Store activations in `Doc`s when `save_activations` is enabled #11002

danieldk · 2022-06-22T08:04:11Z

Description

This change adds the new activations attribute to Doc. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users.

As an example, this change modifies the tagger and senter pipes to add an store_activations option. When this option is enabled, the probabilities and guesses are stored in set_annotations.

Edit: out of draft.

Types of change

Proposal.

Checklist

I confirm that I have the right to submit this contribution under the project's MIT license.
I ran the tests, and all new and existing tests passed.
My changes don't require a change to the documentation, or if they do, I've added all required information.

This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`.

When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored

So that they, like the other tests, also check that we get no activations if no activations were requested.

adrianeboyd · 2022-06-27T13:20:38Z

Adding a field to Doc is breaking, so I think this would need to wait until v4?

danieldk · 2022-06-27T13:21:52Z

Adding a field to Doc is breaking, so I think this would need to wait until v4?

Yeah. I was going to rebase this to v4, but we first need to update v4 to avoid that all the commits end up in this PR:

#11034

danieldk · 2022-06-27T14:49:19Z

Something went awry with the merge, will fix this later.

danieldk · 2022-06-27T17:38:42Z

Something went awry with the merge, will fix this later.

Order restored :).

spacy/pipeline/edit_tree_lemmatizer.py

Before this change, we'd use the senter pipe directly. However, this did not work with the transformer model without modifications (because it clears tensors after backprop). By using the functionality proposed in explosion/spaCy#11002 we can use the activations that are stored by the senter pipe in `Doc`.

svlandeg

Nice idea!

Left a first batch of some comments.

spacy/errors.py

spacy/pipeline/edit_tree_lemmatizer.py

svlandeg

We'll still need to release explosion/thinc#739, bump the Thinc requirement and have the CI go green, but in principle this seems good to me :-)

Will be a great feature to have!

Before this change, we'd use the senter pipe directly. However, this did not work with the transformer model without modifications (because it clears tensors after backprop). By using the functionality proposed in explosion/spaCy#11002 we can use the activations that are stored by the senter pipe in `Doc`.

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

@svlandeg

…sion#11002) * Store activations in Doc when `store_activations` is enabled This change adds the new `activations` attribute to `Doc`. This attribute can be used by trainable pipes to store their activations, probabilities, and guesses for downstream users. As an example, this change modifies the `tagger` and `senter` pipes to add an `store_activations` option. When this option is enabled, the probabilities and guesses are stored in `set_annotations`. * Change type of `store_activations` to `Union[bool, List[str]]` When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored * Formatting fixes in Tagger * Support store_activations in spancat and morphologizer * Make Doc.activations type visible to MyPy * textcat/textcat_multilabel: add store_activations option * trainable_lemmatizer/entity_linker: add store_activations option * parser/ner: do not currently support returning activations * Extend tagger and senter tests So that they, like the other tests, also check that we get no activations if no activations were requested. * Document `Doc.activations` and `store_activations` in the relevant pipes * Start errors/warnings at higher numbers to avoid merge conflicts Between the master and v4 branches. * Add `store_activations` to docstrings. * Replace store_activations setter by set_store_activations method Setters that take a different type than what the getter returns are still problematic for MyPy. Replace the setter by a method, so that type inference works everywhere. * Use dict comprehension suggested by @svlandeg * Revert "Use dict comprehension suggested by @svlandeg" This reverts commit 6e7b958. * EntityLinker: add type annotations to _add_activations * _store_activations: make kwarg-only, remove doc_scores_lens arg * set_annotations: add type annotations * Apply suggestions from code review Co-authored-by: Sofie Van Landeghem <[email protected]> * TextCat.predict: return dict * Make the `TrainablePipe.store_activations` property a bool This means that we can also bring back `store_activations` setter. * Remove `TrainablePipe.activations` We do not need to enumerate the activations anymore since `store_activations` is `bool`. * Add type annotations for activations in predict/set_annotations * Rename `TrainablePipe.store_activations` to `save_activations` * Error E1400 is not used anymore This error was used when activations were still `Union[bool, List[str]]`. * Change wording in API docs after store -> save change * docs: tag (save_)activations as new in spaCy 4.0 * Fix copied line in morphologizer activations test * Don't train in any test_save_activations test * Rename activations - "probs" -> "probabilities" - "guesses" -> "label_ids", except in the edit tree lemmatizer, where "guesses" -> "tree_ids". * Remove unused W400 warning. This warning was used when we still allowed the user to specify which activations to save. * Formatting fixes Co-authored-by: Sofie Van Landeghem <[email protected]> * Replace "kb_ids" by a constant * spancat: replace a cast by an assertion * Fix EOF spacing * Fix comments in test_save_activations tests * Do not set RNG seed in activation saving tests * Revert "spancat: replace a cast by an assertion" This reverts commit 0bd5730. Co-authored-by: Sofie Van Landeghem <[email protected]>

danieldk changed the title ~~Store activations in Doc when store_activations is enabled~~ Store activations in Docs when store_activations is enabled Jun 22, 2022

danieldk added 4 commits June 22, 2022 14:28

Change type of store_activations to Union[bool, List[str]]

c3da32b

When the value is: - A bool: all activations are stored when set to `True`. - A List[str]: the activations named in the list are stored

Formatting fixes in Tagger

789a447

Support store_activations in spancat and morphologizer

acf47e8

Make Doc.activations type visible to MyPy

42526a6

svlandeg added enhancement Feature requests and improvements feat / pipeline Feature: Processing pipeline and components labels Jun 22, 2022

danieldk added 6 commits June 22, 2022 20:40

textcat/textcat_multilabel: add store_activations option

8772b9c

trainable_lemmatizer/entity_linker: add store_activations option

1c9be0d

parser/ner: do not currently support returning activations

009a960

Extend tagger and senter tests

c8a12c5

So that they, like the other tests, also check that we get no activations if no activations were requested.

Merge remote-tracking branch 'upstream/master' into store-activations

508b96f

Document Doc.activations and store_activations in the relevant pipes

3b13f17

danieldk marked this pull request as ready for review June 27, 2022 13:16

danieldk changed the base branch from master to v4 June 27, 2022 14:45

danieldk force-pushed the store-activations branch from 85248bf to 944b0e3 Compare June 27, 2022 14:48

danieldk marked this pull request as draft June 27, 2022 14:49

danieldk force-pushed the v4 branch from 3f76bc1 to 7f3842f Compare June 27, 2022 15:38

Merge remote-tracking branch 'upstream/v4' into store-activations

df513be

danieldk force-pushed the store-activations branch from 944b0e3 to df513be Compare June 27, 2022 17:21

danieldk marked this pull request as ready for review June 27, 2022 17:38

shadeMe reviewed Jun 28, 2022

View reviewed changes

spacy/pipeline/edit_tree_lemmatizer.py Outdated Show resolved Hide resolved

danieldk mentioned this pull request Jun 28, 2022

Support lazy, recursive sentence splitting explosion/spacy-experimental#7

Closed

svlandeg reviewed Jun 29, 2022

View reviewed changes

spacy/errors.py Outdated Show resolved Hide resolved

spacy/pipeline/edit_tree_lemmatizer.py Show resolved Hide resolved

spacy/pipeline/edit_tree_lemmatizer.py Outdated Show resolved Hide resolved

spacy/pipeline/edit_tree_lemmatizer.py Outdated Show resolved Hide resolved

svlandeg approved these changes Sep 8, 2022

View reviewed changes

svlandeg merged commit efdbb72 into explosion:v4 Sep 13, 2022

danieldk deleted the store-activations branch September 13, 2022 08:01

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Store activations in `Doc`s when `save_activations` is enabled #11002

Store activations in `Doc`s when `save_activations` is enabled #11002

danieldk commented Jun 22, 2022 •

edited

Loading

adrianeboyd commented Jun 27, 2022

danieldk commented Jun 27, 2022

danieldk commented Jun 27, 2022

danieldk commented Jun 27, 2022

svlandeg left a comment

svlandeg left a comment

Store activations in Docs when save_activations is enabled #11002

Store activations in Docs when save_activations is enabled #11002

Conversation

danieldk commented Jun 22, 2022 • edited Loading

Description

Types of change

Checklist

adrianeboyd commented Jun 27, 2022

danieldk commented Jun 27, 2022

danieldk commented Jun 27, 2022

danieldk commented Jun 27, 2022

svlandeg left a comment

Choose a reason for hiding this comment

svlandeg left a comment

Choose a reason for hiding this comment

Store activations in `Doc`s when `save_activations` is enabled #11002

Store activations in `Doc`s when `save_activations` is enabled #11002

danieldk commented Jun 22, 2022 •

edited

Loading