `pipeline = ["snek"]` in its +`config.cfg`. For that, you need to be able to tell spaCy where to find the +component `"snek"`. If you don't do this, spaCy will raise an error when you try +to load the model because there's no built-in `"snek"` component. To add an entry to the factories, you can now expose it in your `setup.py` via the `entry_points` dictionary: +> #### Entry point syntax +> +> Python entry points for a group are formatted as a **list of strings**, with +> each string following the syntax of `name = module:object`. In this example, +> the created entry point is named `snek` and points to the function +> `snek_component` in the module `snek`, i.e. `snek.py`. + ```python ### setup.py {highlight="5-7"} from setuptools import setup @@ -370,73 +366,74 @@ from setuptools import setup setup( name="snek", entry_points={ - "spacy_factories": ["snek = snek:SnekFactory"] + "spacy_factories": ["snek = snek:snek_component"] } ) ``` -The entry point definition tells spaCy that the name `snek` can be found in the -module `snek` (i.e. `snek.py`) as `SnekFactory`. The same package can expose -multiple entry points. To make them available to spaCy, all you need to do is -install the package: +The same package can expose multiple entry points, by the way. To make them +available to spaCy, all you need to do is install the package in your +environment: ```bash $ python setup.py develop ``` -spaCy is now able to create the pipeline component `'snek'`: +spaCy is now able to create the pipeline component `"snek"` – even though you +never imported `snek_component`. When you save the +[`nlp.config`](/api/language#config) to disk, it includes an entry for your +`"snek"` component and any model you train with this config will include the +component and know how to load it – if your `snek` package is installed. + +> #### config.cfg (excerpt) +> +> ```diff +> [nlp] +> lang = "en" +> + pipeline = ["snek"] +> +> [components] +> +> + [components.snek] +> + factory = "snek" +> ``` ``` >>> from spacy.lang.en import English >>> nlp = English() ->>> snek = nlp.create_pipe("snek") # this now works! 🐍🎉 ->>> nlp.add_pipe(snek) +>>> nlp.add_pipe("snek") # this now works! 🐍🎉 >>> doc = nlp("I am snek") --..,_ _,.--. - `'.'. .'`__ o `;__. + `'.'. .'`__ o `;__. I am snek '.'. .'.'` '---'` ` '.`'--....--'`.' `'--....--'` ``` -Arguably, this gets even more exciting when you train your `en_core_snek_sm` -model. To make sure `snek` is installed with the model, you can add it to the -model's `setup.py`. You can then tell spaCy to construct the model pipeline with -the `snek` component by setting `"pipeline": ["snek"]` in the `meta.json`. +Instead of making your snek component a simple +[stateless component](/usage/processing-pipelines#custom-components-simple), you +could also make it a +[factory](/usage/processing-pipelines#custom-components-factories) that takes +settings. Your users can then pass in an optional `config` when they add your +component to the pipeline and customize its appearance – for example, the +`snek_style`. -> #### meta.json +> #### config.cfg (excerpt) > > ```diff -> { -> "lang": "en", -> "name": "core_snek_sm", -> "version": "1.0.0", -> + "pipeline": ["snek"] -> } +> [components.snek] +> factory = "snek" +> + snek_style = "basic" > ``` -In theory, the entry point mechanism also lets you overwrite built-in factories -– including the tokenizer. By default, spaCy will output a warning in these -cases, to prevent accidental overwrites and unintended results. - -#### Advanced components with settings {#advanced-cfg} - -The `**cfg` keyword arguments that the factory receives are passed down all the -way from `spacy.load`. This means that the factory can respond to custom -settings defined when loading the model – for example, the style of the snake to -load: - -```python -nlp = spacy.load("en_core_snek_sm", snek_style="cute") -``` - ```python SNEKS = {"basic": snek, "cute": cute_snek} # collection of sneks +@Language.factory("snek", default_config={"snek_style": "basic"}) class SnekFactory: - def __init__(self, nlp, **cfg): + def __init__(self, nlp: Language, name: str, snek_style: str): self.nlp = nlp - self.snek_style = cfg.get("snek_style", "basic") + self.snek_style = snek_style self.snek = SNEKS[self.snek_style] def __call__(self, doc): @@ -444,6 +441,14 @@ class SnekFactory: return doc ``` +```diff +### setup.py +entry_points={ +- "spacy_factories": ["snek = snek:snek_component"] ++ "spacy_factories": ["snek = snek:SnekFactory"] +} +``` + The factory can also implement other pipeline component like `to_disk` and `from_disk` for serialization, or even `update` to make the component trainable. If a component exposes a `from_disk` method and is included in a model's @@ -452,12 +457,12 @@ model. When you save out a model using `nlp.to_disk` and the component exposes a `to_disk` method, it will be called with the disk path. ```python -def to_disk(self, path, **kwargs): +def to_disk(self, path, exclude=tuple()): snek_path = path / "snek.txt" with snek_path.open("w", encoding="utf8") as snek_file: snek_file.write(self.snek) -def from_disk(self, path, **cfg): +def from_disk(self, path, exclude=tuple()): snek_path = path / "snek.txt" with snek_path.open("r", encoding="utf8") as snek_file: self.snek = snek_file.read() @@ -473,24 +478,20 @@ the `snek.txt` and make it available to the component. To stay with the theme of the previous example and [this blog post on entry points](https://amir.rachum.com/blog/2017/07/28/python-entry-points/), let's imagine you wanted to implement your own `SnekLanguage` class for your -custom model – but you don't necessarily want to modify spaCy's code to -[add a language](/usage/adding-languages). In your package, you could then -implement the following: +custom model – but you don't necessarily want to modify spaCy's code to add a +language. In your package, you could then implement the following +[custom language subclass](/usage/linguistic-features#language-subclass): ```python ### snek.py from spacy.language import Language -from spacy.attrs import LANG class SnekDefaults(Language.Defaults): - lex_attr_getters = dict(Language.Defaults.lex_attr_getters) - lex_attr_getters[LANG] = lambda text: "snk" - + stop_words = set(["sss", "hiss"]) class SnekLanguage(Language): lang = "snk" Defaults = SnekDefaults - # Some custom snek language stuff here ``` Alongside the `spacy_factories`, there's also an entry point option for @@ -510,31 +511,12 @@ setup( ) ``` -In spaCy, you can then load the custom `sk` language and it will be resolved to +In spaCy, you can then load the custom `snk` language and it will be resolved to `SnekLanguage` via the custom entry point. This is especially relevant for model -packages, which could then specify `"lang": "snk"` in their `meta.json` without -spaCy raising an error because the language is not available in the core +packages you train, which could then specify `lang = snk` in their `config.cfg` +without spaCy raising an error because the language is not available in the core library. -> #### meta.json -> -> ```diff -> { -> - "lang": "en", -> + "lang": "snk", -> "name": "core_snek_sm", -> "version": "1.0.0", -> "pipeline": ["snek"] -> } -> ``` - -```python -from spacy.util import get_lang_class - -SnekLanguage = get_lang_class("snk") -nlp = SnekLanguage() -``` - ### Custom displaCy colors via entry points {#entry-points-displacy new="2.2"} If you're training a named entity recognition model for a custom domain, you may @@ -611,7 +593,7 @@ manually and place it in the model data directory, or supply a path to it using the `--meta` flag. For more info on this, see the [`package`](/api/cli#package) docs. -> #### meta.json +> #### meta.json (example) > > ```json > { @@ -622,8 +604,7 @@ docs. > "description": "Example model for spaCy", > "author": "You", > "email": "you@example.com", -> "license": "CC BY-SA 3.0", -> "pipeline": ["tagger", "parser", "ner"] +> "license": "CC BY-SA 3.0" > } > ``` @@ -631,66 +612,39 @@ docs. $ python -m spacy package /home/me/data/en_example_model /home/me/my_models ``` -This command will create a model package directory that should look like this: +This command will create a model package directory and will run +`python setup.py sdist` in that directory to create `.tar.gz` archive of your +model package that can be installed using `pip install`. ```yaml ### Directory structure └── / - ├── MANIFEST.in # to include meta.json - ├── meta.json # model meta data - ├── setup.py # setup file for pip installation - └── en_example_model # model directory - ├── __init__.py # init for pip installation - └── en_example_model-1.0.0 # model data + ├── MANIFEST.in # to include meta.json + ├── meta.json # model meta data + ├── setup.py # setup file for pip installation + ├── en_example_model # model directory + │ ├── __init__.py # init for pip installation + │ └── en_example_model-1.0.0 # model data + └── dist + └── en_example_model-1.0.0.tar.gz # installable package ``` -You can also find templates for all files on -[GitHub](https://github.com/explosion/spacy-models/tree/master/template). If -you're creating the package manually, keep in mind that the directories need to -be named according to the naming conventions of `lang_name` and +You can also find templates for all files in the +[`cli/package.py` source](https://github.com/explosion/spacy/tree/master/spacy/cli/package.py). +If you're creating the package manually, keep in mind that the directories need +to be named according to the naming conventions of `lang_name` and `lang_name-version`. ### Customizing the model setup {#models-custom} -The meta.json includes the model details, like name, requirements and license, -and lets you customize how the model should be initialized and loaded. You can -define the language data to be loaded and the -[processing pipeline](/usage/processing-pipelines) to execute. - -| Setting | Type | Description | -| ---------- | ---- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -| `lang` | str | ID of the language class to initialize. | -| `pipeline` | list | A list of strings mapping to the IDs of pipeline factories to apply in that order. If not set, spaCy's [default pipeline](/usage/processing-pipelines) will be used. | - The `load()` method that comes with our model package templates will take care of putting all this together and returning a `Language` object with the loaded pipeline and data. If your model requires custom [pipeline components](/usage/processing-pipelines) or a custom language class, -you can also **ship the code with your model**. For examples of this, check out -the implementations of spaCy's -[`load_model_from_init_py`](/api/top-level#util.load_model_from_init_py) and -[`load_model_from_path`](/api/top-level#util.load_model_from_path) utility -functions. - -### Building the model package {#models-building} - -To build the package, run the following command from within the directory. For -more information on building Python packages, see the docs on Python's -[setuptools](https://setuptools.readthedocs.io/en/latest/). - -```bash -$ python setup.py sdist -``` - -This will create a `.tar.gz` archive in a directory `/dist`. The model can be -installed by pointing pip to the path of the archive: - -```bash -$ pip install /path/to/en_example_model-1.0.0.tar.gz -``` - -You can then load the model via its name, `en_example_model`, or import it -directly as a module and then call its `load()` method. +you can also **ship the code with your model** and include it in the +`__init__.py` – for example, to register custom +[pipeline components](/usage/processing-pipelines#custom-components) before the +`nlp` object is created. ### Loading a custom model package {#loading} diff --git a/website/docs/usage/training.md b/website/docs/usage/training.md index b45788e34..8a741215b 100644 --- a/website/docs/usage/training.md +++ b/website/docs/usage/training.md @@ -328,18 +328,15 @@ spaCy's configs are powered by our machine learning library Thinc's [type hints](https://docs.python.org/3/library/typing.html) and even [advanced type annotations](https://thinc.ai/docs/usage-config#advanced-types) using [`pydantic`](https://github.com/samuelcolvin/pydantic). If your registered -function provides For example, `start: int` in the example above will ensure -that the value received as the argument `start` is an integer. If the value -can't be cast to an integer, spaCy will raise an error. +function provides type hints, the values that are passed in will be checked +against the expected types. For example, `start: int` in the example above will +ensure that the value received as the argument `start` is an integer. If the +value can't be cast to an integer, spaCy will raise an error. `start: pydantic.StrictInt` will force the value to be an integer and raise an error if it's not – for instance, if your config defines a float. -### Defining custom architectures {#custom-architectures} - - - ### Wrapping PyTorch and TensorFlow {#custom-frameworks} @@ -352,6 +349,10 @@ mattis pretium. +### Defining custom architectures {#custom-architectures} + + + ## Parallel Training with Ray {#parallel-training} diff --git a/website/docs/usage/transformers.md b/website/docs/usage/transformers.md new file mode 100644 index 000000000..c54165e72 --- /dev/null +++ b/website/docs/usage/transformers.md @@ -0,0 +1,6 @@ +--- +title: Transformers +teaser: Using transformer models like BERT in spaCy +--- + +TODO: ... diff --git a/website/docs/usage/v3.md b/website/docs/usage/v3.md index 13f6e67af..cd68a5242 100644 --- a/website/docs/usage/v3.md +++ b/website/docs/usage/v3.md @@ -6,6 +6,7 @@ menu: - ['New Features', 'features'] - ['Backwards Incompatibilities', 'incompat'] - ['Migrating from v2.x', 'migrating'] + - ['Migrating plugins', 'plugins'] --- ## Summary {#summary} @@ -31,3 +32,74 @@ relied on them. ## Migrating from v2.x {#migrating} + +## Migration notes for plugin maintainers {#plugins} + +Thanks to everyone who's been contributing to the spaCy ecosystem by developing +and maintaining one of the many awesome [plugins and extensions](/universe). +We've tried to keep breaking changes to a minimum and make it as easy as +possible for you to upgrade your packages for spaCy v3. + +### Custom pipeline components + +The most common use case for plugins is providing pipeline components and +extension attributes. + +- Use the [`@Language.factory`](/api/language#factory) decorator to register + your component and assign it a name. This allows users to refer to your + components by name and serialize pipelines referencing them. Remove all manual + entries to the `Language.factories`. +- Make sure your component factories take at least two **named arguments**: + `nlp` (the current `nlp` object) and `name` (the instance name of the added + component so you can identify multiple instances of the same component). +- Update all references to [`nlp.add_pipe`](/api/language#add_pipe) in your docs + to use **string names** instead of the component functions. + +```python +### {highlight="1-5"} +from spacy.language import Language + +@Language.factory("my_component", default_config={"some_setting": False}) +def create_component(nlp: Language, name: str, some_setting: bool): + return MyCoolComponent(some_setting=some_setting) + + +class MyCoolComponent: + def __init__(self, some_setting): + self.some_setting = some_setting + + def __call__(self, doc): + # Do something to the doc + return doc +``` + +> #### Result in config.cfg +> +> ```ini +> [components.my_component] +> factory = "my_component" +> some_setting = true +> ``` + +```diff +import spacy +from your_plugin import MyCoolComponent + +nlp = spacy.load("en_core_web_sm") +- component = MyCoolComponent(some_setting=True) +- nlp.add_pipe(component) ++ nlp.add_pipe("my_component", config={"some_setting": True}) +``` + + + +The [`@Language.factory`](/api/language#factory) decorator takes care of letting +spaCy know that a component of that name is available. This means that your +users can add it to the pipeline using its **string name**. However, this +requires the decorator to be executed – so users will still have to **import +your plugin**. Alternatively, your plugin could expose an +[entry point](/usage/saving-loading#entry-points), which spaCy can read from. +This means that spaCy knows how to initialize `my_component`, even if your +package isn't imported. + + diff --git a/website/docs/usage/vectors-embeddings.md b/website/docs/usage/vectors-embeddings.md index c3a73d4db..7725068ec 100644 --- a/website/docs/usage/vectors-embeddings.md +++ b/website/docs/usage/vectors-embeddings.md @@ -229,3 +229,5 @@ vectors.data = torch.Tensor(vectors.data).cuda(0) ## Other embeddings {#embeddings} + + diff --git a/website/meta/sidebars.json b/website/meta/sidebars.json index 6685e565b..399c39dcf 100644 --- a/website/meta/sidebars.json +++ b/website/meta/sidebars.json @@ -19,6 +19,7 @@ { "text": "Rule-based Matching", "url": "/usage/rule-based-matching" }, { "text": "Processing Pipelines", "url": "/usage/processing-pipelines" }, { "text": "Vectors & Embeddings", "url": "/usage/vectors-embeddings" }, + { "text": "Transformers", "url": "/usage/transformers", "tag": "new" }, { "text": "Training Models", "url": "/usage/training", "tag": "new" }, { "text": "spaCy Projects", "url": "/usage/projects", "tag": "new" }, { "text": "Saving & Loading", "url": "/usage/saving-loading" }, diff --git a/website/src/styles/layout.sass b/website/src/styles/layout.sass index 95f8e1f92..f3d78625a 100644 --- a/website/src/styles/layout.sass +++ b/website/src/styles/layout.sass @@ -414,7 +414,7 @@ body [id]:target .cm-number color: var(--syntax-number) - .cm-def + .cm-def, .cm-meta color: var(--syntax-function) // Jupyter diff --git a/website/src/styles/table.module.sass b/website/src/styles/table.module.sass index 7a82a26fe..4477c43f1 100644 --- a/website/src/styles/table.module.sass +++ b/website/src/styles/table.module.sass @@ -17,7 +17,8 @@ background: var(--color-subtle-opaque) .footer - background: var(--color-theme-light) + --color-inline-code-bg: var(--color-theme-opaque) + background: var(--color-theme-light) !important border-top: 2px solid var(--color-theme) & > td:first-child From 7dd53d0964c0f25f69ef3fcbb78200142d364393 Mon Sep 17 00:00:00 2001 From: Ines Montani Date: Mon, 27 Jul 2020 00:34:00 +0200 Subject: [PATCH 08/10] Fix typo [ci skip] --- website/docs/usage/training.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/website/docs/usage/training.md b/website/docs/usage/training.md index 8a741215b..be0b7196a 100644 --- a/website/docs/usage/training.md +++ b/website/docs/usage/training.md @@ -149,12 +149,12 @@ not just define static settings, but also construct objects like architectures, schedules, optimizers or any other custom components. The main top-level sections of a config file are: -| Section | Description | -| ------------- | -------------------------------------------------------------------------------------------------------------------- | -| `training` | Settings and controls for the training and evaluation process. | -| `pretraining` | Optional settings and controls for the [language model pretraining](#pretraining). | -| `nlp` | Definition of the `nlp` object, its tokenizer and [processing pipeline](/docs/processing-pipelines) component names. | -| `components` | Definitions of the [pipeline components](/docs/processing-pipelines) and their models. | +| Section | Description | +| ------------- | --------------------------------------------------------------------------------------------------------------------- | +| `training` | Settings and controls for the training and evaluation process. | +| `pretraining` | Optional settings and controls for the [language model pretraining](#pretraining). | +| `nlp` | Definition of the `nlp` object, its tokenizer and [processing pipeline](/usage/processing-pipelines) component names. | +| `components` | Definitions of the [pipeline components](/usage/processing-pipelines) and their models. | From 10b84e1e279c3161df9fd06fc82915a41c533709 Mon Sep 17 00:00:00 2001 From: Ines Montani Date: Mon, 27 Jul 2020 16:52:23 +0200 Subject: [PATCH 09/10] Add flag to toggle sdist creation on package [ci skip] --- spacy/cli/package.py | 14 ++++++++------ website/docs/api/cli.md | 8 +++++--- 2 files changed, 13 insertions(+), 9 deletions(-) diff --git a/spacy/cli/package.py b/spacy/cli/package.py index 82b9337a4..a1162f3e1 100644 --- a/spacy/cli/package.py +++ b/spacy/cli/package.py @@ -19,6 +19,7 @@ def package_cli( meta_path: Optional[Path] = Opt(None, "--meta-path", "--meta", "-m", help="Path to meta.json", exists=True, dir_okay=False), create_meta: bool = Opt(False, "--create-meta", "-c", "-C", help="Create meta.json, even if one exists"), version: Optional[str] = Opt(None, "--version", "-v", help="Package version to override meta"), + no_sdist: bool = Opt(False, "--no-sdist", "-NS", help="Don't build .tar.gz sdist, can be set if you want to run this step manually"), force: bool = Opt(False, "--force", "-f", "-F", help="Force overwriting existing model in output directory"), # fmt: on ): @@ -37,6 +38,7 @@ def package_cli( meta_path=meta_path, version=version, create_meta=create_meta, + create_sdist=not no_sdist, force=force, silent=False, ) @@ -48,6 +50,7 @@ def package( meta_path: Optional[Path] = None, version: Optional[str] = None, create_meta: bool = False, + create_sdist: bool = True, force: bool = False, silent: bool = True, ) -> None: @@ -61,7 +64,6 @@ def package( msg.fail("Output directory not found", output_path, exits=1) if meta_path and not meta_path.exists(): msg.fail("Can't find model meta.json", meta_path, exits=1) - meta_path = meta_path or input_dir / "meta.json" if not meta_path.exists() or not meta_path.is_file(): msg.fail("Can't load model meta.json", meta_path, exits=1) @@ -80,7 +82,6 @@ def package( model_name_v = model_name + "-" + meta["version"] main_path = output_dir / model_name_v package_path = main_path / model_name - if package_path.exists(): if force: shutil.rmtree(str(package_path)) @@ -98,10 +99,11 @@ def package( create_file(main_path / "MANIFEST.in", TEMPLATE_MANIFEST) create_file(package_path / "__init__.py", TEMPLATE_INIT) msg.good(f"Successfully created package '{model_name_v}'", main_path) - with util.working_dir(main_path): - util.run_command([sys.executable, "setup.py", "sdist"]) - zip_file = main_path / "dist" / f"{model_name_v}.tar.gz" - msg.good(f"Successfully created zipped Python package", zip_file) + if create_sdist: + with util.working_dir(main_path): + util.run_command([sys.executable, "setup.py", "sdist"]) + zip_file = main_path / "dist" / f"{model_name_v}.tar.gz" + msg.good(f"Successfully created zipped Python package", zip_file) def create_file(file_path: Path, contents: str) -> None: diff --git a/website/docs/api/cli.md b/website/docs/api/cli.md index 5374ff216..4690029aa 100644 --- a/website/docs/api/cli.md +++ b/website/docs/api/cli.md @@ -535,13 +535,14 @@ then create a `.tar.gz` archive file that you can distribute and install with The `spacy package` command now also builds the `.tar.gz` archive automatically, -so you don't have to run `python setup.py sdist` separately anymore. +so you don't have to run `python setup.py sdist` separately anymore. To disable +this, you can set the `--no-sdist` flag. ```bash $ python -m spacy package [input_dir] [output_dir] [--meta-path] [--create-meta] -[--version] [--force] +[--no-sdist] [--version] [--force] ``` > #### Example @@ -557,7 +558,8 @@ $ python -m spacy package [input_dir] [output_dir] [--meta-path] [--create-meta] | `input_dir` | positional | Path to directory containing model data. | | `output_dir` | positional | Directory to create package folder in. | | `--meta-path`, `-m` 2 | option | Path to `meta.json` file (optional). | -| `--create-meta`, `-c` 2 | flag | Create a `meta.json` file on the command line, even if one already exists in the directory. If an existing file is found, its entries will be shown as the defaults in the command line prompt. | +| `--create-meta`, `-C` 2 | flag | Create a `meta.json` file on the command line, even if one already exists in the directory. If an existing file is found, its entries will be shown as the defaults in the command line prompt. | +| `--no-sdist`, `-NS`, | flag | Don't build the `.tar.gz` sdist automatically. Can be set if you want to run this step manually. | | `--version`, `-v` 3 | option | Package version to override in meta. Useful when training new versions, as it doesn't require editing the meta template. | | `--force`, `-f` | flag | Force overwriting of existing folder in output directory. | | `--help`, `-h` | flag | Show help message and available arguments. | From d8b519c23ccd8d6e5898f8bab97fad24ec636101 Mon Sep 17 00:00:00 2001 From: Ines Montani Date: Mon, 27 Jul 2020 18:11:45 +0200 Subject: [PATCH 10/10] API docs, docstrings and argument consistency --- spacy/pipeline/dep_parser.pyx | 9 +- spacy/pipeline/entity_linker.py | 98 ++++++- spacy/pipeline/entityruler.py | 4 + spacy/pipeline/morphologizer.pyx | 89 ++++++- spacy/pipeline/ner.pyx | 7 + spacy/pipeline/pipe.pyx | 4 +- spacy/pipeline/sentencizer.pyx | 34 ++- spacy/pipeline/senter.pyx | 77 +++++- spacy/pipeline/tagger.pyx | 154 ++++++++++- spacy/pipeline/textcat.py | 113 +++++++- spacy/syntax/nn_parser.pyx | 2 +- website/docs/api/dependencymatcher.md | 8 + website/docs/api/dependencyparser.md | 168 +++++++----- website/docs/api/entitylinker.md | 275 ++++++++++--------- website/docs/api/entityrecognizer.md | 184 ++++++++----- website/docs/api/entityruler.md | 82 ++++-- website/docs/api/language.md | 62 +++++ website/docs/api/lemmatizer.md | 2 + website/docs/api/matcher.md | 11 +- website/docs/api/morphologizer.md | 188 +++++++------ website/docs/api/phrasematcher.md | 11 +- website/docs/api/pipe.md | 6 + website/docs/api/sentencerecognizer.md | 350 ++++++++++++++++++++++++- website/docs/api/sentencizer.md | 97 +++++-- website/docs/api/tagger.md | 239 +++++++++++------ website/docs/api/textcategorizer.md | 235 +++++++++++------ website/docs/api/tok2vec.md | 17 +- website/docs/api/top-level.md | 33 +-- website/docs/api/vectors.md | 58 ++-- website/docs/api/vocab.md | 9 +- website/docs/usage/training.md | 29 +- website/docs/usage/v3.md | 183 ++++++++++++- website/gatsby-node.js | 24 +- website/meta/sidebars.json | 17 +- website/src/components/link.js | 11 + website/src/components/title.js | 124 ++++++--- website/src/components/typography.js | 4 +- website/src/styles/table.module.sass | 10 + website/src/styles/title.module.sass | 13 + website/src/templates/docs.js | 2 + 40 files changed, 2323 insertions(+), 720 deletions(-) create mode 100644 website/docs/api/dependencymatcher.md create mode 100644 website/docs/api/pipe.md diff --git a/spacy/pipeline/dep_parser.pyx b/spacy/pipeline/dep_parser.pyx index 78926a984..7a31c200d 100644 --- a/spacy/pipeline/dep_parser.pyx +++ b/spacy/pipeline/dep_parser.pyx @@ -71,7 +71,6 @@ cdef class DependencyParser(Parser): DOCS: https://spacy.io/api/dependencyparser """ - # cdef classes can't have decorators, so we're defining this here TransitionSystem = ArcEager @property @@ -105,6 +104,14 @@ cdef class DependencyParser(Parser): return tuple(sorted(labels)) def score(self, examples, **kwargs): + """Score a batch of examples. + + examples (Iterable[Example]): The examples to score. + RETURNS (Dict[str, Any]): The scores, produced by Scorer.score_spans + and Scorer.score_deps. + + DOCS: https://spacy.io/api/dependencyparser#score + """ def dep_getter(token, attr): dep = getattr(token, attr) dep = token.vocab.strings.as_string(dep).lower() diff --git a/spacy/pipeline/entity_linker.py b/spacy/pipeline/entity_linker.py index 85a4b6174..6c2276dcd 100644 --- a/spacy/pipeline/entity_linker.py +++ b/spacy/pipeline/entity_linker.py @@ -86,6 +86,19 @@ class EntityLinker(Pipe): incl_prior: bool, incl_context: bool, ) -> None: + """Initialize an entity linker. + + vocab (Vocab): The shared vocabulary. + model (thinc.api.Model): The Thinc Model powering the pipeline component. + name (str): The component instance name, used to add entries to the + losses during training. + kb (KnowledgeBase): TODO: + labels_discard (Iterable[str]): TODO: + incl_prior (bool): TODO: + incl_context (bool): TODO: + + DOCS: https://spacy.io/api/entitylinker#init + """ self.vocab = vocab self.model = model self.name = name @@ -119,6 +132,19 @@ class EntityLinker(Pipe): pipeline: Optional[List[Tuple[str, Callable[[Doc], Doc]]]] = None, sgd: Optional[Optimizer] = None, ) -> Optimizer: + """Initialize the pipe for training, using data examples if available. + + get_examples (Callable[[], Iterable[Example]]): Optional function that + returns gold-standard Example objects. + pipeline (List[Tuple[str, Callable]]): Optional list of pipeline + components that this component is part of. Corresponds to + nlp.pipeline. + sgd (thinc.api.Optimizer): Optional optimizer. Will be created with + create_optimizer if it doesn't exist. + RETURNS (thinc.api.Optimizer): The optimizer. + + DOCS: https://spacy.io/api/entitylinker#begin_training + """ self.require_kb() nO = self.kb.entity_vector_length self.set_output(nO) @@ -136,6 +162,20 @@ class EntityLinker(Pipe): sgd: Optional[Optimizer] = None, losses: Optional[Dict[str, float]] = None, ) -> Dict[str, float]: + """Learn from a batch of documents and gold-standard information, + updating the pipe's model. Delegates to predict and get_loss. + + examples (Iterable[Example]): A batch of Example objects. + drop (float): The dropout rate. + set_annotations (bool): Whether or not to update the Example objects + with the predictions. + sgd (thinc.api.Optimizer): The optimizer. + losses (Dict[str, float]): Optional record of the loss during training. + Updated using the component name as the key. + RETURNS (Dict[str, float]): The updated losses dictionary. + + DOCS: https://spacy.io/api/entitylinker#update + """ self.require_kb() if losses is None: losses = {} @@ -215,18 +255,43 @@ class EntityLinker(Pipe): return loss, gradients def __call__(self, doc: Doc) -> Doc: + """Apply the pipe to a Doc. + + doc (Doc): The document to process. + RETURNS (Doc): The processed Doc. + + DOCS: https://spacy.io/api/entitylinker#call + """ kb_ids = self.predict([doc]) self.set_annotations([doc], kb_ids) return doc - def pipe(self, stream: Iterable[Doc], batch_size: int = 128) -> Iterator[Doc]: + def pipe(self, stream: Iterable[Doc], *, batch_size: int = 128) -> Iterator[Doc]: + """Apply the pipe to a stream of documents. This usually happens under + the hood when the nlp object is called on a text and all components are + applied to the Doc. + + stream (Iterable[Doc]): A stream of documents. + batch_size (int): The number of documents to buffer. + YIELDS (Doc): PRocessed documents in order. + + DOCS: https://spacy.io/api/entitylinker#pipe + """ for docs in util.minibatch(stream, size=batch_size): kb_ids = self.predict(docs) self.set_annotations(docs, kb_ids) yield from docs - def predict(self, docs): - """ Return the KB IDs for each entity in each doc, including NIL if there is no prediction """ + def predict(self, docs: Iterable[Doc]) -> List[str]: + """Apply the pipeline's model to a batch of docs, without modifying them. + Returns the KB IDs for each entity in each doc, including NIL if there is + no prediction. + + docs (Iterable[Doc]): The documents to predict. + RETURNS (List[int]): The models prediction for each document. + + DOCS: https://spacy.io/api/entitylinker#predict + """ self.require_kb() entity_count = 0 final_kb_ids = [] @@ -315,7 +380,14 @@ class EntityLinker(Pipe): raise RuntimeError(err) return final_kb_ids - def set_annotations(self, docs: Iterable[Doc], kb_ids: List[int]) -> None: + def set_annotations(self, docs: Iterable[Doc], kb_ids: List[str]) -> None: + """Modify a batch of documents, using pre-computed scores. + + docs (Iterable[Doc]): The documents to modify. + kb_ids (List[str]): The IDs to set, produced by EntityLinker.predict. + + DOCS: https://spacy.io/api/entitylinker#predict + """ count_ents = len([ent for doc in docs for ent in doc.ents]) if count_ents != len(kb_ids): raise ValueError(Errors.E148.format(ents=count_ents, ids=len(kb_ids))) @@ -328,6 +400,13 @@ class EntityLinker(Pipe): token.ent_kb_id_ = kb_id def to_disk(self, path: Union[str, Path], exclude: Iterable[str] = tuple()) -> None: + """Serialize the pipe to disk. + + path (str / Path): Path to a directory. + exclude (Iterable[str]): String names of serialization fields to exclude. + + DOCS: https://spacy.io/api/entitylinker#to_disk + """ serialize = {} self.cfg["entity_width"] = self.kb.entity_vector_length serialize["cfg"] = lambda p: srsly.write_json(p, self.cfg) @@ -339,6 +418,15 @@ class EntityLinker(Pipe): def from_disk( self, path: Union[str, Path], exclude: Iterable[str] = tuple() ) -> "EntityLinker": + """Load the pipe from disk. Modifies the object in place and returns it. + + path (str / Path): Path to a directory. + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (EntityLinker): The modified EntityLinker object. + + DOCS: https://spacy.io/api/entitylinker#from_disk + """ + def load_model(p): try: self.model.from_bytes(p.open("rb").read()) @@ -359,7 +447,7 @@ class EntityLinker(Pipe): util.from_disk(path, deserialize, exclude) return self - def rehearse(self, examples, sgd=None, losses=None, **config): + def rehearse(self, examples, *, sgd=None, losses=None, **config): raise NotImplementedError def add_label(self, label): diff --git a/spacy/pipeline/entityruler.py b/spacy/pipeline/entityruler.py index 869968136..a9833d31a 100644 --- a/spacy/pipeline/entityruler.py +++ b/spacy/pipeline/entityruler.py @@ -71,6 +71,10 @@ class EntityRuler: nlp (Language): The shared nlp object to pass the vocab to the matchers and process phrase patterns. + name (str): Instance name of the current pipeline component. Typically + passed in automatically from the factory when the component is + added. Used to disable the current entity ruler while creating + phrase patterns with the nlp object. phrase_matcher_attr (int / str): Token attribute to match on, passed to the internal PhraseMatcher as `attr` validate (bool): Whether patterns should be validated, passed to diff --git a/spacy/pipeline/morphologizer.pyx b/spacy/pipeline/morphologizer.pyx index fb80a9d86..d3ec3fc90 100644 --- a/spacy/pipeline/morphologizer.pyx +++ b/spacy/pipeline/morphologizer.pyx @@ -61,6 +61,17 @@ class Morphologizer(Tagger): labels_morph: Optional[dict] = None, labels_pos: Optional[dict] = None, ): + """Initialize a morphologizer. + + vocab (Vocab): The shared vocabulary. + model (thinc.api.Model): The Thinc Model powering the pipeline component. + name (str): The component instance name, used to add entries to the + losses during training. + labels_morph (dict): TODO: + labels_pos (dict): TODO: + + DOCS: https://spacy.io/api/morphologizer#init + """ self.vocab = vocab self.model = model self.name = name @@ -77,9 +88,17 @@ class Morphologizer(Tagger): @property def labels(self): + """RETURNS (Tuple[str]): The labels currently added to the component.""" return tuple(self.cfg["labels_morph"].keys()) def add_label(self, label): + """Add a new label to the pipe. + + label (str): The label to add. + RETURNS (int): 1 + + DOCS: https://spacy.io/api/morphologizer#add_label + """ if not isinstance(label, str): raise ValueError(Errors.E187) if label in self.labels: @@ -99,7 +118,20 @@ class Morphologizer(Tagger): self.cfg["labels_pos"][norm_label] = POS_IDS[pos] return 1 - def begin_training(self, get_examples=lambda: [], pipeline=None, sgd=None): + def begin_training(self, get_examples=lambda: [], *, pipeline=None, sgd=None): + """Initialize the pipe for training, using data examples if available. + + get_examples (Callable[[], Iterable[Example]]): Optional function that + returns gold-standard Example objects. + pipeline (List[Tuple[str, Callable]]): Optional list of pipeline + components that this component is part of. Corresponds to + nlp.pipeline. + sgd (thinc.api.Optimizer): Optional optimizer. Will be created with + create_optimizer if it doesn't exist. + RETURNS (thinc.api.Optimizer): The optimizer. + + DOCS: https://spacy.io/api/morphologizer#begin_training + """ for example in get_examples(): for i, token in enumerate(example.reference): pos = token.pos_ @@ -121,6 +153,13 @@ class Morphologizer(Tagger): return sgd def set_annotations(self, docs, batch_tag_ids): + """Modify a batch of documents, using pre-computed scores. + + docs (Iterable[Doc]): The documents to modify. + batch_tag_ids: The IDs to set, produced by Morphologizer.predict. + + DOCS: https://spacy.io/api/morphologizer#predict + """ if isinstance(docs, Doc): docs = [docs] cdef Doc doc @@ -137,6 +176,15 @@ class Morphologizer(Tagger): doc.is_morphed = True def get_loss(self, examples, scores): + """Find the loss and gradient of loss for the batch of documents and + their predicted scores. + + examples (Iterable[Examples]): The batch of examples. + scores: Scores representing the model's predictions. + RETUTNRS (Tuple[float, float]): The loss and the gradient. + + DOCS: https://spacy.io/api/morphologizer#get_loss + """ loss_func = SequenceCategoricalCrossentropy(names=self.labels, normalize=False) truths = [] for eg in examples: @@ -164,6 +212,15 @@ class Morphologizer(Tagger): return float(loss), d_scores def score(self, examples, **kwargs): + """Score a batch of examples. + + examples (Iterable[Example]): The examples to score. + RETURNS (Dict[str, Any]): The scores, produced by + Scorer.score_token_attr for the attributes "pos" and "morph" and + Scorer.score_token_attr_per_feat for the attribute "morph". + + DOCS: https://spacy.io/api/morphologizer#score + """ results = {} results.update(Scorer.score_token_attr(examples, "pos", **kwargs)) results.update(Scorer.score_token_attr(examples, "morph", **kwargs)) @@ -172,6 +229,13 @@ class Morphologizer(Tagger): return results def to_bytes(self, exclude=tuple()): + """Serialize the pipe to a bytestring. + + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (bytes): The serialized object. + + DOCS: https://spacy.io/api/morphologizer#to_bytes + """ serialize = {} serialize["model"] = self.model.to_bytes serialize["vocab"] = self.vocab.to_bytes @@ -179,6 +243,14 @@ class Morphologizer(Tagger): return util.to_bytes(serialize, exclude) def from_bytes(self, bytes_data, exclude=tuple()): + """Load the pipe from a bytestring. + + bytes_data (bytes): The serialized pipe. + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (Morphologizer): The loaded Morphologizer. + + DOCS: https://spacy.io/api/morphologizer#from_bytes + """ def load_model(b): try: self.model.from_bytes(b) @@ -194,6 +266,13 @@ class Morphologizer(Tagger): return self def to_disk(self, path, exclude=tuple()): + """Serialize the pipe to disk. + + path (str / Path): Path to a directory. + exclude (Iterable[str]): String names of serialization fields to exclude. + + DOCS: https://spacy.io/api/morphologizer#to_disk + """ serialize = { "vocab": lambda p: self.vocab.to_disk(p), "model": lambda p: p.open("wb").write(self.model.to_bytes()), @@ -202,6 +281,14 @@ class Morphologizer(Tagger): util.to_disk(path, serialize, exclude) def from_disk(self, path, exclude=tuple()): + """Load the pipe from disk. Modifies the object in place and returns it. + + path (str / Path): Path to a directory. + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (Morphologizer): The modified Morphologizer object. + + DOCS: https://spacy.io/api/morphologizer#from_disk + """ def load_model(p): with p.open("rb") as file_: try: diff --git a/spacy/pipeline/ner.pyx b/spacy/pipeline/ner.pyx index cb2ca89d8..8d31e441e 100644 --- a/spacy/pipeline/ner.pyx +++ b/spacy/pipeline/ner.pyx @@ -91,4 +91,11 @@ cdef class EntityRecognizer(Parser): return tuple(sorted(labels)) def score(self, examples, **kwargs): + """Score a batch of examples. + + examples (Iterable[Example]): The examples to score. + RETURNS (Dict[str, Any]): The scores, produced by Scorer.score_spans. + + DOCS: https://spacy.io/api/entityrecognizer#score + """ return Scorer.score_spans(examples, "ents", **kwargs) diff --git a/spacy/pipeline/pipe.pyx b/spacy/pipeline/pipe.pyx index e7702aa59..0b0aeec62 100644 --- a/spacy/pipeline/pipe.pyx +++ b/spacy/pipeline/pipe.pyx @@ -23,7 +23,7 @@ class Pipe: name = None - def __init__(self, vocab, model, **cfg): + def __init__(self, vocab, model, name, **cfg): """Create a new pipe instance.""" raise NotImplementedError @@ -79,7 +79,7 @@ class Pipe: def create_optimizer(self): return create_default_optimizer() - def begin_training(self, get_examples=lambda: [], pipeline=None, sgd=None): + def begin_training(self, get_examples=lambda: [], *, pipeline=None, sgd=None): """Initialize the pipe for training, using data exampes if available. If no model has been initialized yet, the model is added.""" self.model.initialize() diff --git a/spacy/pipeline/sentencizer.pyx b/spacy/pipeline/sentencizer.pyx index 70188f856..4c0e432f5 100644 --- a/spacy/pipeline/sentencizer.pyx +++ b/spacy/pipeline/sentencizer.pyx @@ -41,7 +41,7 @@ class Sentencizer(Pipe): '𑩃', '𑪛', '𑪜', '𑱁', '𑱂', '𖩮', '𖩯', '𖫵', '𖬷', '𖬸', '𖭄', '𛲟', '𝪈', '｡', '。'] - def __init__(self, name="sentencizer", *, punct_chars): + def __init__(self, name="sentencizer", *, punct_chars=None): """Initialize the sentencizer. punct_chars (list): Punctuation characters to split on. Will be @@ -62,8 +62,8 @@ class Sentencizer(Pipe): def __call__(self, doc): """Apply the sentencizer to a Doc and set Token.is_sent_start. - example (Doc or Example): The document to process. - RETURNS (Doc or Example): The processed Doc or Example. + doc (Doc): The document to process. + RETURNS (Doc): The processed Doc. DOCS: https://spacy.io/api/sentencizer#call """ @@ -83,14 +83,26 @@ class Sentencizer(Pipe): return doc def pipe(self, stream, batch_size=128): + """Apply the pipe to a stream of documents. This usually happens under + the hood when the nlp object is called on a text and all components are + applied to the Doc. + + stream (Iterable[Doc]): A stream of documents. + batch_size (int): The number of documents to buffer. + YIELDS (Doc): Processed documents in order. + + DOCS: https://spacy.io/api/sentencizer#pipe + """ for docs in util.minibatch(stream, size=batch_size): predictions = self.predict(docs) self.set_annotations(docs, predictions) yield from docs def predict(self, docs): - """Apply the pipeline's model to a batch of docs, without - modifying them. + """Apply the pipe to a batch of docs, without modifying them. + + docs (Iterable[Doc]): The documents to predict. + RETURNS: The predictions for each document. """ if not any(len(doc) for doc in docs): # Handle cases where there are no tokens in any docs. @@ -117,6 +129,11 @@ class Sentencizer(Pipe): return guesses def set_annotations(self, docs, batch_tag_ids): + """Modify a batch of documents, using pre-computed scores. + + docs (Iterable[Doc]): The documents to modify. + scores: The tag IDs produced by Sentencizer.predict. + """ if isinstance(docs, Doc): docs = [docs] cdef Doc doc @@ -132,6 +149,13 @@ class Sentencizer(Pipe): doc.c[j].sent_start = -1 def score(self, examples, **kwargs): + """Score a batch of examples. + + examples (Iterable[Example]): The examples to score. + RETURNS (Dict[str, Any]): The scores, produced by Scorer.score_spans. + + DOCS: https://spacy.io/api/sentencizer#score + """ return Scorer.score_spans(examples, "sents", **kwargs) def to_bytes(self, exclude=tuple()): diff --git a/spacy/pipeline/senter.pyx b/spacy/pipeline/senter.pyx index c065ae72f..ae753ae60 100644 --- a/spacy/pipeline/senter.pyx +++ b/spacy/pipeline/senter.pyx @@ -45,6 +45,15 @@ class SentenceRecognizer(Tagger): DOCS: https://spacy.io/api/sentencerecognizer """ def __init__(self, vocab, model, name="senter"): + """Initialize a sentence recognizer. + + vocab (Vocab): The shared vocabulary. + model (thinc.api.Model): The Thinc Model powering the pipeline component. + name (str): The component instance name, used to add entries to the + losses during training. + + DOCS: https://spacy.io/api/sentencerecognizer#init + """ self.vocab = vocab self.model = model self.name = name @@ -53,12 +62,20 @@ class SentenceRecognizer(Tagger): @property def labels(self): + """RETURNS (Tuple[str]): The labels.""" # labels are numbered by index internally, so this matches GoldParse # and Example where the sentence-initial tag is 1 and other positions # are 0 return tuple(["I", "S"]) def set_annotations(self, docs, batch_tag_ids): + """Modify a batch of documents, using pre-computed scores. + + docs (Iterable[Doc]): The documents to modify. + batch_tag_ids: The IDs to set, produced by SentenceRecognizer.predict. + + DOCS: https://spacy.io/api/sentencerecognizer#predict + """ if isinstance(docs, Doc): docs = [docs] cdef Doc doc @@ -75,6 +92,15 @@ class SentenceRecognizer(Tagger): doc.c[j].sent_start = -1 def get_loss(self, examples, scores): + """Find the loss and gradient of loss for the batch of documents and + their predicted scores. + + examples (Iterable[Examples]): The batch of examples. + scores: Scores representing the model's predictions. + RETUTNRS (Tuple[float, float]): The loss and the gradient. + + DOCS: https://spacy.io/api/sentencerecognizer#get_loss + """ labels = self.labels loss_func = SequenceCategoricalCrossentropy(names=labels, normalize=False) truths = [] @@ -94,7 +120,20 @@ class SentenceRecognizer(Tagger): raise ValueError("nan value when computing loss") return float(loss), d_scores - def begin_training(self, get_examples=lambda: [], pipeline=None, sgd=None): + def begin_training(self, get_examples=lambda: [], *, pipeline=None, sgd=None): + """Initialize the pipe for training, using data examples if available. + + get_examples (Callable[[], Iterable[Example]]): Optional function that + returns gold-standard Example objects. + pipeline (List[Tuple[str, Callable]]): Optional list of pipeline + components that this component is part of. Corresponds to + nlp.pipeline. + sgd (thinc.api.Optimizer): Optional optimizer. Will be created with + create_optimizer if it doesn't exist. + RETURNS (thinc.api.Optimizer): The optimizer. + + DOCS: https://spacy.io/api/sentencerecognizer#begin_training + """ self.set_output(len(self.labels)) self.model.initialize() util.link_vectors_to_models(self.vocab) @@ -106,9 +145,22 @@ class SentenceRecognizer(Tagger): raise NotImplementedError def score(self, examples, **kwargs): + """Score a batch of examples. + + examples (Iterable[Example]): The examples to score. + RETURNS (Dict[str, Any]): The scores, produced by Scorer.score_spans. + DOCS: https://spacy.io/api/sentencerecognizer#score + """ return Scorer.score_spans(examples, "sents", **kwargs) def to_bytes(self, exclude=tuple()): + """Serialize the pipe to a bytestring. + + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (bytes): The serialized object. + + DOCS: https://spacy.io/api/sentencerecognizer#to_bytes + """ serialize = {} serialize["model"] = self.model.to_bytes serialize["vocab"] = self.vocab.to_bytes @@ -116,6 +168,14 @@ class SentenceRecognizer(Tagger): return util.to_bytes(serialize, exclude) def from_bytes(self, bytes_data, exclude=tuple()): + """Load the pipe from a bytestring. + + bytes_data (bytes): The serialized pipe. + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (Tagger): The loaded SentenceRecognizer. + + DOCS: https://spacy.io/api/sentencerecognizer#from_bytes + """ def load_model(b): try: self.model.from_bytes(b) @@ -131,6 +191,13 @@ class SentenceRecognizer(Tagger): return self def to_disk(self, path, exclude=tuple()): + """Serialize the pipe to disk. + + path (str / Path): Path to a directory. + exclude (Iterable[str]): String names of serialization fields to exclude. + + DOCS: https://spacy.io/api/sentencerecognizer#to_disk + """ serialize = { "vocab": lambda p: self.vocab.to_disk(p), "model": lambda p: p.open("wb").write(self.model.to_bytes()), @@ -139,6 +206,14 @@ class SentenceRecognizer(Tagger): util.to_disk(path, serialize, exclude) def from_disk(self, path, exclude=tuple()): + """Load the pipe from disk. Modifies the object in place and returns it. + + path (str / Path): Path to a directory. + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (Tagger): The modified SentenceRecognizer object. + + DOCS: https://spacy.io/api/sentencerecognizer#from_disk + """ def load_model(p): with p.open("rb") as file_: try: diff --git a/spacy/pipeline/tagger.pyx b/spacy/pipeline/tagger.pyx index 7cc11fb84..a80d516df 100644 --- a/spacy/pipeline/tagger.pyx +++ b/spacy/pipeline/tagger.pyx @@ -51,6 +51,16 @@ class Tagger(Pipe): DOCS: https://spacy.io/api/tagger """ def __init__(self, vocab, model, name="tagger", *, set_morphology=False): + """Initialize a part-of-speech tagger. + + vocab (Vocab): The shared vocabulary. + model (thinc.api.Model): The Thinc Model powering the pipeline component. + name (str): The component instance name, used to add entries to the + losses during training. + set_morphology (bool): Whether to set morphological features. + + DOCS: https://spacy.io/api/tagger#init + """ self.vocab = vocab self.model = model self.name = name @@ -60,20 +70,52 @@ class Tagger(Pipe): @property def labels(self): + """The labels currently added to the component. Note that even for a + blank component, this will always include the built-in coarse-grained + part-of-speech tags by default. + + RETURNS (Tuple[str]): The labels. + + DOCS: https://spacy.io/api/tagger#labels + """ return tuple(self.vocab.morphology.tag_names) def __call__(self, doc): + """Apply the pipe to a Doc. + + doc (Doc): The document to process. + RETURNS (Doc): The processed Doc. + + DOCS: https://spacy.io/api/tagger#call + """ tags = self.predict([doc]) self.set_annotations([doc], tags) return doc - def pipe(self, stream, batch_size=128): + def pipe(self, stream, *, batch_size=128): + """Apply the pipe to a stream of documents. This usually happens under + the hood when the nlp object is called on a text and all components are + applied to the Doc. + + stream (Iterable[Doc]): A stream of documents. + batch_size (int): The number of documents to buffer. + YIELDS (Doc): Processed documents in order. + + DOCS: https://spacy.io/api/tagger#pipe + """ for docs in util.minibatch(stream, size=batch_size): tag_ids = self.predict(docs) self.set_annotations(docs, tag_ids) yield from docs def predict(self, docs): + """Apply the pipeline's model to a batch of docs, without modifying them. + + docs (Iterable[Doc]): The documents to predict. + RETURNS: The models prediction for each document. + + DOCS: https://spacy.io/api/tagger#predict + """ if not any(len(doc) for doc in docs): # Handle cases where there are no tokens in any docs. n_labels = len(self.labels) @@ -96,6 +138,13 @@ class Tagger(Pipe): return guesses def set_annotations(self, docs, batch_tag_ids): + """Modify a batch of documents, using pre-computed scores. + + docs (Iterable[Doc]): The documents to modify. + batch_tag_ids: The IDs to set, produced by Tagger.predict. + + DOCS: https://spacy.io/api/tagger#predict + """ if isinstance(docs, Doc): docs = [docs] cdef Doc doc @@ -121,10 +170,23 @@ class Tagger(Pipe): doc.is_tagged = True def update(self, examples, *, drop=0., sgd=None, losses=None, set_annotations=False): + """Learn from a batch of documents and gold-standard information, + updating the pipe's model. Delegates to predict and get_loss. + + examples (Iterable[Example]): A batch of Example objects. + drop (float): The dropout rate. + set_annotations (bool): Whether or not to update the Example objects + with the predictions. + sgd (thinc.api.Optimizer): The optimizer. + losses (Dict[str, float]): Optional record of the loss during training. + Updated using the component name as the key. + RETURNS (Dict[str, float]): The updated losses dictionary. + + DOCS: https://spacy.io/api/tagger#update + """ if losses is None: losses = {} losses.setdefault(self.name, 0.0) - try: if not any(len(eg.predicted) if eg.predicted else 0 for eg in examples): # Handle cases where there are no tokens in any docs. @@ -149,9 +211,20 @@ class Tagger(Pipe): self.set_annotations(docs, self._scores2guesses(tag_scores)) return losses - def rehearse(self, examples, drop=0., sgd=None, losses=None): - """Perform a 'rehearsal' update, where we try to match the output of - an initial model. + def rehearse(self, examples, *, drop=0., sgd=None, losses=None): + """Perform a "rehearsal" update from a batch of data. Rehearsal updates + teach the current model to make predictions similar to an initial model, + to try to address the "catastrophic forgetting" problem. This feature is + experimental. + + examples (Iterable[Example]): A batch of Example objects. + drop (float): The dropout rate. + sgd (thinc.api.Optimizer): The optimizer. + losses (Dict[str, float]): Optional record of the loss during training. + Updated using the component name as the key. + RETURNS (Dict[str, float]): The updated losses dictionary. + + DOCS: https://spacy.io/api/tagger#rehearse """ try: docs = [eg.predicted for eg in examples] @@ -174,6 +247,15 @@ class Tagger(Pipe): losses[self.name] += (gradient**2).sum() def get_loss(self, examples, scores): + """Find the loss and gradient of loss for the batch of documents and + their predicted scores. + + examples (Iterable[Examples]): The batch of examples. + scores: Scores representing the model's predictions. + RETUTNRS (Tuple[float, float]): The loss and the gradient. + + DOCS: https://spacy.io/api/tagger#get_loss + """ loss_func = SequenceCategoricalCrossentropy(names=self.labels, normalize=False) truths = [eg.get_aligned("tag", as_string=True) for eg in examples] d_scores, loss = loss_func(scores, truths) @@ -181,7 +263,20 @@ class Tagger(Pipe): raise ValueError("nan value when computing loss") return float(loss), d_scores - def begin_training(self, get_examples=lambda: [], pipeline=None, sgd=None): + def begin_training(self, get_examples=lambda: [], *, pipeline=None, sgd=None): + """Initialize the pipe for training, using data examples if available. + + get_examples (Callable[[], Iterable[Example]]): Optional function that + returns gold-standard Example objects. + pipeline (List[Tuple[str, Callable]]): Optional list of pipeline + components that this component is part of. Corresponds to + nlp.pipeline. + sgd (thinc.api.Optimizer): Optional optimizer. Will be created with + create_optimizer if it doesn't exist. + RETURNS (thinc.api.Optimizer): The optimizer. + + DOCS: https://spacy.io/api/tagger#begin_training + """ lemma_tables = ["lemma_rules", "lemma_index", "lemma_exc", "lemma_lookup"] if not any(table in self.vocab.lookups for table in lemma_tables): warnings.warn(Warnings.W022) @@ -227,6 +322,15 @@ class Tagger(Pipe): return sgd def add_label(self, label, values=None): + """Add a new label to the pipe. + + label (str): The label to add. + values (Dict[int, str]): Optional values to map to the label, e.g. a + tag map dictionary. + RETURNS (int): 1 + + DOCS: https://spacy.io/api/tagger#add_label + """ if not isinstance(label, str): raise ValueError(Errors.E187) if label in self.labels: @@ -254,6 +358,14 @@ class Tagger(Pipe): yield def score(self, examples, **kwargs): + """Score a batch of examples. + + examples (Iterable[Example]): The examples to score. + RETURNS (Dict[str, Any]): The scores, produced by + Scorer.score_token_attr for the attributes "tag", "pos" and "lemma". + + DOCS: https://spacy.io/api/tagger#score + """ scores = {} scores.update(Scorer.score_token_attr(examples, "tag", **kwargs)) scores.update(Scorer.score_token_attr(examples, "pos", **kwargs)) @@ -261,6 +373,13 @@ class Tagger(Pipe): return scores def to_bytes(self, exclude=tuple()): + """Serialize the pipe to a bytestring. + + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (bytes): The serialized object. + + DOCS: https://spacy.io/api/tagger#to_bytes + """ serialize = {} serialize["model"] = self.model.to_bytes serialize["vocab"] = self.vocab.to_bytes @@ -272,6 +391,14 @@ class Tagger(Pipe): return util.to_bytes(serialize, exclude) def from_bytes(self, bytes_data, exclude=tuple()): + """Load the pipe from a bytestring. + + bytes_data (bytes): The serialized pipe. + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (Tagger): The loaded Tagger. + + DOCS: https://spacy.io/api/tagger#from_bytes + """ def load_model(b): try: self.model.from_bytes(b) @@ -300,6 +427,13 @@ class Tagger(Pipe): return self def to_disk(self, path, exclude=tuple()): + """Serialize the pipe to disk. + + path (str / Path): Path to a directory. + exclude (Iterable[str]): String names of serialization fields to exclude. + + DOCS: https://spacy.io/api/tagger#to_disk + """ tag_map = dict(sorted(self.vocab.morphology.tag_map.items())) morph_rules = dict(self.vocab.morphology.exc) serialize = { @@ -312,6 +446,14 @@ class Tagger(Pipe): util.to_disk(path, serialize, exclude) def from_disk(self, path, exclude=tuple()): + """Load the pipe from disk. Modifies the object in place and returns it. + + path (str / Path): Path to a directory. + exclude (Iterable[str]): String names of serialization fields to exclude. + RETURNS (Tagger): The modified Tagger object. + + DOCS: https://spacy.io/api/tagger#from_disk + """ def load_model(p): with p.open("rb") as file_: try: diff --git a/spacy/pipeline/textcat.py b/spacy/pipeline/textcat.py index 628ff250b..1f1754cb0 100644 --- a/spacy/pipeline/textcat.py +++ b/spacy/pipeline/textcat.py @@ -77,6 +77,16 @@ class TextCategorizer(Pipe): *, labels: Iterable[str], ) -> None: + """Initialize a text categorizer. + + vocab (Vocab): The shared vocabulary. + model (thinc.api.Model): The Thinc Model powering the pipeline component. + name (str): The component instance name, used to add entries to the + losses during training. + labels (Iterable[str]): The labels to use. + + DOCS: https://spacy.io/api/textcategorizer#init + """ self.vocab = vocab self.model = model self.name = name @@ -86,6 +96,10 @@ class TextCategorizer(Pipe): @property def labels(self) -> Tuple[str]: + """RETURNS (Tuple[str]): The labels currently added to the component. + + DOCS: https://spacy.io/api/textcategorizer#labels + """ return tuple(self.cfg.setdefault("labels", [])) def require_labels(self) -> None: @@ -97,13 +111,30 @@ class TextCategorizer(Pipe): def labels(self, value: Iterable[str]) -> None: self.cfg["labels"] = tuple(value) - def pipe(self, stream: Iterator[str], batch_size: int = 128) -> Iterator[Doc]: + def pipe(self, stream: Iterable[Doc], *, batch_size: int = 128) -> Iterator[Doc]: + """Apply the pipe to a stream of documents. This usually happens under + the hood when the nlp object is called on a text and all components are + applied to the Doc. + + stream (Iterable[Doc]): A stream of documents. + batch_size (int): The number of documents to buffer. + YIELDS (Doc): PRocessed documents in order. + + DOCS: https://spacy.io/api/textcategorizer#pipe + """ for docs in util.minibatch(stream, size=batch_size): scores = self.predict(docs) self.set_annotations(docs, scores) yield from docs def predict(self, docs: Iterable[Doc]): + """Apply the pipeline's model to a batch of docs, without modifying them. + + docs (Iterable[Doc]): The documents to predict. + RETURNS: The models prediction for each document. + + DOCS: https://spacy.io/api/textcategorizer#predict + """ tensors = [doc.tensor for doc in docs] if not any(len(doc) for doc in docs): # Handle cases where there are no tokens in any docs. @@ -115,6 +146,13 @@ class TextCategorizer(Pipe): return scores def set_annotations(self, docs: Iterable[Doc], scores) -> None: + """Modify a batch of documents, using pre-computed scores. + + docs (Iterable[Doc]): The documents to modify. + scores: The scores to set, produced by TextCategorizer.predict. + + DOCS: https://spacy.io/api/textcategorizer#predict + """ for i, doc in enumerate(docs): for j, label in enumerate(self.labels): doc.cats[label] = float(scores[i, j]) @@ -128,6 +166,20 @@ class TextCategorizer(Pipe): sgd: Optional[Optimizer] = None, losses: Optional[Dict[str, float]] = None, ) -> Dict[str, float]: + """Learn from a batch of documents and gold-standard information, + updating the pipe's model. Delegates to predict and get_loss. + + examples (Iterable[Example]): A batch of Example objects. + drop (float): The dropout rate. + set_annotations (bool): Whether or not to update the Example objects + with the predictions. + sgd (thinc.api.Optimizer): The optimizer. + losses (Dict[str, float]): Optional record of the loss during training. + Updated using the component name as the key. + RETURNS (Dict[str, float]): The updated losses dictionary. + + DOCS: https://spacy.io/api/textcategorizer#update + """ if losses is None: losses = {} losses.setdefault(self.name, 0.0) @@ -155,10 +207,25 @@ class TextCategorizer(Pipe): def rehearse( self, examples: Iterable[Example], + *, drop: float = 0.0, sgd: Optional[Optimizer] = None, losses: Optional[Dict[str, float]] = None, - ) -> None: + ) -> Dict[str, float]: + """Perform a "rehearsal" update from a batch of data. Rehearsal updates + teach the current model to make predictions similar to an initial model, + to try to address the "catastrophic forgetting" problem. This feature is + experimental. + + examples (Iterable[Example]): A batch of Example objects. + drop (float): The dropout rate. + sgd (thinc.api.Optimizer): The optimizer. + losses (Dict[str, float]): Optional record of the loss during training. + Updated using the component name as the key. + RETURNS (Dict[str, float]): The updated losses dictionary. + + DOCS: https://spacy.io/api/textcategorizer#rehearse + """ if self._rehearsal_model is None: return try: @@ -182,6 +249,7 @@ class TextCategorizer(Pipe): if losses is not None: losses.setdefault(self.name, 0.0) losses[self.name] += (gradient ** 2).sum() + return losses def _examples_to_truth( self, examples: List[Example] @@ -198,6 +266,15 @@ class TextCategorizer(Pipe): return truths, not_missing def get_loss(self, examples: Iterable[Example], scores) -> Tuple[float, float]: + """Find the loss and gradient of loss for the batch of documents and + their predicted scores. + + examples (Iterable[Examples]): The batch of examples. + scores: Scores representing the model's predictions. + RETUTNRS (Tuple[float, float]): The loss and the gradient. + + DOCS: https://spacy.io/api/textcategorizer#get_loss + """ truths, not_missing = self._examples_to_truth(examples) not_missing = self.model.ops.asarray(not_missing) d_scores = (scores - truths) / scores.shape[0] @@ -206,6 +283,13 @@ class TextCategorizer(Pipe): return float(mean_square_error), d_scores def add_label(self, label: str) -> int: + """Add a new label to the pipe. + + label (str): The label to add. + RETURNS (int): 1. + + DOCS: https://spacy.io/api/textcategorizer#add_label + """ if not isinstance(label, str): raise ValueError(Errors.E187) if label in self.labels: @@ -226,10 +310,24 @@ class TextCategorizer(Pipe): def begin_training( self, - get_examples: Callable = lambda: [], + get_examples: Callable[[], Iterable[Example]] = lambda: [], + *, pipeline: Optional[List[Tuple[str, Callable[[Doc], Doc]]]] = None, sgd: Optional[Optimizer] = None, ) -> Optimizer: + """Initialize the pipe for training, using data examples if available. + + get_examples (Callable[[], Iterable[Example]]): Optional function that + returns gold-standard Example objects. + pipeline (List[Tuple[str, Callable]]): Optional list of pipeline + components that this component is part of. Corresponds to + nlp.pipeline. + sgd (thinc.api.Optimizer): Optional optimizer. Will be created with + create_optimizer if it doesn't exist. + RETURNS (thinc.api.Optimizer): The optimizer. + + DOCS: https://spacy.io/api/textcategorizer#begin_training + """ # TODO: begin_training is not guaranteed to see all data / labels ? examples = list(get_examples()) for example in examples: @@ -255,9 +353,18 @@ class TextCategorizer(Pipe): def score( self, examples: Iterable[Example], + *, positive_label: Optional[str] = None, **kwargs, ) -> Dict[str, Any]: + """Score a batch of examples. + + examples (Iterable[Example]): The examples to score. + positive_label (str): Optional positive label. + RETURNS (Dict[str, Any]): The scores, produced by Scorer.score_cats. + + DOCS: https://spacy.io/api/textcategorizer#score + """ return Scorer.score_cats( examples, "cats", diff --git a/spacy/syntax/nn_parser.pyx b/spacy/syntax/nn_parser.pyx index f640e2e8d..7f115acd8 100644 --- a/spacy/syntax/nn_parser.pyx +++ b/spacy/syntax/nn_parser.pyx @@ -160,7 +160,7 @@ cdef class Parser: self.set_annotations([doc], states) return doc - def pipe(self, docs, int batch_size=256): + def pipe(self, docs, *, int batch_size=256): """Process a stream of documents. stream: The sequence of documents to process. diff --git a/website/docs/api/dependencymatcher.md b/website/docs/api/dependencymatcher.md new file mode 100644 index 000000000..3638575df --- /dev/null +++ b/website/docs/api/dependencymatcher.md @@ -0,0 +1,8 @@ +--- +title: DependencyMatcher +teaser: Match sequences of tokens, based on the dependency parse +tag: class +source: spacy/matcher/dependencymatcher.pyx +--- + +TODO: write diff --git a/website/docs/api/dependencyparser.md b/website/docs/api/dependencyparser.md index 52442fe49..2a65f61d7 100644 --- a/website/docs/api/dependencyparser.md +++ b/website/docs/api/dependencyparser.md @@ -2,18 +2,37 @@ title: DependencyParser tag: class source: spacy/pipeline/dep_parser.pyx +teaser: 'Pipeline component for syntactic dependency parsing' +api_base_class: /api/pipe +api_string_name: parser +api_trainable: true --- -This class is a subclass of `Pipe` and follows the same API. The pipeline -component is available in the [processing pipeline](/usage/processing-pipelines) -via the ID `"parser"`. +## Config and implementation {#config} -## Implementation and defaults {#implementation} +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). See the +[model architectures](/api/architectures) documentation for details on the +architectures and their arguments and hyperparameters. -See the [model architectures](/api/architectures) documentation for details on -the architectures and their arguments and hyperparameters. To learn more about -how to customize the config and train custom models, check out the -[training config](/usage/training#config) docs. +> #### Example +> +> ```python +> from spacy.pipeline.dep_parser import DEFAULT_PARSER_MODEL +> config = { +> "moves": None, +> # TODO: rest +> "model": DEFAULT_PARSER_MODEL, +> } +> nlp.add_pipe("parser", config=config) +> ``` + +| Setting | Type | Description | Default | +| ------- | ------------------------------------------ | ----------------- | ----------------------------------------------------------------- | +| `moves` | list | | `None` | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The model to use. | [TransitionBasedParser](/api/architectures#TransitionBasedParser) | ```python https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/dep_parser.pyx @@ -30,18 +49,27 @@ https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/dep_parser.pyx > # Construction via add_pipe with custom model > config = {"model": {"@architectures": "my_parser"}} > parser = nlp.add_pipe("parser", config=config) +> +> # Construction from class +> from spacy.pipeline import DependencyParser +> parser = DependencyParser(nlp.vocab, model) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and [`nlp.add_pipe`](/api/language#add_pipe). -| Name | Type | Description | -| ----------- | ------------------ | ------------------------------------------------------------------------------- | -| `vocab` | `Vocab` | The shared vocabulary. | -| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | -| `**cfg` | - | Configuration parameters. | -| **RETURNS** | `DependencyParser` | The newly constructed object. | +| Name | Type | Description | +| ----------------------------- | ------------------------------------------ | ------------------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | The shared vocabulary. | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | +| `name` | str | String name of the component instance. Used to add entries to the `losses` during training. | +| `moves` | list | | +| _keyword-only_ | | | +| `update_with_oracle_cut_size` | int | | +| `multitasks` | `Iterable` | | +| `learn_tokens` | bool | | +| `min_action_freq` | int | | ## DependencyParser.\_\_call\_\_ {#call tag="method"} @@ -56,8 +84,8 @@ and all pipeline components are applied to the `Doc` in order. Both > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) > doc = nlp("This is a sentence.") +> parser = nlp.add_pipe("parser") > # This usually happens under the hood > processed = parser(doc) > ``` @@ -79,16 +107,37 @@ applied to the `Doc` in order. Both [`__call__`](/api/dependencyparser#call) and > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > for doc in parser.pipe(docs, batch_size=50): > pass > ``` -| Name | Type | Description | -| ------------ | --------------- | ------------------------------------------------------ | -| `stream` | `Iterable[Doc]` | A stream of documents. | -| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | -| **YIELDS** | `Doc` | Processed documents in the order of the original text. | +| Name | Type | Description | +| -------------- | --------------- | ------------------------------------------------------ | +| `stream` | `Iterable[Doc]` | A stream of documents. | +| _keyword-only_ | | | +| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | +| **YIELDS** | `Doc` | Processed documents in the order of the original text. | + +## DependencyParser.begin_training {#begin_training tag="method"} + +Initialize the pipe for training, using data examples if available. Return an +[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. + +> #### Example +> +> ```python +> parser = nlp.add_pipe("parser") +> optimizer = parser.begin_training(pipeline=nlp.pipeline) +> ``` + +| Name | Type | Description | +| -------------- | --------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------- | +| `get_examples` | `Callable[[], Iterable[Example]]` | Optional function that returns gold-standard annotations in the form of [`Example`](/api/example) objects. | +| _keyword-only_ | | | +| `pipeline` | `List[Tuple[str, Callable]]` | Optional list of pipeline components that this component is part of. | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | An optional optimizer. Will be created via [`create_optimizer`](/api/dependencyparser#create_optimizer) if not set. | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## DependencyParser.predict {#predict tag="method"} @@ -97,7 +146,7 @@ Apply the pipeline's model to a batch of docs, without modifying them. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > scores = parser.predict([doc1, doc2]) > ``` @@ -113,7 +162,7 @@ Modify a batch of documents, using pre-computed scores. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > scores = parser.predict([doc1, doc2]) > parser.set_annotations([doc1, doc2], scores) > ``` @@ -132,7 +181,7 @@ model. Delegates to [`predict`](/api/dependencyparser#predict) and > #### Example > > ```python -> parser = DependencyParser(nlp.vocab, parser_model) +> parser = nlp.add_pipe("parser") > optimizer = nlp.begin_training() > losses = parser.update(examples, sgd=optimizer) > ``` @@ -144,7 +193,7 @@ model. Delegates to [`predict`](/api/dependencyparser#predict) and | `drop` | float | The dropout rate. | | `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/dependencyparser#set_annotations). | | `sgd` | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | -| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. Updated using the component name as the key. | | **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | ## DependencyParser.get_loss {#get_loss tag="method"} @@ -155,36 +204,31 @@ predicted scores. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > scores = parser.predict([eg.predicted for eg in examples]) > loss, d_loss = parser.get_loss(examples, scores) > ``` -| Name | Type | Description | -| ----------- | ------------------- | --------------------------------------------------- | -| `examples` | `Iterable[Example]` | The batch of examples. | -| `scores` | `syntax.StateClass` | Scores representing the model's predictions. | -| **RETURNS** | tuple | The loss and the gradient, i.e. `(loss, gradient)`. | +| Name | Type | Description | +| ----------- | --------------------- | --------------------------------------------------- | +| `examples` | `Iterable[Example]` | The batch of examples. | +| `scores` | `syntax.StateClass` | Scores representing the model's predictions. | +| **RETURNS** | `Tuple[float, float]` | The loss and the gradient, i.e. `(loss, gradient)`. | -## DependencyParser.begin_training {#begin_training tag="method"} +## DependencyParser.score {#score tag="method" new="3"} -Initialize the pipe for training, using data examples if available. Return an -[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. +Score a batch of examples. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) -> nlp.pipeline.append(parser) -> optimizer = parser.begin_training(pipeline=nlp.pipeline) +> scores = parser.score(examples) > ``` -| Name | Type | Description | -| -------------- | ----------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -| `get_examples` | `Iterable[Example]` | Optional gold-standard annotations in the form of [`Example`](/api/example) objects. | -| `pipeline` | `List[(str, callable)]` | Optional list of pipeline components that this component is part of. | -| `sgd` | `Optimizer` | An optional [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. Will be created via [`create_optimizer`](/api/dependencyparser#create_optimizer) if not set. | -| **RETURNS** | `Optimizer` | An optimizer. | +| Name | Type | Description | +| ----------- | ------------------- | -------------------------------------------------------------------------------------------------------------------------- | +| `examples` | `Iterable[Example]` | The examples to score. | +| **RETURNS** | `Dict[str, Any]` | The scores, produced by [`Scorer.score_spans`](/api/scorer#score_spans) and [`Scorer.score_deps`](/api/scorer#score_deps). | ## DependencyParser.create_optimizer {#create_optimizer tag="method"} @@ -194,13 +238,13 @@ component. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > optimizer = parser.create_optimizer() > ``` -| Name | Type | Description | -| ----------- | ----------- | --------------------------------------------------------------- | -| **RETURNS** | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | +| Name | Type | Description | +| ----------- | --------------------------------------------------- | -------------- | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## DependencyParser.use_params {#use_params tag="method, contextmanager"} @@ -225,7 +269,7 @@ Add a new label to the pipe. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > parser.add_label("MY_LABEL") > ``` @@ -240,14 +284,14 @@ Serialize the pipe to disk. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > parser.to_disk("/path/to/parser") > ``` -| Name | Type | Description | -| --------- | ------------ | --------------------------------------------------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| Name | Type | Description | +| --------- | --------------- | --------------------------------------------------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | ## DependencyParser.from_disk {#from_disk tag="method"} @@ -256,14 +300,14 @@ Load the pipe from disk. Modifies the object in place and returns it. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > parser.from_disk("/path/to/parser") > ``` | Name | Type | Description | | ----------- | ------------------ | -------------------------------------------------------------------------- | | `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | | **RETURNS** | `DependencyParser` | The modified `DependencyParser` object. | ## DependencyParser.to_bytes {#to_bytes tag="method"} @@ -271,16 +315,16 @@ Load the pipe from disk. Modifies the object in place and returns it. > #### Example > > ```python -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > parser_bytes = parser.to_bytes() > ``` Serialize the pipe to a bytestring. -| Name | Type | Description | -| ----------- | ----- | ------------------------------------------------------------------------- | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | bytes | The serialized form of the `DependencyParser` object. | +| Name | Type | Description | +| ----------- | --------------- | ------------------------------------------------------------------------- | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | bytes | The serialized form of the `DependencyParser` object. | ## DependencyParser.from_bytes {#from_bytes tag="method"} @@ -290,14 +334,14 @@ Load the pipe from a bytestring. Modifies the object in place and returns it. > > ```python > parser_bytes = parser.to_bytes() -> parser = DependencyParser(nlp.vocab) +> parser = nlp.add_pipe("parser") > parser.from_bytes(parser_bytes) > ``` | Name | Type | Description | | ------------ | ------------------ | ------------------------------------------------------------------------- | | `bytes_data` | bytes | The data to load from. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | | **RETURNS** | `DependencyParser` | The `DependencyParser` object. | ## DependencyParser.labels {#labels tag="property"} diff --git a/website/docs/api/entitylinker.md b/website/docs/api/entitylinker.md index bbc6cc632..433ff6f72 100644 --- a/website/docs/api/entitylinker.md +++ b/website/docs/api/entitylinker.md @@ -1,23 +1,44 @@ --- title: EntityLinker -teaser: - Functionality to disambiguate a named entity in text to a unique knowledge - base identifier. tag: class source: spacy/pipeline/entity_linker.py new: 2.2 +teaser: 'Pipeline component for named entity linking and disambiguation' +api_base_class: /api/pipe +api_string_name: entity_linker +api_trainable: true --- -This class is a subclass of `Pipe` and follows the same API. The pipeline -component is available in the [processing pipeline](/usage/processing-pipelines) -via the ID `"entity_linker"`. +## Config and implementation {#config} -## Implementation and defaults {#implementation} +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). See the +[model architectures](/api/architectures) documentation for details on the +architectures and their arguments and hyperparameters. -See the [model architectures](/api/architectures) documentation for details on -the architectures and their arguments and hyperparameters. To learn more about -how to customize the config and train custom models, check out the -[training config](/usage/training#config) docs. +> #### Example +> +> ```python +> from spacy.pipeline.entity_linker import DEFAULT_NEL_MODEL +> config = { +> "kb": None, +> "labels_discard": [], +> "incl_prior": True, +> "incl_context": True, +> "model": DEFAULT_NEL_MODEL, +> } +> nlp.add_pipe("entity_linker", config=config) +> ``` + +| Setting | Type | Description | Default | +| ---------------- | ------------------------------------------ | ----------------- | ----------------------------------------------- | +| `kb` | `KnowledgeBase` | | `None` | +| `labels_discard` | `Iterable[str]` | | `[]` | +| `incl_prior` | bool | |  `True` | +| `incl_context` | bool | | `True` | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The model to use. | [EntityLinker](/api/architectures#EntityLinker) | ```python https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/entity_linker.py @@ -34,19 +55,26 @@ https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/entity_linker.py > # Construction via add_pipe with custom model > config = {"model": {"@architectures": "my_el"}} > entity_linker = nlp.add_pipe("entity_linker", config=config) +> +> # Construction from class +> from spacy.pipeline import EntityLinker +> entity_linker = EntityLinker(nlp.vocab, model) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and [`nlp.add_pipe`](/api/language#add_pipe). -| Name | Type | Description | -| ------- | ------- | ------------------------------------------------------------------------------- | -| `vocab` | `Vocab` | The shared vocabulary. | -| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | -| `**cfg` | - | Configuration parameters. | - -| **RETURNS** | `EntityLinker` | The newly constructed object. | +| Name | Type | Description | +| ---------------- | --------------- | ------------------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | The shared vocabulary. | +| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | +| `name` | str | String name of the component instance. Used to add entries to the `losses` during training. | +| _keyword-only_ | | | +| `kb` | `KnowlegeBase` | | +| `labels_discard` | `Iterable[str]` | | +| `incl_prior` | bool | | +| `incl_context` | bool | | ## EntityLinker.\_\_call\_\_ {#call tag="method"} @@ -60,8 +88,8 @@ delegate to the [`predict`](/api/entitylinker#predict) and > #### Example > > ```python -> entity_linker = EntityLinker(nlp.vocab) > doc = nlp("This is a sentence.") +> entity_linker = nlp.add_pipe("entity_linker") > # This usually happens under the hood > processed = entity_linker(doc) > ``` @@ -83,91 +111,17 @@ applied to the `Doc` in order. Both [`__call__`](/api/entitylinker#call) and > #### Example > > ```python -> entity_linker = EntityLinker(nlp.vocab) +> entity_linker = nlp.add_pipe("entity_linker") > for doc in entity_linker.pipe(docs, batch_size=50): > pass > ``` -| Name | Type | Description | -| ------------ | --------------- | ------------------------------------------------------ | -| `stream` | `Iterable[Doc]` | A stream of documents. | -| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | -| **YIELDS** | `Doc` | Processed documents in the order of the original text. | - -## EntityLinker.predict {#predict tag="method"} - -Apply the pipeline's model to a batch of docs, without modifying them. - -> #### Example -> -> ```python -> entity_linker = EntityLinker(nlp.vocab) -> kb_ids = entity_linker.predict([doc1, doc2]) -> ``` - -| Name | Type | Description | -| ----------- | --------------- | ------------------------------------------------------------ | -| `docs` | `Iterable[Doc]` | The documents to predict. | -| **RETURNS** | `Iterable[str]` | The predicted KB identifiers for the entities in the `docs`. | - -## EntityLinker.set_annotations {#set_annotations tag="method"} - -Modify a batch of documents, using pre-computed entity IDs for a list of named -entities. - -> #### Example -> -> ```python -> entity_linker = EntityLinker(nlp.vocab) -> kb_ids = entity_linker.predict([doc1, doc2]) -> entity_linker.set_annotations([doc1, doc2], kb_ids) -> ``` - -| Name | Type | Description | -| -------- | --------------- | ------------------------------------------------------------------------------------------------- | -| `docs` | `Iterable[Doc]` | The documents to modify. | -| `kb_ids` | `Iterable[str]` | The knowledge base identifiers for the entities in the docs, predicted by `EntityLinker.predict`. | - -## EntityLinker.update {#update tag="method"} - -Learn from a batch of [`Example`](/api/example) objects, updating both the -pipe's entity linking model and context encoder. Delegates to -[`predict`](/api/entitylinker#predict) and -[`get_loss`](/api/entitylinker#get_loss). - -> #### Example -> -> ```python -> entity_linker = EntityLinker(nlp.vocab, nel_model) -> optimizer = nlp.begin_training() -> losses = entity_linker.update(examples, sgd=optimizer) -> ``` - -| Name | Type | Description | -| ----------------- | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------ | -| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | -| _keyword-only_ | | | -| `drop` | float | The dropout rate. | -| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/entitylinker#set_annotations). | -| `sgd` | `Optimizer` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | -| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | -| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | - -## EntityLinker.set_kb {#set_kb tag="method"} - -Define the knowledge base (KB) used for disambiguating named entities to KB -identifiers. - -> #### Example -> -> ```python -> entity_linker = EntityLinker(nlp.vocab) -> entity_linker.set_kb(kb) -> ``` - -| Name | Type | Description | -| ---- | --------------- | ------------------------------- | -| `kb` | `KnowledgeBase` | The [`KnowledgeBase`](/api/kb). | +| Name | Type | Description | +| -------------- | --------------- | ------------------------------------------------------ | +| `stream` | `Iterable[Doc]` | A stream of documents. | +| _keyword-only_ | | | +| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | +| **YIELDS** | `Doc` | Processed documents in the order of the original text. | ## EntityLinker.begin_training {#begin_training tag="method"} @@ -184,12 +138,89 @@ method, a knowledge base should have been defined with > optimizer = entity_linker.begin_training(pipeline=nlp.pipeline) > ``` -| Name | Type | Description | -| -------------- | ----------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------- | -| `get_examples` | `Iterable[Example]` | Optional gold-standard annotations in the form of [`Example`](/api/example) objects. | -| `pipeline` | `List[(str, callable)]` | Optional list of pipeline components that this component is part of. | -| `sgd` | `Optimizer` | An optional [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. Will be created via [`create_optimizer`](/api/entitylinker#create_optimizer) if not set. | -| **RETURNS** | `Optimizer` | An optimizer. | | +| Name | Type | Description | +| -------------- | --------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------- | +| `get_examples` | `Callable[[], Iterable[Example]]` | Optional function that returns gold-standard annotations in the form of [`Example`](/api/example) objects. | +| _keyword-only_ | | | +| `pipeline` | `List[Tuple[str, Callable]]` | Optional list of pipeline components that this component is part of. | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | An optional optimizer. Will be created via [`create_optimizer`](/api/dependencyparser#create_optimizer) if not set. | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | + +## EntityLinker.predict {#predict tag="method"} + +Apply the pipeline's model to a batch of docs, without modifying them. Returns +the KB IDs for each entity in each doc, including `NIL` if there is no +prediction. + +> #### Example +> +> ```python +> entity_linker = nlp.add_pipe("entity_linker") +> kb_ids = entity_linker.predict([doc1, doc2]) +> ``` + +| Name | Type | Description | +| ----------- | --------------- | ------------------------------------------------------------ | +| `docs` | `Iterable[Doc]` | The documents to predict. | +| **RETURNS** | `List[str]` | The predicted KB identifiers for the entities in the `docs`. | + +## EntityLinker.set_annotations {#set_annotations tag="method"} + +Modify a batch of documents, using pre-computed entity IDs for a list of named +entities. + +> #### Example +> +> ```python +> entity_linker = nlp.add_pipe("entity_linker") +> kb_ids = entity_linker.predict([doc1, doc2]) +> entity_linker.set_annotations([doc1, doc2], kb_ids) +> ``` + +| Name | Type | Description | +| -------- | --------------- | ------------------------------------------------------------------------------------------------- | +| `docs` | `Iterable[Doc]` | The documents to modify. | +| `kb_ids` | `List[str]` | The knowledge base identifiers for the entities in the docs, predicted by `EntityLinker.predict`. | + +## EntityLinker.update {#update tag="method"} + +Learn from a batch of [`Example`](/api/example) objects, updating both the +pipe's entity linking model and context encoder. Delegates to +[`predict`](/api/entitylinker#predict). + +> #### Example +> +> ```python +> entity_linker = nlp.add_pipe("entity_linker") +> optimizer = nlp.begin_training() +> losses = entity_linker.update(examples, sgd=optimizer) +> ``` + +| Name | Type | Description | +| ----------------- | --------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------- | +| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | +| _keyword-only_ | | | +| `drop` | float | The dropout rate. | +| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/textcategorizer#set_annotations). | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. Updated using the component name as the key. | +| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | + +## EntityLinker.set_kb {#set_kb tag="method"} + +Define the knowledge base (KB) used for disambiguating named entities to KB +identifiers. + +> #### Example +> +> ```python +> entity_linker = nlp.add_pipe("entity_linker") +> entity_linker.set_kb(kb) +> ``` + +| Name | Type | Description | +| ---- | --------------- | ------------------------------- | +| `kb` | `KnowledgeBase` | The [`KnowledgeBase`](/api/kb). | ## EntityLinker.create_optimizer {#create_optimizer tag="method"} @@ -198,13 +229,13 @@ Create an optimizer for the pipeline component. > #### Example > > ```python -> entity_linker = EntityLinker(nlp.vocab) +> entity_linker = nlp.add_pipe("entity_linker") > optimizer = entity_linker.create_optimizer() > ``` -| Name | Type | Description | -| ----------- | ----------- | --------------------------------------------------------------- | -| **RETURNS** | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | +| Name | Type | Description | +| ----------- | --------------------------------------------------- | -------------- | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## EntityLinker.use_params {#use_params tag="method, contextmanager"} @@ -213,7 +244,7 @@ Modify the pipe's EL model, to use the given parameter values. > #### Example > > ```python -> entity_linker = EntityLinker(nlp.vocab) +> entity_linker = nlp.add_pipe("entity_linker") > with entity_linker.use_params(optimizer.averages): > entity_linker.to_disk("/best_model") > ``` @@ -229,14 +260,14 @@ Serialize the pipe to disk. > #### Example > > ```python -> entity_linker = EntityLinker(nlp.vocab) +> entity_linker = nlp.add_pipe("entity_linker") > entity_linker.to_disk("/path/to/entity_linker") > ``` -| Name | Type | Description | -| --------- | ------------ | --------------------------------------------------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| Name | Type | Description | +| --------- | --------------- | --------------------------------------------------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | ## EntityLinker.from_disk {#from_disk tag="method"} @@ -245,15 +276,15 @@ Load the pipe from disk. Modifies the object in place and returns it. > #### Example > > ```python -> entity_linker = EntityLinker(nlp.vocab) +> entity_linker = nlp.add_pipe("entity_linker") > entity_linker.from_disk("/path/to/entity_linker") > ``` -| Name | Type | Description | -| ----------- | -------------- | -------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | `EntityLinker` | The modified `EntityLinker` object. | +| Name | Type | Description | +| ----------- | --------------- | -------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | `EntityLinker` | The modified `EntityLinker` object. | ## Serialization fields {#serialization-fields} diff --git a/website/docs/api/entityrecognizer.md b/website/docs/api/entityrecognizer.md index 6b5b1f93c..4651af03c 100644 --- a/website/docs/api/entityrecognizer.md +++ b/website/docs/api/entityrecognizer.md @@ -2,18 +2,37 @@ title: EntityRecognizer tag: class source: spacy/pipeline/ner.pyx +teaser: 'Pipeline component for named entity recognition' +api_base_class: /api/pipe +api_string_name: ner +api_trainable: true --- -This class is a subclass of `Pipe` and follows the same API. The pipeline -component is available in the [processing pipeline](/usage/processing-pipelines) -via the ID `"ner"`. +## Config and implementation {#config} -## Implementation and defaults {#implementation} +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). See the +[model architectures](/api/architectures) documentation for details on the +architectures and their arguments and hyperparameters. -See the [model architectures](/api/architectures) documentation for details on -the architectures and their arguments and hyperparameters. To learn more about -how to customize the config and train custom models, check out the -[training config](/usage/training#config) docs. +> #### Example +> +> ```python +> from spacy.pipeline.ner import DEFAULT_NER_MODEL +> config = { +> "moves": None, +> # TODO: rest +> "model": DEFAULT_NER_MODEL, +> } +> nlp.add_pipe("ner", config=config) +> ``` + +| Setting | Type | Description | Default | +| ------- | ------------------------------------------ | ----------------- | ----------------------------------------------------------------- | +| `moves` | list | | `None` | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The model to use. | [TransitionBasedParser](/api/architectures#TransitionBasedParser) | ```python https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/ner.pyx @@ -30,18 +49,27 @@ https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/ner.pyx > # Construction via add_pipe with custom model > config = {"model": {"@architectures": "my_ner"}} > parser = nlp.add_pipe("ner", config=config) +> +> # Construction from class +> from spacy.pipeline import EntityRecognizer +> ner = EntityRecognizer(nlp.vocab, model) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and [`nlp.add_pipe`](/api/language#add_pipe). -| Name | Type | Description | -| ----------- | ------------------ | ------------------------------------------------------------------------------- | -| `vocab` | `Vocab` | The shared vocabulary. | -| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | -| `**cfg` | - | Configuration parameters. | -| **RETURNS** | `EntityRecognizer` | The newly constructed object. | +| Name | Type | Description | +| ----------------------------- | ------------------------------------------ | ------------------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | The shared vocabulary. | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | +| `name` | str | String name of the component instance. Used to add entries to the `losses` during training. | +| `moves` | list | | +| _keyword-only_ | | | +| `update_with_oracle_cut_size` | int | | +| `multitasks` | `Iterable` | | +| `learn_tokens` | bool | | +| `min_action_freq` | int | | ## EntityRecognizer.\_\_call\_\_ {#call tag="method"} @@ -56,8 +84,8 @@ and all pipeline components are applied to the `Doc` in order. Both > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) > doc = nlp("This is a sentence.") +> ner = nlp.add_pipe("ner") > # This usually happens under the hood > processed = ner(doc) > ``` @@ -79,16 +107,37 @@ applied to the `Doc` in order. Both [`__call__`](/api/entityrecognizer#call) and > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > for doc in ner.pipe(docs, batch_size=50): > pass > ``` -| Name | Type | Description | -| ------------ | --------------- | ------------------------------------------------------ | -| `stream` | `Iterable[Doc]` | A stream of documents. | -| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | -| **YIELDS** | `Doc` | Processed documents in the order of the original text. | +| Name | Type | Description | +| -------------- | --------------- | ------------------------------------------------------ | +| `docs` | `Iterable[Doc]` | A stream of documents. | +| _keyword-only_ | | | +| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | +| **YIELDS** | `Doc` | Processed documents in the order of the original text. | + +## EntityRecognizer.begin_training {#begin_training tag="method"} + +Initialize the pipe for training, using data examples if available. Return an +[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. + +> #### Example +> +> ```python +> ner = nlp.add_pipe("ner") +> optimizer = ner.begin_training(pipeline=nlp.pipeline) +> ``` + +| Name | Type | Description | +| -------------- | --------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------- | +| `get_examples` | `Callable[[], Iterable[Example]]` | Optional function that returns gold-standard annotations in the form of [`Example`](/api/example) objects. | +| _keyword-only_ | | | +| `pipeline` | `List[Tuple[str, Callable]]` | Optional list of pipeline components that this component is part of. | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | An optional optimizer. Will be created via [`create_optimizer`](/api/entityrecognizer#create_optimizer) if not set. | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## EntityRecognizer.predict {#predict tag="method"} @@ -97,7 +146,7 @@ Apply the pipeline's model to a batch of docs, without modifying them. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > scores = ner.predict([doc1, doc2]) > ``` @@ -113,7 +162,7 @@ Modify a batch of documents, using pre-computed scores. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > scores = ner.predict([doc1, doc2]) > ner.set_annotations([doc1, doc2], scores) > ``` @@ -132,20 +181,20 @@ model. Delegates to [`predict`](/api/entityrecognizer#predict) and > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab, ner_model) +> ner = nlp.add_pipe("ner") > optimizer = nlp.begin_training() > losses = ner.update(examples, sgd=optimizer) > ``` -| Name | Type | Description | -| ----------------- | ------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------- | -| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | -| _keyword-only_ | | | -| `drop` | float | The dropout rate. | -| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/entityrecognizer#set_annotations). | -| `sgd` | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | -| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | -| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | +| Name | Type | Description | +| ----------------- | --------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------- | +| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | +| _keyword-only_ | | | +| `drop` | float | The dropout rate. | +| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/entityrecognizer#set_annotations). | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. Updated using the component name as the key. | +| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | ## EntityRecognizer.get_loss {#get_loss tag="method"} @@ -155,36 +204,31 @@ predicted scores. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > scores = ner.predict([eg.predicted for eg in examples]) > loss, d_loss = ner.get_loss(examples, scores) > ``` -| Name | Type | Description | -| ----------- | ------------------- | --------------------------------------------------- | -| `examples` | `Iterable[Example]` | The batch of examples. | -| `scores` | `List[StateClass]` | Scores representing the model's predictions. | -| **RETURNS** | tuple | The loss and the gradient, i.e. `(loss, gradient)`. | +| Name | Type | Description | +| ----------- | --------------------- | --------------------------------------------------- | +| `examples` | `Iterable[Example]` | The batch of examples. | +| `scores` | `List[StateClass]` | Scores representing the model's predictions. | +| **RETURNS** | `Tuple[float, float]` | The loss and the gradient, i.e. `(loss, gradient)`. | -## EntityRecognizer.begin_training {#begin_training tag="method"} +## EntityRecognizer.score {#score tag="method" new="3"} -Initialize the pipe for training, using data examples if available. Return an -[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. +Score a batch of examples. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) -> nlp.pipeline.append(ner) -> optimizer = ner.begin_training(pipeline=nlp.pipeline) +> scores = ner.score(examples) > ``` -| Name | Type | Description | -| -------------- | ----------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -| `get_examples` | `Iterable[Example]` | Optional gold-standard annotations in the form of [`Example`](/api/example) objects. | -| `pipeline` | `List[(str, callable)]` | Optional list of pipeline components that this component is part of. | -| `sgd` | `Optimizer` | An optional [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. Will be created via [`create_optimizer`](/api/entityrecognizer#create_optimizer) if not set. | -| **RETURNS** | `Optimizer` | An optimizer. | +| Name | Type | Description | +| ----------- | ------------------- | ------------------------------------------------------------------------ | +| `examples` | `Iterable[Example]` | The examples to score. | +| **RETURNS** | `Dict[str, Any]` | The scores, produced by [`Scorer.score_spans`](/api/scorer#score_spans). | ## EntityRecognizer.create_optimizer {#create_optimizer tag="method"} @@ -193,13 +237,13 @@ Create an optimizer for the pipeline component. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > optimizer = ner.create_optimizer() > ``` -| Name | Type | Description | -| ----------- | ----------- | --------------------------------------------------------------- | -| **RETURNS** | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | +| Name | Type | Description | +| ----------- | --------------------------------------------------- | -------------- | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## EntityRecognizer.use_params {#use_params tag="method, contextmanager"} @@ -224,7 +268,7 @@ Add a new label to the pipe. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > ner.add_label("MY_LABEL") > ``` @@ -239,14 +283,14 @@ Serialize the pipe to disk. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > ner.to_disk("/path/to/ner") > ``` -| Name | Type | Description | -| --------- | ------------ | --------------------------------------------------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| Name | Type | Description | +| --------- | --------------- | --------------------------------------------------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | ## EntityRecognizer.from_disk {#from_disk tag="method"} @@ -255,14 +299,14 @@ Load the pipe from disk. Modifies the object in place and returns it. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > ner.from_disk("/path/to/ner") > ``` | Name | Type | Description | | ----------- | ------------------ | -------------------------------------------------------------------------- | | `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | | **RETURNS** | `EntityRecognizer` | The modified `EntityRecognizer` object. | ## EntityRecognizer.to_bytes {#to_bytes tag="method"} @@ -270,16 +314,16 @@ Load the pipe from disk. Modifies the object in place and returns it. > #### Example > > ```python -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > ner_bytes = ner.to_bytes() > ``` Serialize the pipe to a bytestring. -| Name | Type | Description | -| ----------- | ----- | ------------------------------------------------------------------------- | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | bytes | The serialized form of the `EntityRecognizer` object. | +| Name | Type | Description | +| ----------- | --------------- | ------------------------------------------------------------------------- | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | bytes | The serialized form of the `EntityRecognizer` object. | ## EntityRecognizer.from_bytes {#from_bytes tag="method"} @@ -289,14 +333,14 @@ Load the pipe from a bytestring. Modifies the object in place and returns it. > > ```python > ner_bytes = ner.to_bytes() -> ner = EntityRecognizer(nlp.vocab) +> ner = nlp.add_pipe("ner") > ner.from_bytes(ner_bytes) > ``` | Name | Type | Description | | ------------ | ------------------ | ------------------------------------------------------------------------- | | `bytes_data` | bytes | The data to load from. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | | **RETURNS** | `EntityRecognizer` | The `EntityRecognizer` object. | ## EntityRecognizer.labels {#labels tag="property"} diff --git a/website/docs/api/entityruler.md b/website/docs/api/entityruler.md index 12a6ab748..0b4369ec4 100644 --- a/website/docs/api/entityruler.md +++ b/website/docs/api/entityruler.md @@ -3,17 +3,48 @@ title: EntityRuler tag: class source: spacy/pipeline/entityruler.py new: 2.1 +teaser: 'Pipeline component for rule-based named entity recognition' +api_string_name: entity_ruler +api_trainable: false --- -The EntityRuler lets you add spans to the [`Doc.ents`](/api/doc#ents) using +The entity ruler lets you add spans to the [`Doc.ents`](/api/doc#ents) using token-based rules or exact phrase matches. It can be combined with the statistical [`EntityRecognizer`](/api/entityrecognizer) to boost accuracy, or -used on its own to implement a purely rule-based entity recognition system. The -pipeline component is available in the -[processing pipeline](/usage/processing-pipelines) via the ID `"entity_ruler"`. -For usage examples, see the docs on +used on its own to implement a purely rule-based entity recognition system. For +usage examples, see the docs on [rule-based entity recognition](/usage/rule-based-matching#entityruler). +## Config and implementation {#config} + +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). + +> #### Example +> +> ```python +> config = { +> "phrase_matcher_attr": None, +> "validation": True, +> "overwrite_ents": False, +> "ent_id_sep": "||", +> } +> nlp.add_pipe("entity_ruler", config=config) +> ``` + +| Setting | Type | Description | Default | +| --------------------- | ---- | ------------------------------------------------------------------------------------------------------------------------------------------- | ------- | +| `phrase_matcher_attr` | str | Optional attribute name match on for the internal [`PhraseMatcher`](/api/phrasematcher), e.g. `LOWER` to match on the lowercase token text. | `None` | +| `validation` | bool | Whether patterns should be validated, passed to Matcher and PhraseMatcher as `validate`. | `False` | +| `overwrite_ents` | bool | If existing entities are present, e.g. entities added by the model, overwrite them by matches if necessary. | `False` | +| `ent_id_sep` | str | Separator used internally for entity IDs. | `"||"` | + +```python +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/entityruler.py +``` + ## EntityRuler.\_\_init\_\_ {#init tag="method"} Initialize the entity ruler. If patterns are supplied here, they need to be a @@ -32,15 +63,16 @@ be a token pattern (list) or a phrase pattern (string). For example: > ruler = EntityRuler(nlp, overwrite_ents=True) > ``` -| Name | Type | Description | -| --------------------- | ------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------- | -| `nlp` | `Language` | The shared nlp object to pass the vocab to the matchers and process phrase patterns. | -| `patterns` | iterable | Optional patterns to load in. | -| `phrase_matcher_attr` | int / str | Optional attr to pass to the internal [`PhraseMatcher`](/api/phrasematcher). defaults to `None` | -| `validate` | bool | Whether patterns should be validated, passed to Matcher and PhraseMatcher as `validate`. Defaults to `False`. | -| `overwrite_ents` | bool | If existing entities are present, e.g. entities added by the model, overwrite them by matches if necessary. Defaults to `False`. | -| `**cfg` | - | Other config parameters. If pipeline component is loaded as part of a model pipeline, this will include all keyword arguments passed to `spacy.load`. | -| **RETURNS** | `EntityRuler` | The newly constructed object. | +| Name | Type | Description | +| --------------------------------- | ---------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `nlp` | `Language` | The shared nlp object to pass the vocab to the matchers and process phrase patterns. | +| `name` 3 | str | Instance name of the current pipeline component. Typically passed in automatically from the factory when the component is added. Used to disable the current entity ruler while creating phrase patterns with the nlp object. | +| _keyword-only_ | | | +| `phrase_matcher_attr` | int / str | Optional attribute name match on for the internal [`PhraseMatcher`](/api/phrasematcher), e.g. `LOWER` to match on the lowercase token text. Defaults to `None`. | +| `validate` | bool | Whether patterns should be validated, passed to Matcher and PhraseMatcher as `validate`. Defaults to `False`. | +| `overwrite_ents` | bool | If existing entities are present, e.g. entities added by the model, overwrite them by matches if necessary. Defaults to `False`. | +| `ent_id_sep` | str | Separator used internally for entity IDs. Defaults to `"||"`. | +| `patterns` | iterable | Optional patterns to load in on initialization. | ## EntityRuler.\_\len\_\_ {#len tag="method"} @@ -49,7 +81,7 @@ The number of all patterns added to the entity ruler. > #### Example > > ```python -> ruler = EntityRuler(nlp) +> ruler = nlp.add_pipe("entity_ruler") > assert len(ruler) == 0 > ruler.add_patterns([{"label": "ORG", "pattern": "Apple"}]) > assert len(ruler) == 1 @@ -66,7 +98,7 @@ Whether a label is present in the patterns. > #### Example > > ```python -> ruler = EntityRuler(nlp) +> ruler = nlp.add_pipe("entity_ruler") > ruler.add_patterns([{"label": "ORG", "pattern": "Apple"}]) > assert "ORG" in ruler > assert not "PERSON" in ruler @@ -116,7 +148,7 @@ of dicts) or a phrase pattern (string). For more details, see the usage guide on > {"label": "ORG", "pattern": "Apple"}, > {"label": "GPE", "pattern": [{"lower": "san"}, {"lower": "francisco"}]} > ] -> ruler = EntityRuler(nlp) +> ruler = nlp.add_pipe("entity_ruler") > ruler.add_patterns(patterns) > ``` @@ -134,7 +166,7 @@ only the patterns are saved as JSONL. If a directory name is provided, a > #### Example > > ```python -> ruler = EntityRuler(nlp) +> ruler = nlp.add_pipe("entity_ruler") > ruler.to_disk("/path/to/patterns.jsonl") # saves patterns only > ruler.to_disk("/path/to/entity_ruler") # saves patterns and config > ``` @@ -153,7 +185,7 @@ configuration. > #### Example > > ```python -> ruler = EntityRuler(nlp) +> ruler = nlp.add_pipe("entity_ruler") > ruler.from_disk("/path/to/patterns.jsonl") # loads patterns only > ruler.from_disk("/path/to/entity_ruler") # loads patterns and config > ``` @@ -170,7 +202,7 @@ Serialize the entity ruler patterns to a bytestring. > #### Example > > ```python -> ruler = EntityRuler(nlp) +> ruler = nlp.add_pipe("entity_ruler") > ruler_bytes = ruler.to_bytes() > ``` @@ -186,14 +218,14 @@ Load the pipe from a bytestring. Modifies the object in place and returns it. > > ```python > ruler_bytes = ruler.to_bytes() -> ruler = EntityRuler(nlp) +> ruler = nlp.add_pipe("enity_ruler") > ruler.from_bytes(ruler_bytes) > ``` -| Name | Type | Description | -| ---------------- | ------------- | ---------------------------------- | -| `patterns_bytes` | bytes | The bytestring to load. | -| **RETURNS** | `EntityRuler` | The modified `EntityRuler` object. | +| Name | Type | Description | +| ------------ | ------------- | ---------------------------------- | +| `bytes_data` | bytes | The bytestring to load. | +| **RETURNS** | `EntityRuler` | The modified `EntityRuler` object. | ## EntityRuler.labels {#labels tag="property"} diff --git a/website/docs/api/language.md b/website/docs/api/language.md index c41b349e5..1df7bcb48 100644 --- a/website/docs/api/language.md +++ b/website/docs/api/language.md @@ -495,6 +495,51 @@ As of spaCy v3.0, the `disable_pipes` method has been renamed to `select_pipes`: | `enable` | str / list | Names(s) of pipeline components that will not be disabled. | | **RETURNS** | `DisabledPipes` | The disabled pipes that can be restored by calling the object's `.restore()` method. | +## Language.get_factory_meta {#get_factory_meta tag="classmethod" new="3"} + +Get the factory meta information for a given pipeline component name. Expects +the name of the component **factory**. The factory meta is an instance of the +[`FactoryMeta`](/api/language#factorymeta) dataclass and contains the +information about the component and its default provided by the +[`@Language.component`](/api/language#component) or +[`@Language.factory`](/api/language#factory) decorator. + +> #### Example +> +> ```python +> factory_meta = Language.get_factory_meta("ner") +> assert factory_meta.factory == "ner" +> print(factory_meta.default_config) +> ``` + +| Name | Type | Description | +| ----------- | ----------------------------- | ------------------ | +| `name` | str | The factory name. | +| **RETURNS** | [`FactoryMeta`](#factorymeta) |  The factory meta. | + +## Language.get_pipe_meta {#get_pipe_meta tag="method" new="3"} + +Get the factory meta information for a given pipeline component name. Expects +the name of the component **instance** in the pipeline. The factory meta is an +instance of the [`FactoryMeta`](/api/language#factorymeta) dataclass and +contains the information about the component and its default provided by the +[`@Language.component`](/api/language#component) or +[`@Language.factory`](/api/language#factory) decorator. + +> #### Example +> +> ```python +> nlp.add_pipe("ner", name="entity_recognizer") +> factory_meta = nlp.get_pipe_meta("entity_recognizer") +> assert factory_meta.factory == "ner" +> print(factory_meta.default_config) +> ``` + +| Name | Type | Description | +| ----------- | ----------------------------- | ---------------------------- | +| `name` | str | The pipeline component name. | +| **RETURNS** | [`FactoryMeta`](#factorymeta) |  The factory meta. | + ## Language.meta {#meta tag="property"} Custom meta data for the Language class. If a model is loaded, contains meta @@ -622,6 +667,7 @@ available to the loaded object. | `pipe_names` 2 | `List[str]` | List of pipeline component names, in order. | | `pipe_labels` 2.2 | `Dict[str, List[str]]` | List of labels set by the pipeline components, if available, keyed by component name. | | `pipe_factories` 2.2 | `Dict[str, str]` | Dictionary of pipeline component names, mapped to their factory names. | +| `factories` | `Dict[str, Callable]` | All available factory functions, keyed by name. | | `factory_names` 3 | `List[str]` | List of all available factory names. | | `path` 2 | `Path` | Path to the model data directory, if a model is loaded. Otherwise `None`. | @@ -712,3 +758,19 @@ serialization by passing in the string names via the `exclude` argument. | `tokenizer` | Tokenization rules and exceptions. | | `meta` | The meta data, available as `Language.meta`. | | ... | String names of pipeline components, e.g. `"ner"`. | + +## FactoryMeta {#factorymeta new="3" tag="dataclass"} + +The `FactoryMeta` contains the information about the component and its default +provided by the [`@Language.component`](/api/language#component) or +[`@Language.factory`](/api/language#factory) decorator. It's created whenever a +component is added to the pipeline and stored on the `Language` class for each +component instance and factory instance. + +| Name | Type | Description | +| ---------------- | ---------------- | ---------------------------------------------------------------------------------------------------------------------------------------------- | +| `factory` | str | The name of the registered component factory. | +| `default_config` | `Dict[str, Any]` | The default config, describing the default values of the factory arguments. | +| `assigns` | `Iterable[str]` | `Doc` or `Token` attributes assigned by this component, e.g. `["token.ent_id"]`. Used for pipeline analysis. | +| `requires` | `Iterable[str]` | `Doc` or `Token` attributes required by this component, e.g. `["token.ent_id"]`. Used for pipeline analysis.   | +| `retokenizes` | bool | Whether the component changes tokenization. Used for pipeline analysis.   | diff --git a/website/docs/api/lemmatizer.md b/website/docs/api/lemmatizer.md index fa376d246..237bfa468 100644 --- a/website/docs/api/lemmatizer.md +++ b/website/docs/api/lemmatizer.md @@ -5,6 +5,8 @@ tag: class source: spacy/lemmatizer.py --- + + The `Lemmatizer` supports simple part-of-speech-sensitive suffix rules and lookup tables. diff --git a/website/docs/api/matcher.md b/website/docs/api/matcher.md index 636354496..c59a58c81 100644 --- a/website/docs/api/matcher.md +++ b/website/docs/api/matcher.md @@ -142,11 +142,12 @@ patterns = [[{"TEXT": "Google"}, {"TEXT": "Now"}], [{"TEXT": "GoogleNow"}]] -| Name | Type | Description | -| ---------- | ------------------ | --------------------------------------------------------------------------------------------- | -| `match_id` | str | An ID for the thing you're matching. | -| `patterns` | list | Match pattern. A pattern consists of a list of dicts, where each dict describes a token. | -| `on_match` | callable or `None` | Callback function to act on matches. Takes the arguments `matcher`, `doc`, `i` and `matches`. | +| Name | Type | Description | +| -------------- | ------------------ | --------------------------------------------------------------------------------------------- | +| `match_id` | str | An ID for the thing you're matching. | +| `patterns` | list | Match pattern. A pattern consists of a list of dicts, where each dict describes a token. | +| _keyword-only_ | | | +| `on_match` | callable or `None` | Callback function to act on matches. Takes the arguments `matcher`, `doc`, `i` and `matches`. | ## Matcher.remove {#remove tag="method" new="2"} diff --git a/website/docs/api/morphologizer.md b/website/docs/api/morphologizer.md index 3585c76f9..64843a3e0 100644 --- a/website/docs/api/morphologizer.md +++ b/website/docs/api/morphologizer.md @@ -3,22 +3,38 @@ title: Morphologizer tag: class source: spacy/pipeline/morphologizer.pyx new: 3 +teaser: 'Pipeline component for predicting morphological features' +api_base_class: /api/tagger +api_string_name: morphologizer +api_trainable: true --- A trainable pipeline component to predict morphological features and coarse-grained POS tags following the Universal Dependencies [UPOS](https://universaldependencies.org/u/pos/index.html) and [FEATS](https://universaldependencies.org/format.html#morphological-annotation) -annotation guidelines. This class is a subclass of `Pipe` and follows the same -API. The pipeline component is available in the -[processing pipeline](/usage/processing-pipelines) via the ID `"morphologizer"`. +annotation guidelines. -## Implementation and defaults {#implementation} +## Config and implementation {#config} -See the [model architectures](/api/architectures) documentation for details on -the architectures and their arguments and hyperparameters. To learn more about -how to customize the config and train custom models, check out the -[training config](/usage/training#config) docs. +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). See the +[model architectures](/api/architectures) documentation for details on the +architectures and their arguments and hyperparameters. + +> #### Example +> +> ```python +> from spacy.pipeline.morphologizer import DEFAULT_MORPH_MODEL +> config = {"model": DEFAULT_MORPH_MODEL} +> nlp.add_pipe("morphologizer", config=config) +> ``` + +| Setting | Type | Description | Default | +| ------- | ------------------------------------------ | ----------------- | ----------------------------------- | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The model to use. | [Tagger](/api/architectures#Tagger) | ```python https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/morphologizer.pyx @@ -31,20 +47,30 @@ Initialize the morphologizer. > #### Example > > ```python -> # Construction via add_pipe +> # Construction via add_pipe with default model > morphologizer = nlp.add_pipe("morphologizer") +> +> # Construction via create_pipe with custom model +> config = {"model": {"@architectures": "my_morphologizer"}} +> morphologizer = nlp.add_pipe("morphologizer", config=config) +> +> # Construction from class +> from spacy.pipeline import Morphologizer +> morphologizer = Morphologizer(nlp.vocab, model) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and [`nlp.add_pipe`](/api/language#add_pipe). -| Name | Type | Description | -| ----------- | --------------- | ------------------------------------------------------------------------------- | -| `vocab` | `Vocab` | The shared vocabulary. | -| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | -| `**cfg` | - | Configuration parameters. | -| **RETURNS** | `Morphologizer` | The newly constructed object. | +| Name | Type | Description | +| -------------- | ------- | ------------------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | The shared vocabulary. | +| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | +| `name` | str | String name of the component instance. Used to add entries to the `losses` during training. | +| _keyword-only_ | | | +| `labels_morph` | dict | | +| `labels_pos` | dict | | ## Morphologizer.\_\_call\_\_ {#call tag="method"} @@ -58,8 +84,8 @@ delegate to the [`predict`](/api/morphologizer#predict) and > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) > doc = nlp("This is a sentence.") +> morphologizer = nlp.add_pipe("morphologizer") > # This usually happens under the hood > processed = morphologizer(doc) > ``` @@ -81,16 +107,38 @@ applied to the `Doc` in order. Both [`__call__`](/api/morphologizer#call) and > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > for doc in morphologizer.pipe(docs, batch_size=50): > pass > ``` -| Name | Type | Description | -| ------------ | --------------- | ------------------------------------------------------ | -| `stream` | `Iterable[Doc]` | A stream of documents. | -| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | -| **YIELDS** | `Doc` | Processed documents in the order of the original text. | +| Name | Type | Description | +| -------------- | --------------- | ------------------------------------------------------ | +| `stream` | `Iterable[Doc]` | A stream of documents. | +| _keyword-only_ | | | +| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | +| **YIELDS** | `Doc` | Processed documents in the order of the original text. | + +## Morphologizer.begin_training {#begin_training tag="method"} + +Initialize the pipe for training, using data examples if available. Return an +[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. + +> #### Example +> +> ```python +> morphologizer = nlp.add_pipe("morphologizer") +> nlp.pipeline.append(morphologizer) +> optimizer = morphologizer.begin_training(pipeline=nlp.pipeline) +> ``` + +| Name | Type | Description | +| -------------- | --------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------- | +| `get_examples` | `Callable[[], Iterable[Example]]` | Optional function that returns gold-standard annotations in the form of [`Example`](/api/example) objects. | +| _keyword-only_ | | | +| `pipeline` | `List[Tuple[str, Callable]]` | Optional list of pipeline components that this component is part of. | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | An optional optimizer. Will be created via [`create_optimizer`](/api/sentencerecognizer#create_optimizer) if not set. | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## Morphologizer.predict {#predict tag="method"} @@ -99,7 +147,7 @@ Apply the pipeline's model to a batch of docs, without modifying them. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > scores = morphologizer.predict([doc1, doc2]) > ``` @@ -115,7 +163,7 @@ Modify a batch of documents, using pre-computed scores. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > scores = morphologizer.predict([doc1, doc2]) > morphologizer.set_annotations([doc1, doc2], scores) > ``` @@ -134,20 +182,20 @@ pipe's model. Delegates to [`predict`](/api/morphologizer#predict) and > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab, morphologizer_model) +> morphologizer = nlp.add_pipe("morphologizer") > optimizer = nlp.begin_training() > losses = morphologizer.update(examples, sgd=optimizer) > ``` -| Name | Type | Description | -| ----------------- | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------- | -| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | -| _keyword-only_ | | | -| `drop` | float | The dropout rate. | -| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/morphologizer#set_annotations). | -| `sgd` | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | -| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | -| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | +| Name | Type | Description | +| ----------------- | --------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ | +| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | +| _keyword-only_ | | | +| `drop` | float | The dropout rate. | +| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/sentencerecognizer#set_annotations). | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | +| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | ## Morphologizer.get_loss {#get_loss tag="method"} @@ -157,36 +205,16 @@ predicted scores. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > scores = morphologizer.predict([eg.predicted for eg in examples]) > loss, d_loss = morphologizer.get_loss(examples, scores) > ``` -| Name | Type | Description | -| ----------- | ------------------- | --------------------------------------------------- | -| `examples` | `Iterable[Example]` | The batch of examples. | -| `scores` | - | Scores representing the model's predictions. | -| **RETURNS** | tuple | The loss and the gradient, i.e. `(loss, gradient)`. | - -## Morphologizer.begin_training {#begin_training tag="method"} - -Initialize the pipe for training, using data examples if available. Return an -[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. - -> #### Example -> -> ```python -> morphologizer = Morphologizer(nlp.vocab) -> nlp.pipeline.append(morphologizer) -> optimizer = morphologizer.begin_training(pipeline=nlp.pipeline) -> ``` - -| Name | Type | Description | -| -------------- | ----------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------- | -| `get_examples` | `Iterable[Example]` | Optional gold-standard annotations in the form of [`Example`](/api/example) objects. | -| `pipeline` | `List[(str, callable)]` | Optional list of pipeline components that this component is part of. | -| `sgd` | `Optimizer` | An optional [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. Will be created via [`create_optimizer`](/api/morphologizer#create_optimizer) if not set. | -| **RETURNS** | `Optimizer` | An optimizer. | +| Name | Type | Description | +| ----------- | --------------------- | --------------------------------------------------- | +| `examples` | `Iterable[Example]` | The batch of examples. | +| `scores` | - | Scores representing the model's predictions. | +| **RETURNS** | `Tuple[float, float]` | The loss and the gradient, i.e. `(loss, gradient)`. | ## Morphologizer.create_optimizer {#create_optimizer tag="method"} @@ -195,13 +223,13 @@ Create an optimizer for the pipeline component. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > optimizer = morphologizer.create_optimizer() > ``` -| Name | Type | Description | -| ----------- | ----------- | --------------------------------------------------------------- | -| **RETURNS** | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | +| Name | Type | Description | +| ----------- | --------------------------------------------------- | -------------- | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## Morphologizer.use_params {#use_params tag="method, contextmanager"} @@ -210,7 +238,7 @@ Modify the pipe's model, to use the given parameter values. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > with morphologizer.use_params(): > morphologizer.to_disk("/best_model") > ``` @@ -227,7 +255,7 @@ both `pos` and `morph`, the label should include the UPOS as the feature `POS`. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > morphologizer.add_label("Mood=Ind|POS=VERB|Tense=Past|VerbForm=Fin") > ``` @@ -242,14 +270,14 @@ Serialize the pipe to disk. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > morphologizer.to_disk("/path/to/morphologizer") > ``` -| Name | Type | Description | -| --------- | ------------ | --------------------------------------------------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| Name | Type | Description | +| --------- | --------------- | --------------------------------------------------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | ## Morphologizer.from_disk {#from_disk tag="method"} @@ -258,14 +286,14 @@ Load the pipe from disk. Modifies the object in place and returns it. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > morphologizer.from_disk("/path/to/morphologizer") > ``` | Name | Type | Description | | ----------- | --------------- | -------------------------------------------------------------------------- | | `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | | **RETURNS** | `Morphologizer` | The modified `Morphologizer` object. | ## Morphologizer.to_bytes {#to_bytes tag="method"} @@ -273,16 +301,16 @@ Load the pipe from disk. Modifies the object in place and returns it. > #### Example > > ```python -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > morphologizer_bytes = morphologizer.to_bytes() > ``` Serialize the pipe to a bytestring. -| Name | Type | Description | -| ----------- | ----- | ------------------------------------------------------------------------- | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | bytes | The serialized form of the `Morphologizer` object. | +| Name | Type | Description | +| ----------- | --------------- | ------------------------------------------------------------------------- | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | bytes | The serialized form of the `Morphologizer` object. | ## Morphologizer.from_bytes {#from_bytes tag="method"} @@ -292,14 +320,14 @@ Load the pipe from a bytestring. Modifies the object in place and returns it. > > ```python > morphologizer_bytes = morphologizer.to_bytes() -> morphologizer = Morphologizer(nlp.vocab) +> morphologizer = nlp.add_pipe("morphologizer") > morphologizer.from_bytes(morphologizer_bytes) > ``` | Name | Type | Description | | ------------ | --------------- | ------------------------------------------------------------------------- | | `bytes_data` | bytes | The data to load from. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | | **RETURNS** | `Morphologizer` | The `Morphologizer` object. | ## Morphologizer.labels {#labels tag="property"} diff --git a/website/docs/api/phrasematcher.md b/website/docs/api/phrasematcher.md index 9c722297d..991016094 100644 --- a/website/docs/api/phrasematcher.md +++ b/website/docs/api/phrasematcher.md @@ -165,11 +165,12 @@ patterns = [nlp("health care reform"), nlp("healthcare reform")] -| Name | Type | Description | -| ---------- | ------------------ | --------------------------------------------------------------------------------------------- | -| `match_id` | str | An ID for the thing you're matching. | -| `docs` | list | `Doc` objects of the phrases to match. | -| `on_match` | callable or `None` | Callback function to act on matches. Takes the arguments `matcher`, `doc`, `i` and `matches`. | +| Name | Type | Description | +| -------------- | ------------------ | --------------------------------------------------------------------------------------------- | +| `match_id` | str | An ID for the thing you're matching. | +| `docs` | list | `Doc` objects of the phrases to match. | +| _keyword-only_ | | | +| `on_match` | callable or `None` | Callback function to act on matches. Takes the arguments `matcher`, `doc`, `i` and `matches`. | ## PhraseMatcher.remove {#remove tag="method" new="2.2"} diff --git a/website/docs/api/pipe.md b/website/docs/api/pipe.md new file mode 100644 index 000000000..f87cf7d20 --- /dev/null +++ b/website/docs/api/pipe.md @@ -0,0 +1,6 @@ +--- +title: Pipe +tag: class +--- + +TODO: write diff --git a/website/docs/api/sentencerecognizer.md b/website/docs/api/sentencerecognizer.md index 2cea97aaf..5927fbaac 100644 --- a/website/docs/api/sentencerecognizer.md +++ b/website/docs/api/sentencerecognizer.md @@ -3,19 +3,35 @@ title: SentenceRecognizer tag: class source: spacy/pipeline/senter.pyx new: 3 +teaser: 'Pipeline component for sentence segmentation' +api_base_class: /api/tagger +api_string_name: senter +api_trainable: true --- A trainable pipeline component for sentence segmentation. For a simpler, -ruse-based strategy, see the [`Sentencizer`](/api/sentencizer). This class is a -subclass of `Pipe` and follows the same API. The component is also available via -the string name `"senter"`. +ruse-based strategy, see the [`Sentencizer`](/api/sentencizer). -## Implementation and defaults {#implementation} +## Config and implementation {#config} -See the [model architectures](/api/architectures) documentation for details on -the architectures and their arguments and hyperparameters. To learn more about -how to customize the config and train custom models, check out the -[training config](/usage/training#config) docs. +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). See the +[model architectures](/api/architectures) documentation for details on the +architectures and their arguments and hyperparameters. + +> #### Example +> +> ```python +> from spacy.pipeline.senter import DEFAULT_SENTER_MODEL +> config = {"model": DEFAULT_SENTER_MODEL,} +> nlp.add_pipe("senter", config=config) +> ``` + +| Setting | Type | Description | Default | +| ------- | ------------------------------------------ | ----------------- | ----------------------------------- | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The model to use. | [Tagger](/api/architectures#Tagger) | ```python https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/senter.pyx @@ -28,8 +44,322 @@ Initialize the sentence recognizer. > #### Example > > ```python -> # Construction via add_pipe +> # Construction via add_pipe with default model > senter = nlp.add_pipe("senter") +> +> # Construction via create_pipe with custom model +> config = {"model": {"@architectures": "my_senter"}} +> senter = nlp.add_pipe("senter", config=config) +> +> # Construction from class +> from spacy.pipeline import SentenceRecognizer +> senter = SentenceRecognizer(nlp.vocab, model) > ``` - +Create a new pipeline instance. In your application, you would normally use a +shortcut for this and instantiate the component using its string name and +[`nlp.add_pipe`](/api/language#add_pipe). + +| Name | Type | Description | +| ------- | ------- | ------------------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | The shared vocabulary. | +| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | +| `name` | str | String name of the component instance. Used to add entries to the `losses` during training. | + +## SentenceRecognizer.\_\_call\_\_ {#call tag="method"} + +Apply the pipe to one document. The document is modified in place, and returned. +This usually happens under the hood when the `nlp` object is called on a text +and all pipeline components are applied to the `Doc` in order. Both +[`__call__`](/api/sentencerecognizer#call) and +[`pipe`](/api/sentencerecognizer#pipe) delegate to the +[`predict`](/api/sentencerecognizer#predict) and +[`set_annotations`](/api/sentencerecognizer#set_annotations) methods. + +> #### Example +> +> ```python +> doc = nlp("This is a sentence.") +> senter = nlp.add_pipe("senter") +> # This usually happens under the hood +> processed = senter(doc) +> ``` + +| Name | Type | Description | +| ----------- | ----- | ------------------------ | +| `doc` | `Doc` | The document to process. | +| **RETURNS** | `Doc` | The processed document. | + +## SentenceRecognizer.pipe {#pipe tag="method"} + +Apply the pipe to a stream of documents. This usually happens under the hood +when the `nlp` object is called on a text and all pipeline components are +applied to the `Doc` in order. Both [`__call__`](/api/sentencerecognizer#call) +and [`pipe`](/api/sentencerecognizer#pipe) delegate to the +[`predict`](/api/sentencerecognizer#predict) and +[`set_annotations`](/api/sentencerecognizer#set_annotations) methods. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> for doc in senter.pipe(docs, batch_size=50): +> pass +> ``` + +| Name | Type | Description | +| -------------- | --------------- | ------------------------------------------------------ | +| `stream` | `Iterable[Doc]` | A stream of documents. | +| _keyword-only_ | | | +| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | +| **YIELDS** | `Doc` | Processed documents in the order of the original text. | + +## SentenceRecognizer.begin_training {#begin_training tag="method"} + +Initialize the pipe for training, using data examples if available. Return an +[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> optimizer = senter.begin_training(pipeline=nlp.pipeline) +> ``` + +| Name | Type | Description | +| -------------- | --------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------- | +| `get_examples` | `Callable[[], Iterable[Example]]` | Optional function that returns gold-standard annotations in the form of [`Example`](/api/example) objects. | +| _keyword-only_ | | | +| `pipeline` | `List[Tuple[str, Callable]]` | Optional list of pipeline components that this component is part of. | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | An optional optimizer. Will be created via [`create_optimizer`](/api/sentencerecognizer#create_optimizer) if not set. | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | + +## SentenceRecognizer.predict {#predict tag="method"} + +Apply the pipeline's model to a batch of docs, without modifying them. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> scores = senter.predict([doc1, doc2]) +> ``` + +| Name | Type | Description | +| ----------- | --------------- | ----------------------------------------- | +| `docs` | `Iterable[Doc]` | The documents to predict. | +| **RETURNS** | - | The model's prediction for each document. | + +## SentenceRecognizer.set_annotations {#set_annotations tag="method"} + +Modify a batch of documents, using pre-computed scores. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> scores = senter.predict([doc1, doc2]) +> senter.set_annotations([doc1, doc2], scores) +> ``` + +| Name | Type | Description | +| -------- | --------------- | ------------------------------------------------------------ | +| `docs` | `Iterable[Doc]` | The documents to modify. | +| `scores` | - | The scores to set, produced by `SentenceRecognizer.predict`. | + +## SentenceRecognizer.update {#update tag="method"} + +Learn from a batch of documents and gold-standard information, updating the +pipe's model. Delegates to [`predict`](/api/sentencerecognizer#predict) and +[`get_loss`](/api/sentencerecognizer#get_loss). + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> optimizer = nlp.begin_training() +> losses = senter.update(examples, sgd=optimizer) +> ``` + +| Name | Type | Description | +| ----------------- | --------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ | +| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | +| _keyword-only_ | | | +| `drop` | float | The dropout rate. | +| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/sentencerecognizer#set_annotations). | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | +| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | + +## SentenceRecognizer.rehearse {#rehearse tag="method,experimental"} + +Perform a "rehearsal" update from a batch of data. Rehearsal updates teach the +current model to make predictions similar to an initial model, to try to address +the "catastrophic forgetting" problem. This feature is experimental. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> optimizer = nlp.begin_training() +> losses = senter.rehearse(examples, sgd=optimizer) +> ``` + +| Name | Type | Description | +| -------------- | --------------------------------------------------- | ----------------------------------------------------------------------------------------- | +| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | +| _keyword-only_ | | | +| `drop` | float | The dropout rate. | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. Updated using the component name as the key. | +| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | + +## SentenceRecognizer.get_loss {#get_loss tag="method"} + +Find the loss and gradient of loss for the batch of documents and their +predicted scores. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> scores = senter.predict([eg.predicted for eg in examples]) +> loss, d_loss = senter.get_loss(examples, scores) +> ``` + +| Name | Type | Description | +| ----------- | --------------------- | --------------------------------------------------- | +| `examples` | `Iterable[Example]` | The batch of examples. | +| `scores` | - | Scores representing the model's predictions. | +| **RETURNS** | `Tuple[float, float]` | The loss and the gradient, i.e. `(loss, gradient)`. | + +## SentenceRecognizer.score {#score tag="method" new="3"} + +Score a batch of examples. + +> #### Example +> +> ```python +> scores = senter.score(examples) +> ``` + +| Name | Type | Description | +| ----------- | ------------------- | ------------------------------------------------------------------------ | +| `examples` | `Iterable[Example]` | The examples to score. | +| **RETURNS** | `Dict[str, Any]` | The scores, produced by [`Scorer.score_spans`](/api/scorer#score_spans). | + +## SentenceRecognizer.create_optimizer {#create_optimizer tag="method"} + +Create an optimizer for the pipeline component. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> optimizer = senter.create_optimizer() +> ``` + +| Name | Type | Description | +| ----------- | --------------------------------------------------- | -------------- | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | + +## SentenceRecognizer.use_params {#use_params tag="method, contextmanager"} + +Modify the pipe's model, to use the given parameter values. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> with senter.use_params(): +> senter.to_disk("/best_model") +> ``` + +| Name | Type | Description | +| -------- | ---- | ---------------------------------------------------------------------------------------------------------- | +| `params` | - | The parameter values to use in the model. At the end of the context, the original parameters are restored. | + +## SentenceRecognizer.to_disk {#to_disk tag="method"} + +Serialize the pipe to disk. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> senter.to_disk("/path/to/senter") +> ``` + +| Name | Type | Description | +| --------- | --------------- | --------------------------------------------------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | + +## SentenceRecognizer.from_disk {#from_disk tag="method"} + +Load the pipe from disk. Modifies the object in place and returns it. + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> senter.from_disk("/path/to/senter") +> ``` + +| Name | Type | Description | +| ----------- | -------------------- | -------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | `SentenceRecognizer` | The modified `SentenceRecognizer` object. | + +## SentenceRecognizer.to_bytes {#to_bytes tag="method"} + +> #### Example +> +> ```python +> senter = nlp.add_pipe("senter") +> senter_bytes = senter.to_bytes() +> ``` + +Serialize the pipe to a bytestring. + +| Name | Type | Description | +| ----------- | --------------- | ------------------------------------------------------------------------- | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | bytes | The serialized form of the `SentenceRecognizer` object. | + +## SentenceRecognizer.from_bytes {#from_bytes tag="method"} + +Load the pipe from a bytestring. Modifies the object in place and returns it. + +> #### Example +> +> ```python +> senter_bytes = senter.to_bytes() +> senter = nlp.add_pipe("senter") +> senter.from_bytes(senter_bytes) +> ``` + +| Name | Type | Description | +| ------------ | -------------------- | ------------------------------------------------------------------------- | +| `bytes_data` | bytes | The data to load from. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | `SentenceRecognizer` | The `SentenceRecognizer` object. | + +## Serialization fields {#serialization-fields} + +During serialization, spaCy will export several data fields used to restore +different aspects of the object. If needed, you can exclude them from +serialization by passing in the string names via the `exclude` argument. + +> #### Example +> +> ```python +> data = senter.to_disk("/path", exclude=["vocab"]) +> ``` + +| Name | Description | +| ------- | -------------------------------------------------------------- | +| `vocab` | The shared [`Vocab`](/api/vocab). | +| `cfg` | The config file. You usually don't want to exclude this. | +| `model` | The binary model data. You usually don't want to exclude this. | diff --git a/website/docs/api/sentencizer.md b/website/docs/api/sentencizer.md index 181babe06..198215cfa 100644 --- a/website/docs/api/sentencizer.md +++ b/website/docs/api/sentencizer.md @@ -1,15 +1,40 @@ --- title: Sentencizer tag: class -source: spacy/pipeline/pipes.pyx +source: spacy/pipeline/sentencizer.pyx +teaser: 'Pipeline component for rule-based sentence boundary detection' +api_base_class: /api/pipe +api_string_name: sentencizer +api_trainable: false --- A simple pipeline component, to allow custom sentence boundary detection logic that doesn't require the dependency parse. By default, sentence segmentation is performed by the [`DependencyParser`](/api/dependencyparser), so the `Sentencizer` lets you implement a simpler, rule-based strategy that doesn't -require a statistical model to be loaded. The component is also available via -the string name `"sentencizer"`. +require a statistical model to be loaded. + +## Config and implementation {#config} + +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). + +> #### Example +> +> ```python +> config = {"punct_chars": None} +> nlp.add_pipe("entity_ruler", config=config) +> ``` + +| Setting | Type | Description | Default | +| ------------- | ----------- | ---------------------------------------------------------------------------------------------------------- | ------- | +| `punct_chars` | `List[str]` | Optional custom list of punctuation characters that mark sentence ends. See below for defaults if not set. | `None` | + +```python +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/sentencizer.pyx +``` ## Sentencizer.\_\_init\_\_ {#init tag="method"} @@ -20,12 +45,16 @@ Initialize the sentencizer. > ```python > # Construction via add_pipe > sentencizer = nlp.add_pipe("sentencizer") +> +> # Construction from class +> from spacy.pipeline import Sentencizer +> sentencizer = Sentencizer() > ``` -| Name | Type | Description | -| ------------- | ------------- | ----------------------------------------------------------------------------------------------- | -| `punct_chars` | list | Optional custom list of punctuation characters that mark sentence ends. See below for defaults. | -| **RETURNS** | `Sentencizer` | The newly constructed object. | +| Name | Type | Description | +| -------------- | ----------- | ----------------------------------------------------------------------------------------------- | +| _keyword-only_ | | | +| `punct_chars` | `List[str]` | Optional custom list of punctuation characters that mark sentence ends. See below for defaults. | ```python ### punct_chars defaults @@ -63,6 +92,42 @@ the component has been added to the pipeline using | `doc` | `Doc` | The `Doc` object to process, e.g. the `Doc` in the pipeline. | | **RETURNS** | `Doc` | The modified `Doc` with added sentence boundaries. | +## Sentencizer.pipe {#pipe tag="method"} + +Apply the pipe to a stream of documents. This usually happens under the hood +when the `nlp` object is called on a text and all pipeline components are +applied to the `Doc` in order. + +> #### Example +> +> ```python +> sentencizer = nlp.add_pipe("sentencizer") +> for doc in sentencizer.pipe(docs, batch_size=50): +> pass +> ``` + +| Name | Type | Description | +| -------------- | --------------- | ----------------------------------------------------- | +| `stream` | `Iterable[Doc]` | A stream of documents. | +| _keyword-only_ | | | +| `batch_size` | int | The number of documents to buffer. Defaults to `128`. | +| **YIELDS** | `Doc` | The processed documents in order. | + +## Sentencizer.score {#score tag="method" new="3"} + +Score a batch of examples. + +> #### Example +> +> ```python +> scores = sentencizer.score(examples) +> ``` + +| Name | Type | Description | +| ----------- | ------------------- | ------------------------------------------------------------------------ | +| `examples` | `Iterable[Example]` | The examples to score. | +| **RETURNS** | `Dict[str, Any]` | The scores, produced by [`Scorer.score_spans`](/api/scorer#score_spans). | + ## Sentencizer.to_disk {#to_disk tag="method"} Save the sentencizer settings (punctuation characters) a directory. Will create @@ -72,13 +137,14 @@ a file `sentencizer.json`. This also happens automatically when you save an > #### Example > > ```python -> sentencizer = Sentencizer(punct_chars=[".", "?", "!", "。"]) -> sentencizer.to_disk("/path/to/sentencizer.jsonl") +> config = {"punct_chars": [".", "?", "!", "。"]} +> sentencizer = nlp.add_pipe("sentencizer", config=config) +> sentencizer.to_disk("/path/to/sentencizer.json") > ``` -| Name | Type | Description | -| ------ | ------------ | ---------------------------------------------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a file, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | +| Name | Type | Description | +| ------ | ------------ | --------------------------------------------------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a JSON file, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | ## Sentencizer.from_disk {#from_disk tag="method"} @@ -89,7 +155,7 @@ added to its pipeline. > #### Example > > ```python -> sentencizer = Sentencizer() +> sentencizer = nlp.add_pipe("sentencizer") > sentencizer.from_disk("/path/to/sentencizer.json") > ``` @@ -105,7 +171,8 @@ Serialize the sentencizer settings to a bytestring. > #### Example > > ```python -> sentencizer = Sentencizer(punct_chars=[".", "?", "!", "。"]) +> config = {"punct_chars": [".", "?", "!", "。"]} +> sentencizer = nlp.add_pipe("sentencizer", config=config) > sentencizer_bytes = sentencizer.to_bytes() > ``` @@ -121,7 +188,7 @@ Load the pipe from a bytestring. Modifies the object in place and returns it. > > ```python > sentencizer_bytes = sentencizer.to_bytes() -> sentencizer = Sentencizer() +> sentencizer = nlp.add_pipe("sentencizer") > sentencizer.from_bytes(sentencizer_bytes) > ``` diff --git a/website/docs/api/tagger.md b/website/docs/api/tagger.md index 82658b8e0..9de4667bf 100644 --- a/website/docs/api/tagger.md +++ b/website/docs/api/tagger.md @@ -2,11 +2,40 @@ title: Tagger tag: class source: spacy/pipeline/tagger.pyx +teaser: 'Pipeline component for part-of-speech tagging' +api_base_class: /api/pipe +api_string_name: tagger +api_trainable: true --- -This class is a subclass of `Pipe` and follows the same API. The pipeline -component is available in the [processing pipeline](/usage/processing-pipelines) -via the ID `"tagger"`. +## Config and implementation {#config} + +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). See the +[model architectures](/api/architectures) documentation for details on the +architectures and their arguments and hyperparameters. + +> #### Example +> +> ```python +> from spacy.pipeline.tagger import DEFAULT_TAGGER_MODEL +> config = { +> "set_morphology": False, +> "model": DEFAULT_TAGGER_MODEL, +> } +> nlp.add_pipe("tagger", config=config) +> ``` + +| Setting | Type | Description | Default | +| ---------------- | ------------------------------------------ | -------------------------------------- | ----------------------------------- | +| `set_morphology` | bool | Whether to set morphological features. | `False` | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The model to use. | [Tagger](/api/architectures#Tagger) | + +```python +https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/tagger.pyx +``` ## Tagger.\_\_init\_\_ {#init tag="method"} @@ -18,19 +47,24 @@ via the ID `"tagger"`. > > # Construction via create_pipe with custom model > config = {"model": {"@architectures": "my_tagger"}} -> parser = nlp.add_pipe("tagger", config) +> parser = nlp.add_pipe("tagger", config=config) +> +> # Construction from class +> from spacy.pipeline import Tagger +> tagger = Tagger(nlp.vocab, model) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and [`nlp.add_pipe`](/api/language#add_pipe). -| Name | Type | Description | -| ----------- | -------- | ------------------------------------------------------------------------------- | -| `vocab` | `Vocab` | The shared vocabulary. | -| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | -| `**cfg` | - | Configuration parameters. | -| **RETURNS** | `Tagger` | The newly constructed object. | +| Name | Type | Description | +| ---------------- | ------- | ------------------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | The shared vocabulary. | +| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | +| `name` | str | String name of the component instance. Used to add entries to the `losses` during training. | +| _keyword-only_ | | | +| `set_morphology` | bool | Whether to set morphological features. | ## Tagger.\_\_call\_\_ {#call tag="method"} @@ -44,8 +78,8 @@ and all pipeline components are applied to the `Doc` in order. Both > #### Example > > ```python -> tagger = Tagger(nlp.vocab) > doc = nlp("This is a sentence.") +> tagger = nlp.add_pipe("tagger") > # This usually happens under the hood > processed = tagger(doc) > ``` @@ -66,16 +100,37 @@ applied to the `Doc` in order. Both [`__call__`](/api/tagger#call) and > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > for doc in tagger.pipe(docs, batch_size=50): > pass > ``` -| Name | Type | Description | -| ------------ | --------------- | ------------------------------------------------------ | -| `stream` | `Iterable[Doc]` | A stream of documents. | -| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | -| **YIELDS** | `Doc` | Processed documents in the order of the original text. | +| Name | Type | Description | +| -------------- | --------------- | ------------------------------------------------------ | +| `stream` | `Iterable[Doc]` | A stream of documents. | +| _keyword-only_ | | | +| `batch_size` | int | The number of texts to buffer. Defaults to `128`. | +| **YIELDS** | `Doc` | Processed documents in the order of the original text. | + +## Tagger.begin_training {#begin_training tag="method"} + +Initialize the pipe for training, using data examples if available. Return an +[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. + +> #### Example +> +> ```python +> tagger = nlp.add_pipe("tagger") +> optimizer = tagger.begin_training(pipeline=nlp.pipeline) +> ``` + +| Name | Type | Description | +| -------------- | --------------------------------------------------- | ---------------------------------------------------------------------------------------------------------- | +| `get_examples` | `Callable[[], Iterable[Example]]` | Optional function that returns gold-standard annotations in the form of [`Example`](/api/example) objects. | +| _keyword-only_ | | | +| `pipeline` | `List[Tuple[str, Callable]]` | Optional list of pipeline components that this component is part of. | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | An optional optimizer. Will be created via [`create_optimizer`](/api/tagger#create_optimizer) if not set. | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## Tagger.predict {#predict tag="method"} @@ -84,7 +139,7 @@ Apply the pipeline's model to a batch of docs, without modifying them. > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > scores = tagger.predict([doc1, doc2]) > ``` @@ -100,7 +155,7 @@ Modify a batch of documents, using pre-computed scores. > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > scores = tagger.predict([doc1, doc2]) > tagger.set_annotations([doc1, doc2], scores) > ``` @@ -119,20 +174,43 @@ pipe's model. Delegates to [`predict`](/api/tagger#predict) and > #### Example > > ```python -> tagger = Tagger(nlp.vocab, tagger_model) +> tagger = nlp.add_pipe("tagger") > optimizer = nlp.begin_training() > losses = tagger.update(examples, sgd=optimizer) > ``` -| Name | Type | Description | -| ----------------- | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------ | -| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | -| _keyword-only_ | | | -| `drop` | float | The dropout rate. | -| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/tagger#set_annotations). | -| `sgd` | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | -| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | -| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | +| Name | Type | Description | +| ----------------- | --------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------ | +| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | +| _keyword-only_ | | | +| `drop` | float | The dropout rate. | +| `set_annotations` | bool | Whether or not to update the `Example` objects with the predictions, delegating to [`set_annotations`](/api/tagger#set_annotations). | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. The value keyed by the model's name is updated. | +| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | + +## Tagger.rehearse {#rehearse tag="method,experimental"} + +Perform a "rehearsal" update from a batch of data. Rehearsal updates teach the +current model to make predictions similar to an initial model, to try to address +the "catastrophic forgetting" problem. This feature is experimental. + +> #### Example +> +> ```python +> tagger = nlp.add_pipe("tagger") +> optimizer = nlp.begin_training() +> losses = tagger.rehearse(examples, sgd=optimizer) +> ``` + +| Name | Type | Description | +| -------------- | --------------------------------------------------- | ----------------------------------------------------------------------------------------- | +| `examples` | `Iterable[Example]` | A batch of [`Example`](/api/example) objects to learn from. | +| _keyword-only_ | | | +| `drop` | float | The dropout rate. | +| `sgd` | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | +| `losses` | `Dict[str, float]` | Optional record of the loss during training. Updated using the component name as the key. | +| **RETURNS** | `Dict[str, float]` | The updated `losses` dictionary. | ## Tagger.get_loss {#get_loss tag="method"} @@ -142,36 +220,31 @@ predicted scores. > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > scores = tagger.predict([eg.predicted for eg in examples]) > loss, d_loss = tagger.get_loss(examples, scores) > ``` -| Name | Type | Description | -| ----------- | ------------------- | --------------------------------------------------- | -| `examples` | `Iterable[Example]` | The batch of examples. | -| `scores` | - | Scores representing the model's predictions. | -| **RETURNS** | tuple | The loss and the gradient, i.e. `(loss, gradient)`. | +| Name | Type | Description | +| ----------- | --------------------- | --------------------------------------------------- | +| `examples` | `Iterable[Example]` | The batch of examples. | +| `scores` | - | Scores representing the model's predictions. | +| **RETURNS** | `Tuple[float, float]` | The loss and the gradient, i.e. `(loss, gradient)`. | -## Tagger.begin_training {#begin_training tag="method"} +## Tagger.score {#score tag="method" new="3"} -Initialize the pipe for training, using data examples if available. Return an -[`Optimizer`](https://thinc.ai/docs/api-optimizers) object. +Score a batch of examples. > #### Example > > ```python -> tagger = Tagger(nlp.vocab) -> nlp.pipeline.append(tagger) -> optimizer = tagger.begin_training(pipeline=nlp.pipeline) +> scores = tagger.score(examples) > ``` -| Name | Type | Description | -| -------------- | ----------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------- | -| `get_examples` | `Iterable[Example]` | Optional gold-standard annotations in the form of [`Example`](/api/example) objects. | -| `pipeline` | `List[(str, callable)]` | Optional list of pipeline components that this component is part of. | -| `sgd` | `Optimizer` | An optional [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. Will be created via [`create_optimizer`](/api/tagger#create_optimizer) if not set. | -| **RETURNS** | `Optimizer` | An optimizer. | +| Name | Type | Description | +| ----------- | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------ | +| `examples` | `Iterable[Example]` | The examples to score. | +| **RETURNS** | `Dict[str, Any]` | The scores, produced by [`Scorer.score_token_attr`](/api/scorer#score_token_attr) for the attributes `"pos"`, `"tag"` and `"lemma"`. | ## Tagger.create_optimizer {#create_optimizer tag="method"} @@ -180,13 +253,13 @@ Create an optimizer for the pipeline component. > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > optimizer = tagger.create_optimizer() > ``` -| Name | Type | Description | -| ----------- | ----------- | --------------------------------------------------------------- | -| **RETURNS** | `Optimizer` | The [`Optimizer`](https://thinc.ai/docs/api-optimizers) object. | +| Name | Type | Description | +| ----------- | --------------------------------------------------- | -------------- | +| **RETURNS** | [`Optimizer`](https://thinc.ai/docs/api-optimizers) | The optimizer. | ## Tagger.use_params {#use_params tag="method, contextmanager"} @@ -195,7 +268,7 @@ Modify the pipe's model, to use the given parameter values. > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > with tagger.use_params(): > tagger.to_disk("/best_model") > ``` @@ -212,14 +285,14 @@ Add a new label to the pipe. > > ```python > from spacy.symbols import POS -> tagger = Tagger(nlp.vocab) -> tagger.add_label("MY_LABEL", {POS: 'NOUN'}) +> tagger = nlp.add_pipe("tagger") +> tagger.add_label("MY_LABEL", {POS: "NOUN"}) > ``` -| Name | Type | Description | -| -------- | ---- | --------------------------------------------------------------- | -| `label` | str | The label to add. | -| `values` | dict | Optional values to map to the label, e.g. a tag map dictionary. | +| Name | Type | Description | +| -------- | ---------------- | --------------------------------------------------------------- | +| `label` | str | The label to add. | +| `values` | `Dict[int, str]` | Optional values to map to the label, e.g. a tag map dictionary. | ## Tagger.to_disk {#to_disk tag="method"} @@ -228,14 +301,14 @@ Serialize the pipe to disk. > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > tagger.to_disk("/path/to/tagger") > ``` -| Name | Type | Description | -| --------- | ------------ | --------------------------------------------------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | +| Name | Type | Description | +| --------- | --------------- | --------------------------------------------------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | ## Tagger.from_disk {#from_disk tag="method"} @@ -244,31 +317,31 @@ Load the pipe from disk. Modifies the object in place and returns it. > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > tagger.from_disk("/path/to/tagger") > ``` -| Name | Type | Description | -| ----------- | ------------ | -------------------------------------------------------------------------- | -| `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | `Tagger` | The modified `Tagger` object. | +| Name | Type | Description | +| ----------- | --------------- | -------------------------------------------------------------------------- | +| `path` | str / `Path` | A path to a directory. Paths may be either strings or `Path`-like objects. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | `Tagger` | The modified `Tagger` object. | ## Tagger.to_bytes {#to_bytes tag="method"} > #### Example > > ```python -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > tagger_bytes = tagger.to_bytes() > ``` Serialize the pipe to a bytestring. -| Name | Type | Description | -| ----------- | ----- | ------------------------------------------------------------------------- | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | bytes | The serialized form of the `Tagger` object. | +| Name | Type | Description | +| ----------- | --------------- | ------------------------------------------------------------------------- | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | bytes | The serialized form of the `Tagger` object. | ## Tagger.from_bytes {#from_bytes tag="method"} @@ -278,15 +351,15 @@ Load the pipe from a bytestring. Modifies the object in place and returns it. > > ```python > tagger_bytes = tagger.to_bytes() -> tagger = Tagger(nlp.vocab) +> tagger = nlp.add_pipe("tagger") > tagger.from_bytes(tagger_bytes) > ``` -| Name | Type | Description | -| ------------ | -------- | ------------------------------------------------------------------------- | -| `bytes_data` | bytes | The data to load from. | -| `exclude` | list | String names of [serialization fields](#serialization-fields) to exclude. | -| **RETURNS** | `Tagger` | The `Tagger` object. | +| Name | Type | Description | +| ------------ | --------------- | ------------------------------------------------------------------------- | +| `bytes_data` | bytes | The data to load from. | +| `exclude` | `Iterable[str]` | String names of [serialization fields](#serialization-fields) to exclude. | +| **RETURNS** | `Tagger` | The `Tagger` object. | ## Tagger.labels {#labels tag="property"} @@ -301,9 +374,9 @@ tags by default, e.g. `VERB`, `NOUN` and so on. > assert "MY_LABEL" in tagger.labels > ``` -| Name | Type | Description | -| ----------- | ----- | ---------------------------------- | -| **RETURNS** | tuple | The labels added to the component. | +| Name | Type | Description | +| ----------- | ------------ | ---------------------------------- | +| **RETURNS** | `Tuple[str]` | The labels added to the component. | ## Serialization fields {#serialization-fields} diff --git a/website/docs/api/textcategorizer.md b/website/docs/api/textcategorizer.md index 52b7e399c..5f981de82 100644 --- a/website/docs/api/textcategorizer.md +++ b/website/docs/api/textcategorizer.md @@ -3,18 +3,36 @@ title: TextCategorizer tag: class source: spacy/pipeline/textcat.py new: 2 +teaser: 'Pipeline component for text classification' +api_base_class: /api/pipe +api_string_name: textcat +api_trainable: true --- -This class is a subclass of `Pipe` and follows the same API. The pipeline -component is available in the [processing pipeline](/usage/processing-pipelines) -via the ID `"textcat"`. +## Config and implementation {#config} -## Implementation and defaults {#implementation} +The default config is defined by the pipeline component factory and describes +how the component should be configured. You can override its settings via the +`config` argument on [`nlp.add_pipe`](/api/language#add_pipe) or in your +[`config.cfg` for training](/usage/training#config). See the +[model architectures](/api/architectures) documentation for details on the +architectures and their arguments and hyperparameters. -See the [model architectures](/api/architectures) documentation for details on -the architectures and their arguments and hyperparameters. To learn more about -how to customize the config and train custom models, check out the -[training config](/usage/training#config) docs. +> #### Example +> +> ```python +> from spacy.pipeline.textcat import DEFAULT_TEXTCAT_MODEL +> config = { +> "labels": [], +> "model": DEFAULT_TEXTCAT_MODEL, +> } +> nlp.add_pipe("textcat", config=config) +> ``` + +| Setting | Type | Description | Default | +| -------- | ------------------------------------------ | ------------------ | ----------------------------------------------------- | +| `labels` | `Iterable[str]` | The labels to use. | `[]` | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The model to use. | [TextCatEnsemble](/api/architectures#TextCatEnsemble) | ```python https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/textcat.py @@ -31,18 +49,23 @@ https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/textcat.py > # Construction via add_pipe with custom model > config = {"model": {"@architectures": "my_textcat"}} > parser = nlp.add_pipe("textcat", config=config) +> +> # Construction from class +> from spacy.pipeline import TextCategorizer +> textcat = TextCategorizer(nlp.vocab, model) > ``` Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and [`nlp.add_pipe`](/api/language#create_pipe). -| Name | Type | Description | -| ----------- | ----------------- | ------------------------------------------------------------------------------- | -| `vocab` | `Vocab` | The shared vocabulary. | -| `model` | `Model` | The [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | -| `**cfg` | - | Configuration parameters. | -| **RETURNS** | `TextCategorizer` | The newly constructed object. | +| Name | Type | Description | +| -------------- | ------------------------------------------ | ------------------------------------------------------------------------------------------- | +| `vocab` | `Vocab` | The shared vocabulary. | +| `model` | [`Model`](https://thinc.ai/docs/api-model) | The Thinc [`Model`](https://thinc.ai/docs/api-model) powering the pipeline component. | +| `name` | str | String name of the component instance. Used to add entries to the `losses` during training. | +| _keyword-only_ | | | +| `labels` | `Iterable[str]` | The labels to use. | - -## Implementation and defaults {#implementation} - -See the [model architectures](/api/architectures) documentation for details on -the architectures and their arguments and hyperparameters. To learn more about -how to customize the config and train custom models, check out the -[training config](/usage/training#config) docs. - -```python -https://github.com/explosion/spaCy/blob/develop/spacy/pipeline/tok2vec.py -``` +TODO: diff --git a/website/docs/api/top-level.md b/website/docs/api/top-level.md index 9c4380626..a463441c7 100644 --- a/website/docs/api/top-level.md +++ b/website/docs/api/top-level.md @@ -31,11 +31,12 @@ loaded in via [`Language.from_disk`](/api/language#from_disk). > nlp = spacy.load("en_core_web_sm", disable=["parser", "tagger"]) > ``` -| Name | Type | Description | -| ----------- | ------------ | --------------------------------------------------------------------------------- | -| `name` | str / `Path` | Model to load, i.e. package name or path. | -| `disable` | `List[str]` | Names of pipeline components to [disable](/usage/processing-pipelines#disabling). | -| **RETURNS** | `Language` | A `Language` object with the loaded model. | +| Name | Type | Description | +| ------------------------------------------ | ----------------- | --------------------------------------------------------------------------------- | +| `name` | str / `Path` | Model to load, i.e. package name or path. | +| `disable` | `List[str]` | Names of pipeline components to [disable](/usage/processing-pipelines#disabling). | +| `component_cfg` 3 | `Dict[str, dict]` | Optional config overrides for pipeline components, keyed by component names. | +| **RETURNS** | `Language` | A `Language` object with the loaded model. | Essentially, `spacy.load()` is a convenience wrapper that reads the language ID and pipeline components from a model's `meta.json`, initializes the `Language` @@ -43,10 +44,10 @@ class, loads in the model data and returns it. ```python ### Abstract example -cls = util.get_lang_class(lang) # get language for ID, e.g. 'en' -nlp = cls() # initialise the language +cls = util.get_lang_class(lang) # get language for ID, e.g. "en" +nlp = cls() # initialize the language for name in pipeline: - nlp.add_pipe(name) # add component to pipeline + nlp.add_pipe(name) # add component to pipeline nlp.from_disk(model_data_path) # load in model data ``` @@ -58,15 +59,14 @@ Create a blank model of a given language class. This function is the twin of > #### Example > > ```python -> nlp_en = spacy.blank("en") -> nlp_de = spacy.blank("de") +> nlp_en = spacy.blank("en") # equivalent to English() +> nlp_de = spacy.blank("de") # equivalent to German() > ``` -| Name | Type | Description | -| ----------- | ----------- | ------------------------------------------------------------------------------------------------ | -| `name` | str | [ISO code](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) of the language class to load. | -| `disable` | `List[str]` | Names of pipeline components to [disable](/usage/processing-pipelines#disabling). | -| **RETURNS** | `Language` | An empty `Language` object of the appropriate subclass. | +| Name | Type | Description | +| ----------- | ---------- | ------------------------------------------------------------------------------------------------ | +| `name` | str | [ISO code](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) of the language class to load. | +| **RETURNS** | `Language` | An empty `Language` object of the appropriate subclass. | #### spacy.info {#spacy.info tag="function"} @@ -80,13 +80,14 @@ meta data as a dictionary instead, you can use the `meta` attribute on your > ```python > spacy.info() > spacy.info("en_core_web_sm") -> spacy.info(markdown=True) +> markdown = spacy.info(markdown=True, silent=True) > ``` | Name | Type | Description | | ---------- | ---- | ------------------------------------------------ | | `model` | str | A model, i.e. a package name or path (optional). | | `markdown` | bool | Print information as Markdown. | +| `silent` | bool | Don't print anything, just return. | ### spacy.explain {#spacy.explain tag="function"} diff --git a/website/docs/api/vectors.md b/website/docs/api/vectors.md index a893a1fd2..a0f7ef88b 100644 --- a/website/docs/api/vectors.md +++ b/website/docs/api/vectors.md @@ -30,13 +30,14 @@ you can add vectors to later. > vectors = Vectors(data=data, keys=keys) > ``` -| Name | Type | Description | -| ----------- | ---------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -| `data` | `ndarray[ndim=1, dtype='float32']` | The vector data. | -| `keys` | iterable | A sequence of keys aligned with the data. | -| `shape` | tuple | Size of the table as `(n_entries, n_columns)`, the number of entries and number of columns. Not required if you're initializing the object with `data` and `keys`. | -| `name` | str | A name to identify the vectors table. | -| **RETURNS** | `Vectors` | The newly created object. | +| Name | Type | Description | +| -------------- | ---------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------ | +| _keyword-only_ | | | +| `shape` | tuple | Size of the table as `(n_entries, n_columns)`, the number of entries and number of columns. Not required if you're initializing the object with `data` and `keys`. | +| `data` | `ndarray[ndim=1, dtype='float32']` | The vector data. | +| `keys` | iterable | A sequence of keys aligned with the data. | +| `name` | str | A name to identify the vectors table. | +| **RETURNS** | `Vectors` | The newly created object. | ## Vectors.\_\_getitem\_\_ {#getitem tag="method"} @@ -138,12 +139,13 @@ mapping separately. If you need to manage the strings, you should use the > nlp.vocab.vectors.add("dog", row=0) > ``` -| Name | Type | Description | -| ----------- | ---------------------------------- | ----------------------------------------------------- | -| `key` | str / int | The key to add. | -| `vector` | `ndarray[ndim=1, dtype='float32']` | An optional vector to add for the key. | -| `row` | int | An optional row number of a vector to map the key to. | -| **RETURNS** | int | The row the vector was added to. | +| Name | Type | Description | +| -------------- | ---------------------------------- | ----------------------------------------------------- | +| `key` | str / int | The key to add. | +| _keyword-only_ | | | +| `vector` | `ndarray[ndim=1, dtype='float32']` | An optional vector to add for the key. | +| `row` | int | An optional row number of a vector to map the key to. | +| **RETURNS** | int | The row the vector was added to. | ## Vectors.resize {#resize tag="method"} @@ -225,13 +227,14 @@ Look up one or more keys by row, or vice versa. > keys = nlp.vocab.vectors.find(rows=[18, 256, 985]) > ``` -| Name | Type | Description | -| ----------- | ------------------------------------- | ------------------------------------------------------------------------ | -| `key` | str / int | Find the row that the given key points to. Returns int, `-1` if missing. | -| `keys` | iterable | Find rows that the keys point to. Returns `ndarray`. | -| `row` | int | Find the first key that points to the row. Returns int. | -| `rows` | iterable | Find the keys that point to the rows. Returns ndarray. | -| **RETURNS** | The requested key, keys, row or rows. | +| Name | Type | Description | +| -------------- | ------------------------------------- | ------------------------------------------------------------------------ | +| _keyword-only_ | | | +| `key` | str / int | Find the row that the given key points to. Returns int, `-1` if missing. | +| `keys` | iterable | Find rows that the keys point to. Returns `ndarray`. | +| `row` | int | Find the first key that points to the row. Returns int. | +| `rows` | iterable | Find the keys that point to the rows. Returns ndarray. | +| **RETURNS** | The requested key, keys, row or rows. | ## Vectors.shape {#shape tag="property"} @@ -318,13 +321,14 @@ performed in chunks, to avoid consuming too much memory. You can set the > most_similar = nlp.vocab.vectors.most_similar(queries, n=10) > ``` -| Name | Type | Description | -| ------------ | --------- | ------------------------------------------------------------------ | -| `queries` | `ndarray` | An array with one or more vectors. | -| `batch_size` | int | The batch size to use. Default to `1024`. | -| `n` | int | The number of entries to return for each query. Defaults to `1`. | -| `sort` | bool | Whether to sort the entries returned by score. Defaults to `True`. | -| **RETURNS** | tuple | The most similar entries as a `(keys, best_rows, scores)` tuple. | +| Name | Type | Description | +| -------------- | --------- | ------------------------------------------------------------------ | +| `queries` | `ndarray` | An array with one or more vectors. | +| _keyword-only_ | | | +| `batch_size` | int | The batch size to use. Default to `1024`. | +| `n` | int | The number of entries to return for each query. Defaults to `1`. | +| `sort` | bool | Whether to sort the entries returned by score. Defaults to `True`. | +| **RETURNS** | tuple | The most similar entries as a `(keys, best_rows, scores)` tuple. | ## Vectors.to_disk {#to_disk tag="method"} diff --git a/website/docs/api/vocab.md b/website/docs/api/vocab.md index a77b9f244..af9feb82c 100644 --- a/website/docs/api/vocab.md +++ b/website/docs/api/vocab.md @@ -136,10 +136,11 @@ have to call this to change the size of the vectors. Only one of the `width` and > nlp.vocab.reset_vectors(width=300) > ``` -| Name | Type | Description | -| ------- | ---- | -------------------------------------- | -| `width` | int | The new width (keyword argument only). | -| `shape` | int | The new shape (keyword argument only). | +| Name | Type | Description | +| -------------- | ---- | -------------------------------------- | +| _keyword-only_ | | | +| `width` | int | The new width (keyword argument only). | +| `shape` | int | The new shape (keyword argument only). | ## Vocab.prune_vectors {#prune_vectors tag="method" new="2"} diff --git a/website/docs/usage/training.md b/website/docs/usage/training.md index be0b7196a..948a13086 100644 --- a/website/docs/usage/training.md +++ b/website/docs/usage/training.md @@ -446,19 +446,6 @@ annotations: -> - **Training data**: The training examples. -> - **Text and label**: The current example. -> - **Doc**: A `Doc` object created from the example text. -> - **Example**: An `Example` object holding both predictions and gold-standard -> annotations. -> - **nlp**: The `nlp` object with the model. -> - **Optimizer**: A function that holds state between updates. -> - **Update**: Update the model's weights. - - - -![The training loop](../images/training-loop.svg) - Of course, it's not enough to only show a model a single example once. Especially if you only have few examples, you'll want to train for a **number of iterations**. At each iteration, the training data is **shuffled** to ensure the @@ -469,12 +456,16 @@ it harder for the model to memorize the training data. For example, a `0.25` dropout means that each feature or internal representation has a 1/4 likelihood of being dropped. -> - [`begin_training`](/api/language#begin_training): Start the training and -> return an [`Optimizer`](https://thinc.ai/docs/api-optimizers) object to -> update the model's weights. -> - [`update`](/api/language#update): Update the model with the training -> examplea. -> - [`to_disk`](/api/language#to_disk): Save the updated model to a directory. +> - [`nlp`](/api/language): The `nlp` object with the model. +> - [`nlp.begin_training`](/api/language#begin_training): Start the training and +> return an optimizer to update the model's weights. +> - [`Optimizer`](https://thinc.ai/docs/api-optimizers): Function that holds +> state between updates. +> - [`nlp.update`](/api/language#update): Update model with examples. +> - [`Example`](/api/example): object holding predictions and gold-standard +> annotations. +> - [`nlp.to_disk`](/api/language#to_disk): Save the updated model to a +> directory. ```python ### Example training loop diff --git a/website/docs/usage/v3.md b/website/docs/usage/v3.md index cd68a5242..24803e953 100644 --- a/website/docs/usage/v3.md +++ b/website/docs/usage/v3.md @@ -15,24 +15,183 @@ menu: ## Backwards Incompatibilities {#incompat} -### Removed deprecated methods, attributes and arguments {#incompat-removed} +### Removed or renamed objects, methods, attributes and arguments {#incompat-removed} + +| Removed | Replacement | +| -------------------------------------------------------- | ----------------------------------------- | +| `GoldParse` | [`Example`](/api/example) | +| `spacy debug-data` | [`spacy debug data`](/api/cli#debug-data) | +| `spacy link`, `util.set_data_path`, `util.get_data_path` | not needed, model symlinks are deprecated | + +### Removed deprecated methods, attributes and arguments {#incompat-removed-deprecated} The following deprecated methods, attributes and arguments were removed in v3.0. -Most of them have been deprecated for quite a while now and many would -previously raise errors. Many of them were also mostly internals. If you've been -working with more recent versions of spaCy v2.x, it's unlikely that your code -relied on them. +Most of them have been **deprecated for a while** and many would previously +raise errors. Many of them were also mostly internals. If you've been working +with more recent versions of spaCy v2.x, it's **unlikely** that your code relied +on them. -| Class | Removed | -| --------------------- | ------------------------------------------------------- | -| [`Doc`](/api/doc) | `Doc.tokens_from_list`, `Doc.merge` | -| [`Span`](/api/span) | `Span.merge`, `Span.upper`, `Span.lower`, `Span.string` | -| [`Token`](/api/token) | `Token.string` | - - +| Removed | Replacement | +| ----------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `Doc.tokens_from_list` | [`Doc.__init__`](/api/doc#init) | +| `Doc.merge`, `Span.merge` | [`Doc.retokenize`](/api/doc#retokenize) | +| `Token.string`, `Span.string`, `Span.upper`, `Span.lower` | [`Span.text`](/api/span#attributes), [`Token.text`](/api/token#attributes) | +| `Language.tagger`, `Language.parser`, `Language.entity` | [`Language.get_pipe`](/api/language#get_pipe) | +| keyword-arguments like `vocab=False` on `to_disk`, `from_disk`, `to_bytes`, `from_bytes` | `exclude=["vocab"]` | +| `n_threads` argument on [`Tokenizer`](/api/tokenizer), [`Matcher`](/api/matcher), [`PhraseMatcher`](/api/phrasematcher) | `n_process` | +| `SentenceSegmenter` hook, `SimilarityHook` | [user hooks](/usage/processing-pipelines#custom-components-user-hooks), [`Sentencizer`](/api/sentencizer), [`SentenceRecognizer`](/api/sentenceregognizer), | ## Migrating from v2.x {#migrating} +### Custom pipeline components and factories {#migrating-pipeline-components} + +Custom pipeline components now have to be registered explicitly using the +[`@Language.component`](/api/language#component) or +[`@Language.factory`](/api/language#factory) decorator. For simple functions +that take a `Doc` and return it, all you have to do is add the +`@Language.component` decorator to it and assign it a name: + +```diff +### Stateless function components ++ from spacy.language import Language + ++ @Language.component("my_component") +def my_component(doc): + return doc +``` + +For class components that are initialized with settings and/or the shared `nlp` +object, you can use the `@Language.factory` decorator. Also make sure that that +the method used to initialize the factory has **two named arguments**: `nlp` +(the current `nlp` object) and `name` (the string name of the component +instance). + +```diff +### Stateful class components ++ from spacy.language import Language + ++ @Language.factory("my_component") +class MyComponent: +- def __init__(self, nlp): ++ def __init__(self, nlp, name): + self.nlp = nlp + + def __call__(self, doc): + return doc +``` + +Instead of decorating your class, you could also add a factory function that +takes the arguments `nlp` and `name` and returns an instance of your component: + +```diff +### Stateful class components with factory function ++ from spacy.language import Language + ++ @Language.factory("my_component") ++ def create_my_component(nlp, name): ++ return MyComponent(nlp) + +class MyComponent: + def __init__(self, nlp): + self.nlp = nlp + + def __call__(self, doc): + return doc +``` + +The `@Language.component` and `@Language.factory` decorators now take care of +adding an entry to the component factories, so spaCy knows how to load a +component back in from its string name. You won't have to write to +`Language.factories` manually anymore. + +```diff +- Language.factories["my_component"] = lambda nlp, **cfg: MyComponent(nlp) +``` + +#### Adding components to the pipeline {#migrating-add-pipe} + +The [`nlp.add_pipe`](/api/language#add_pipe) method now takes the **string +name** of the component factory instead of a callable component. This allows +spaCy to track and serialize components that have been added and their settings. + +```diff ++ @Language.component("my_component") +def my_component(doc): + return doc + +- nlp.add_pipe(my_component) ++ nlp.add_pipe("my_component") +``` + +[`nlp.add_pipe`](/api/language#add_pipe) now also returns the pipeline component +itself, so you can access its attributes. The +[`nlp.create_pipe`](/api/language#create_pipe) method is now mostly internals +and you typically shouldn't have to use it in your code. + +```diff +- parser = nlp.create_pipe("parser") +- nlp.add_pipe(parser) ++ parser = nlp.add_pipe("parser") +``` + +### Training models {#migrating-training} + +To train your models, you should now pretty much always use the +[`spacy train`](/api/cli#train) CLI. You shouldn't have to put together your own +training scripts anymore, unless you _really_ want to. The training commands now +use a [flexible config file](/usage/training#config) that describes all training +settings and hyperparameters, as well as your pipeline, model components and +architectures to use. The `--code` argument lets you pass in code containing +[custom registered functions](/usage/training#custom-code) that you can +reference in your config. + +#### Binary .spacy training data format {#migrating-training-format} + +spaCy now uses a new +[binary training data format](/api/data-formats#binary-training), which is much +smaller and consists of `Doc` objects, serialized via the +[`DocBin`](/api/docbin). You can convert your existing JSON-formatted data using +the [`spacy convert`](/api/cli#convert) command, which outputs `.spacy` files: + +```bash +$ python -m spacy convert ./training.json ./output +``` + +#### Training config {#migrating-training-config} + + + +```diff +### {wrap="true"} +- python -m spacy train en ./output ./train.json ./dev.json --pipeline tagger,parser --cnn-window 1 --bilstm-depth 0 ++ python -m spacy train ./train.spacy ./dev.spacy ./config.cfg --output ./output +``` + + + +The easiest way to get started with an end-to-end training process is to clone a +[project](/usage/projects) template. Projects let you manage multi-step +workflows, from data preprocessing to training and packaging your model. + + + +#### Migrating training scripts to CLI command and config {#migrating-training-scripts} + + + +#### Packaging models {#migrating-training-packaging} + +The [`spacy package`](/api/cli#package) command now automatically builds the +installable `.tar.gz` sdist of the Python package, so you don't have to run this +step manually anymore. You can disable the behavior by setting the `--no-sdist` +flag. + +```diff +python -m spacy package ./model ./packages +- cd /output/en_model-0.0.0 +- python setup.py sdist +``` + ## Migration notes for plugin maintainers {#plugins} Thanks to everyone who's been contributing to the spaCy ecosystem by developing diff --git a/website/gatsby-node.js b/website/gatsby-node.js index fe9f22888..1f6fcf1f9 100644 --- a/website/gatsby-node.js +++ b/website/gatsby-node.js @@ -14,6 +14,10 @@ function getNodeTitle({ childMdx }) { return (frontmatter.title || '').replace("'", '’') } +function findNode(pages, slug) { + return slug ? pages.find(({ node }) => node.fields.slug === slug) : null +} + exports.createPages = ({ graphql, actions }) => { const { createPage } = actions @@ -70,6 +74,9 @@ exports.createPages = ({ graphql, actions }) => { title teaser source + api_base_class + api_string_name + api_trainable tag new next @@ -115,10 +122,18 @@ exports.createPages = ({ graphql, actions }) => { const section = frontmatter.section || page.node.relativeDirectory const sectionMeta = sections[section] || {} const title = getNodeTitle(page.node) - const next = frontmatter.next - ? pages.find(({ node }) => node.fields.slug === frontmatter.next) - : null - + const next = findNode(pages, frontmatter.next) + const baseClass = findNode(pages, frontmatter.api_base_class) + const apiDetails = { + stringName: frontmatter.api_string_name, + baseClass: baseClass + ? { + title: getNodeTitle(baseClass.node), + slug: frontmatter.api_base_class, + } + : null, + trainable: frontmatter.api_trainable, + } createPage({ path: replacePath(page.node.fields.slug), component: DEFAULT_TEMPLATE, @@ -131,6 +146,7 @@ exports.createPages = ({ graphql, actions }) => { sectionTitle: sectionMeta.title, menu: frontmatter.menu || [], teaser: frontmatter.teaser, + apiDetails, source: frontmatter.source, sidebar: frontmatter.sidebar, tag: frontmatter.tag, diff --git a/website/meta/sidebars.json b/website/meta/sidebars.json index 399c39dcf..96e1ea8d6 100644 --- a/website/meta/sidebars.json +++ b/website/meta/sidebars.json @@ -67,6 +67,7 @@ { "label": "Containers", "items": [ + { "text": "Language", "url": "/api/language" }, { "text": "Doc", "url": "/api/doc" }, { "text": "Token", "url": "/api/token" }, { "text": "Span", "url": "/api/span" }, @@ -78,7 +79,6 @@ { "label": "Pipeline", "items": [ - { "text": "Language", "url": "/api/language" }, { "text": "Tokenizer", "url": "/api/tokenizer" }, { "text": "Tok2Vec", "url": "/api/tok2vec" }, { "text": "Lemmatizer", "url": "/api/lemmatizer" }, @@ -86,14 +86,21 @@ { "text": "Tagger", "url": "/api/tagger" }, { "text": "DependencyParser", "url": "/api/dependencyparser" }, { "text": "EntityRecognizer", "url": "/api/entityrecognizer" }, + { "text": "EntityRuler", "url": "/api/entityruler" }, { "text": "EntityLinker", "url": "/api/entitylinker" }, { "text": "TextCategorizer", "url": "/api/textcategorizer" }, - { "text": "Matcher", "url": "/api/matcher" }, - { "text": "PhraseMatcher", "url": "/api/phrasematcher" }, - { "text": "EntityRuler", "url": "/api/entityruler" }, { "text": "Sentencizer", "url": "/api/sentencizer" }, { "text": "SentenceRecognizer", "url": "/api/sentencerecognizer" }, - { "text": "Other Functions", "url": "/api/pipeline-functions" } + { "text": "Other Functions", "url": "/api/pipeline-functions" }, + { "text": "Pipe", "url": "/api/pipe" } + ] + }, + { + "label": "Matchers", + "items": [ + { "text": "Matcher", "url": "/api/matcher" }, + { "text": "PhraseMatcher", "url": "/api/phrasematcher" }, + { "text": "DependencyMatcher", "url": "/api/dependencymatcher" } ] }, { diff --git a/website/src/components/link.js b/website/src/components/link.js index 4c4aa9492..a477cf1d2 100644 --- a/website/src/components/link.js +++ b/website/src/components/link.js @@ -87,6 +87,17 @@ const Link = ({ ) } +export const OptionalLink = ({ to, href, children, ...props }) => { + const dest = to || href + return dest ? ( + + {children} + + ) : ( + children || null + ) +} + Link.defaultProps = { hidden: false, hideIcon: false, diff --git a/website/src/components/title.js b/website/src/components/title.js index 361d24a30..94592cf51 100644 --- a/website/src/components/title.js +++ b/website/src/components/title.js @@ -4,43 +4,105 @@ import classNames from 'classnames' import Button from './button' import Tag from './tag' -import { H1 } from './typography' +import { OptionalLink } from './link' +import { InlineCode } from './code' +import { H1, Label, InlineList, Help } from './typography' +import Icon from './icon' import classes from '../styles/title.module.sass' -const Title = ({ id, title, tag, version, teaser, source, image, children, ...props }) => ( -
- {(image || source) && ( -
- {source && ( - - )} - - {image && ( -
- -
- )} -
+const MetaItem = ({ label, url, children, help }) => ( + + + {children} + {help && ( + <> + {' '} + {help} + )} -

- {title} -

- {tag && {tag}} - {version && ( - - {version} - - )} - - {teaser &&
{teaser}
} - - {children} -
+ ) +const Title = ({ + id, + title, + tag, + version, + teaser, + source, + image, + apiDetails, + children, + ...props +}) => { + const hasApiDetails = Object.values(apiDetails).some(v => v) + const metaIconProps = { className: classes.metaIcon, width: 18 } + return ( +
+ {(image || source) && ( +
+ {source && ( + + )} + + {image && ( +
+ +
+ )} +
+ )} +

+ {title} +

+ {(tag || version) && ( +
+ {tag && {tag}} + {version && ( + + {version} + + )} +
+ )} + + {hasApiDetails && ( + + {apiDetails.stringName && ( + + {apiDetails.stringName} + + )} + {apiDetails.baseClass && ( + + {apiDetails.baseClass.title} + + )} + {apiDetails.trainable != null && ( + + + {apiDetails.trainable ? ( + + ) : ( + + )} + + + )} + + )} + {teaser &&
{teaser}
} + {children} +
+ ) +} + Title.propTypes = { title: PropTypes.string, tag: PropTypes.string, diff --git a/website/src/components/typography.js b/website/src/components/typography.js index eabb3a5ae..41464473f 100644 --- a/website/src/components/typography.js +++ b/website/src/components/typography.js @@ -59,9 +59,9 @@ export const InlineList = ({ Component = 'p', gutterBottom = true, className, ch return {children} } -export const Help = ({ children }) => ( +export const Help = ({ children, size = 16 }) => ( - + ) diff --git a/website/src/styles/table.module.sass b/website/src/styles/table.module.sass index 4477c43f1..6306e2a15 100644 --- a/website/src/styles/table.module.sass +++ b/website/src/styles/table.module.sass @@ -25,6 +25,9 @@ font-family: var(--font-secondary) color: var(--color-theme) + & > td:nth-child(2) a + border: 0 + .td padding: 1rem @@ -80,6 +83,13 @@ white-space: nowrap z-index: 5 + &:first-child // directly after thead/tr + border-bottom: 0 + + td em + top: -7px + + // Responsive table // Shadows adapted from "CSS only Responsive Tables" by David Bushell // http://codepen.io/dbushell/pen/wGaamR diff --git a/website/src/styles/title.module.sass b/website/src/styles/title.module.sass index a29c3c94d..77c0b440a 100644 --- a/website/src/styles/title.module.sass +++ b/website/src/styles/title.module.sass @@ -34,3 +34,16 @@ .corner float: right + +.label + display: inline + margin-right: var(--spacing-xs) + +.meta-icon + position: relative + top: -2px + +.tags + display: inline-block + position: relative + top: 0.5rem diff --git a/website/src/templates/docs.js b/website/src/templates/docs.js index 89bf66605..cc78f339c 100644 --- a/website/src/templates/docs.js +++ b/website/src/templates/docs.js @@ -30,6 +30,7 @@ const Docs = ({ pageContext, children }) => ( menu, theme, version, + apiDetails, } = pageContext const { sidebars = [], modelsRepo, languages, nightly } = site.siteMetadata const isModels = section === 'models' @@ -102,6 +103,7 @@ const Docs = ({ pageContext, children }) => ( tag={tag} version={version} id="_title" + apiDetails={apiDetails} /> {children} {subFooter}