spaCy/spacy/tests/test_tok2vec.py

172 lines
5.4 KiB
Python
Raw Normal View History

import pytest
Default settings to configurations (#4995) * fix grad_clip naming * cleaning up pretrained_vectors out of cfg * further refactoring Model init's * move Model building out of pipes * further refactor to require a model config when creating a pipe * small fixes * making cfg in nn_parser more consistent * fixing nr_class for parser * fixing nn_parser's nO * fix printing of loss * architectures in own file per type, consistent naming * convenience methods default_tagger_config and default_tok2vec_config * let create_pipe access default config if available for that component * default_parser_config * move defaults to separate folder * allow reading nlp from package or dir with argument 'name' * architecture spacy.VocabVectors.v1 to read static vectors from file * cleanup * default configs for nel, textcat, morphologizer, tensorizer * fix imports * fixing unit tests * fixes and clean up * fixing defaults, nO, fix unit tests * restore parser IO * fix IO * 'fix' serialization test * add *.cfg to manifest * fix example configs with additional arguments * replace Morpohologizer with Tagger * add IO bit when testing overfitting of tagger (currently failing) * fix IO - don't initialize when reading from disk * expand overfitting tests to also check IO goes OK * remove dropout from HashEmbed to fix Tagger performance * add defaults for sentrec * update thinc * always pass a Model instance to a Pipe * fix piped_added statement * remove obsolete W029 * remove obsolete errors * restore byte checking tests (work again) * clean up test * further test cleanup * convert from config to Model in create_pipe * bring back error when component is not initialized * cleanup * remove calls for nlp2.begin_training * use thinc.api in imports * allow setting charembed's nM and nC * fix for hardcoded nM/nC + unit test * formatting fixes * trigger build
2020-02-27 20:42:27 +03:00
from spacy.ml.models.tok2vec import build_Tok2Vec_model
2020-07-29 14:47:37 +03:00
from spacy.ml.models.tok2vec import MultiHashEmbed, CharacterEmbed
from spacy.ml.models.tok2vec import MishWindowEncoder, MaxoutWindowEncoder
from spacy.pipeline.tok2vec import Tok2Vec, Tok2VecListener
from spacy.vocab import Vocab
from spacy.tokens import Doc
from spacy.training import Example
from spacy import util
from spacy.lang.en import English
from .util import get_batch
from thinc.api import Config
from numpy.testing import assert_equal
def test_empty_doc():
width = 128
embed_size = 2000
vocab = Vocab()
doc = Doc(vocab, words=[])
tok2vec = build_Tok2Vec_model(
2020-07-29 00:06:46 +03:00
MultiHashEmbed(
width=width,
rows=embed_size,
also_use_static_vectors=False,
2020-08-05 17:00:59 +03:00
also_embed_subwords=True,
2020-07-29 00:06:46 +03:00
),
2020-08-05 17:00:59 +03:00
MaxoutWindowEncoder(width=width, depth=4, window_size=1, maxout_pieces=3),
)
tok2vec.initialize()
vectors, backprop = tok2vec.begin_update([doc])
assert len(vectors) == 1
assert vectors[0].shape == (0, width)
@pytest.mark.parametrize(
"batch_size,width,embed_size", [[1, 128, 2000], [2, 128, 2000], [3, 8, 63]]
)
def test_tok2vec_batch_sizes(batch_size, width, embed_size):
batch = get_batch(batch_size)
Default settings to configurations (#4995) * fix grad_clip naming * cleaning up pretrained_vectors out of cfg * further refactoring Model init's * move Model building out of pipes * further refactor to require a model config when creating a pipe * small fixes * making cfg in nn_parser more consistent * fixing nr_class for parser * fixing nn_parser's nO * fix printing of loss * architectures in own file per type, consistent naming * convenience methods default_tagger_config and default_tok2vec_config * let create_pipe access default config if available for that component * default_parser_config * move defaults to separate folder * allow reading nlp from package or dir with argument 'name' * architecture spacy.VocabVectors.v1 to read static vectors from file * cleanup * default configs for nel, textcat, morphologizer, tensorizer * fix imports * fixing unit tests * fixes and clean up * fixing defaults, nO, fix unit tests * restore parser IO * fix IO * 'fix' serialization test * add *.cfg to manifest * fix example configs with additional arguments * replace Morpohologizer with Tagger * add IO bit when testing overfitting of tagger (currently failing) * fix IO - don't initialize when reading from disk * expand overfitting tests to also check IO goes OK * remove dropout from HashEmbed to fix Tagger performance * add defaults for sentrec * update thinc * always pass a Model instance to a Pipe * fix piped_added statement * remove obsolete W029 * remove obsolete errors * restore byte checking tests (work again) * clean up test * further test cleanup * convert from config to Model in create_pipe * bring back error when component is not initialized * cleanup * remove calls for nlp2.begin_training * use thinc.api in imports * allow setting charembed's nM and nC * fix for hardcoded nM/nC + unit test * formatting fixes * trigger build
2020-02-27 20:42:27 +03:00
tok2vec = build_Tok2Vec_model(
2020-07-29 00:06:46 +03:00
MultiHashEmbed(
width=width,
rows=embed_size,
also_use_static_vectors=False,
2020-08-05 17:00:59 +03:00
also_embed_subwords=True,
2020-07-29 00:06:46 +03:00
),
MaxoutWindowEncoder(width=width, depth=4, window_size=1, maxout_pieces=3),
Default settings to configurations (#4995) * fix grad_clip naming * cleaning up pretrained_vectors out of cfg * further refactoring Model init's * move Model building out of pipes * further refactor to require a model config when creating a pipe * small fixes * making cfg in nn_parser more consistent * fixing nr_class for parser * fixing nn_parser's nO * fix printing of loss * architectures in own file per type, consistent naming * convenience methods default_tagger_config and default_tok2vec_config * let create_pipe access default config if available for that component * default_parser_config * move defaults to separate folder * allow reading nlp from package or dir with argument 'name' * architecture spacy.VocabVectors.v1 to read static vectors from file * cleanup * default configs for nel, textcat, morphologizer, tensorizer * fix imports * fixing unit tests * fixes and clean up * fixing defaults, nO, fix unit tests * restore parser IO * fix IO * 'fix' serialization test * add *.cfg to manifest * fix example configs with additional arguments * replace Morpohologizer with Tagger * add IO bit when testing overfitting of tagger (currently failing) * fix IO - don't initialize when reading from disk * expand overfitting tests to also check IO goes OK * remove dropout from HashEmbed to fix Tagger performance * add defaults for sentrec * update thinc * always pass a Model instance to a Pipe * fix piped_added statement * remove obsolete W029 * remove obsolete errors * restore byte checking tests (work again) * clean up test * further test cleanup * convert from config to Model in create_pipe * bring back error when component is not initialized * cleanup * remove calls for nlp2.begin_training * use thinc.api in imports * allow setting charembed's nM and nC * fix for hardcoded nM/nC + unit test * formatting fixes * trigger build
2020-02-27 20:42:27 +03:00
)
Update spaCy for thinc 8.0.0 (#4920) * Add load_from_config function * Add train_from_config script * Merge configs and expose via spacy.config * Fix script * Suggest create_evaluation_callback * Hard-code for NER * Fix errors * Register command * Add TODO * Update train-from-config todos * Fix imports * Allow delayed setting of parser model nr_class * Get train-from-config working * Tidy up and fix scores and printing * Hide traceback if cancelled * Fix weighted score formatting * Fix score formatting * Make output_path optional * Add Tok2Vec component * Tidy up and add tok2vec_tensors * Add option to copy docs in nlp.update * Copy docs in nlp.update * Adjust nlp.update() for set_annotations * Don't shuffle pipes in nlp.update, decruft * Support set_annotations arg in component update * Support set_annotations in parser update * Add get_gradients method * Add get_gradients to parser * Update errors.py * Fix problems caused by merge * Add _link_components method in nlp * Add concept of 'listeners' and ControlledModel * Support optional attributes arg in ControlledModel * Try having tok2vec component in pipeline * Fix tok2vec component * Fix config * Fix tok2vec * Update for Example * Update for Example * Update config * Add eg2doc util * Update and add schemas/types * Update schemas * Fix nlp.update * Fix tagger * Remove hacks from train-from-config * Remove hard-coded config str * Calculate loss in tok2vec component * Tidy up and use function signatures instead of models * Support union types for registry models * Minor cleaning in Language.update * Make ControlledModel specifically Tok2VecListener * Fix train_from_config * Fix tok2vec * Tidy up * Add function for bilstm tok2vec * Fix type * Fix syntax * Fix pytorch optimizer * Add example configs * Update for thinc describe changes * Update for Thinc changes * Update for dropout/sgd changes * Update for dropout/sgd changes * Unhack gradient update * Work on refactoring _ml * Remove _ml.py module * WIP upgrade cli scripts for thinc * Move some _ml stuff to util * Import link_vectors from util * Update train_from_config * Import from util * Import from util * Temporarily add ml.component_models module * Move ml methods * Move typedefs * Update load vectors * Update gitignore * Move imports * Add PrecomputableAffine * Fix imports * Fix imports * Fix imports * Fix missing imports * Update CLI scripts * Update spacy.language * Add stubs for building the models * Update model definition * Update create_default_optimizer * Fix import * Fix comment * Update imports in tests * Update imports in spacy.cli * Fix import * fix obsolete thinc imports * update srsly pin * from thinc to ml_datasets for example data such as imdb * update ml_datasets pin * using STATE.vectors * small fix * fix Sentencizer.pipe * black formatting * rename Affine to Linear as in thinc * set validate explicitely to True * rename with_square_sequences to with_list2padded * rename with_flatten to with_list2array * chaining layernorm * small fixes * revert Optimizer import * build_nel_encoder with new thinc style * fixes using model's get and set methods * Tok2Vec in component models, various fixes * fix up legacy tok2vec code * add model initialize calls * add in build_tagger_model * small fixes * setting model dims * fixes for ParserModel * various small fixes * initialize thinc Models * fixes * consistent naming of window_size * fixes, removing set_dropout * work around Iterable issue * remove legacy tok2vec * util fix * fix forward function of tok2vec listener * more fixes * trying to fix PrecomputableAffine (not succesful yet) * alloc instead of allocate * add morphologizer * rename residual * rename fixes * Fix predict function * Update parser and parser model * fixing few more tests * Fix precomputable affine * Update component model * Update parser model * Move backprop padding to own function, for test * Update test * Fix p. affine * Update NEL * build_bow_text_classifier and extract_ngrams * Fix parser init * Fix test add label * add build_simple_cnn_text_classifier * Fix parser init * Set gpu off by default in example * Fix tok2vec listener * Fix parser model * Small fixes * small fix for PyTorchLSTM parameters * revert my_compounding hack (iterable fixed now) * fix biLSTM * Fix uniqued * PyTorchRNNWrapper fix * small fixes * use helper function to calculate cosine loss * small fixes for build_simple_cnn_text_classifier * putting dropout default at 0.0 to ensure the layer gets built * using thinc util's set_dropout_rate * moving layer normalization inside of maxout definition to optimize dropout * temp debugging in NEL * fixed NEL model by using init defaults ! * fixing after set_dropout_rate refactor * proper fix * fix test_update_doc after refactoring optimizers in thinc * Add CharacterEmbed layer * Construct tagger Model * Add missing import * Remove unused stuff * Work on textcat * fix test (again :)) after optimizer refactor * fixes to allow reading Tagger from_disk without overwriting dimensions * don't build the tok2vec prematuraly * fix CharachterEmbed init * CharacterEmbed fixes * Fix CharacterEmbed architecture * fix imports * renames from latest thinc update * one more rename * add initialize calls where appropriate * fix parser initialization * Update Thinc version * Fix errors, auto-format and tidy up imports * Fix validation * fix if bias is cupy array * revert for now * ensure it's a numpy array before running bp in ParserStepModel * no reason to call require_gpu twice * use CupyOps.to_numpy instead of cupy directly * fix initialize of ParserModel * remove unnecessary import * fixes for CosineDistance * fix device renaming * use refactored loss functions (Thinc PR 251) * overfitting test for tagger * experimental settings for the tagger: avoid zero-init and subword normalization * clean up tagger overfitting test * use previous default value for nP * remove toy config * bringing layernorm back (had a bug - fixed in thinc) * revert setting nP explicitly * remove setting default in constructor * restore values as they used to be * add overfitting test for NER * add overfitting test for dep parser * add overfitting test for textcat * fixing init for linear (previously affine) * larger eps window for textcat * ensure doc is not None * Require newer thinc * Make float check vaguer * Slop the textcat overfit test more * Fix textcat test * Fix exclusive classes for textcat * fix after renaming of alloc methods * fixing renames and mandatory arguments (staticvectors WIP) * upgrade to thinc==8.0.0.dev3 * refer to vocab.vectors directly instead of its name * rename alpha to learn_rate * adding hashembed and staticvectors dropout * upgrade to thinc 8.0.0.dev4 * add name back to avoid warning W020 * thinc dev4 * update srsly * using thinc 8.0.0a0 ! Co-authored-by: Matthew Honnibal <honnibal+gh@gmail.com> Co-authored-by: Ines Montani <ines@ines.io>
2020-01-29 19:06:46 +03:00
tok2vec.initialize()
vectors, backprop = tok2vec.begin_update(batch)
assert len(vectors) == len(batch)
for doc_vec, doc in zip(vectors, batch):
assert doc_vec.shape == (len(doc), width)
Default settings to configurations (#4995) * fix grad_clip naming * cleaning up pretrained_vectors out of cfg * further refactoring Model init's * move Model building out of pipes * further refactor to require a model config when creating a pipe * small fixes * making cfg in nn_parser more consistent * fixing nr_class for parser * fixing nn_parser's nO * fix printing of loss * architectures in own file per type, consistent naming * convenience methods default_tagger_config and default_tok2vec_config * let create_pipe access default config if available for that component * default_parser_config * move defaults to separate folder * allow reading nlp from package or dir with argument 'name' * architecture spacy.VocabVectors.v1 to read static vectors from file * cleanup * default configs for nel, textcat, morphologizer, tensorizer * fix imports * fixing unit tests * fixes and clean up * fixing defaults, nO, fix unit tests * restore parser IO * fix IO * 'fix' serialization test * add *.cfg to manifest * fix example configs with additional arguments * replace Morpohologizer with Tagger * add IO bit when testing overfitting of tagger (currently failing) * fix IO - don't initialize when reading from disk * expand overfitting tests to also check IO goes OK * remove dropout from HashEmbed to fix Tagger performance * add defaults for sentrec * update thinc * always pass a Model instance to a Pipe * fix piped_added statement * remove obsolete W029 * remove obsolete errors * restore byte checking tests (work again) * clean up test * further test cleanup * convert from config to Model in create_pipe * bring back error when component is not initialized * cleanup * remove calls for nlp2.begin_training * use thinc.api in imports * allow setting charembed's nM and nC * fix for hardcoded nM/nC + unit test * formatting fixes * trigger build
2020-02-27 20:42:27 +03:00
# fmt: off
@pytest.mark.parametrize(
2020-07-29 14:47:37 +03:00
"width,embed_arch,embed_config,encode_arch,encode_config",
[
2020-07-29 14:47:37 +03:00
(8, MultiHashEmbed, {"rows": 100, "also_embed_subwords": True, "also_use_static_vectors": False}, MaxoutWindowEncoder, {"window_size": 1, "maxout_pieces": 3, "depth": 2}),
(8, MultiHashEmbed, {"rows": 100, "also_embed_subwords": True, "also_use_static_vectors": False}, MishWindowEncoder, {"window_size": 1, "depth": 6}),
(8, CharacterEmbed, {"rows": 100, "nM": 64, "nC": 8}, MaxoutWindowEncoder, {"window_size": 1, "maxout_pieces": 3, "depth": 3}),
(8, CharacterEmbed, {"rows": 100, "nM": 16, "nC": 2}, MishWindowEncoder, {"window_size": 1, "depth": 3}),
],
)
Default settings to configurations (#4995) * fix grad_clip naming * cleaning up pretrained_vectors out of cfg * further refactoring Model init's * move Model building out of pipes * further refactor to require a model config when creating a pipe * small fixes * making cfg in nn_parser more consistent * fixing nr_class for parser * fixing nn_parser's nO * fix printing of loss * architectures in own file per type, consistent naming * convenience methods default_tagger_config and default_tok2vec_config * let create_pipe access default config if available for that component * default_parser_config * move defaults to separate folder * allow reading nlp from package or dir with argument 'name' * architecture spacy.VocabVectors.v1 to read static vectors from file * cleanup * default configs for nel, textcat, morphologizer, tensorizer * fix imports * fixing unit tests * fixes and clean up * fixing defaults, nO, fix unit tests * restore parser IO * fix IO * 'fix' serialization test * add *.cfg to manifest * fix example configs with additional arguments * replace Morpohologizer with Tagger * add IO bit when testing overfitting of tagger (currently failing) * fix IO - don't initialize when reading from disk * expand overfitting tests to also check IO goes OK * remove dropout from HashEmbed to fix Tagger performance * add defaults for sentrec * update thinc * always pass a Model instance to a Pipe * fix piped_added statement * remove obsolete W029 * remove obsolete errors * restore byte checking tests (work again) * clean up test * further test cleanup * convert from config to Model in create_pipe * bring back error when component is not initialized * cleanup * remove calls for nlp2.begin_training * use thinc.api in imports * allow setting charembed's nM and nC * fix for hardcoded nM/nC + unit test * formatting fixes * trigger build
2020-02-27 20:42:27 +03:00
# fmt: on
2020-07-29 14:47:37 +03:00
def test_tok2vec_configs(width, embed_arch, embed_config, encode_arch, encode_config):
embed_config["width"] = width
encode_config["width"] = width
docs = get_batch(3)
2020-07-29 14:47:37 +03:00
tok2vec = build_Tok2Vec_model(
embed_arch(**embed_config),
encode_arch(**encode_config)
)
tok2vec.initialize(docs)
vectors, backprop = tok2vec.begin_update(docs)
assert len(vectors) == len(docs)
2020-07-29 14:47:37 +03:00
assert vectors[0].shape == (len(docs[0]), width)
backprop(vectors)
def test_init_tok2vec():
# Simple test to initialize the default tok2vec
nlp = English()
tok2vec = nlp.add_pipe("tok2vec")
assert tok2vec.listeners == []
nlp.begin_training()
assert tok2vec.model.get_dim("nO")
cfg_string = """
[nlp]
lang = "en"
pipeline = ["tok2vec","tagger"]
[components]
[components.tagger]
factory = "tagger"
[components.tagger.model]
@architectures = "spacy.Tagger.v1"
nO = null
[components.tagger.model.tok2vec]
@architectures = "spacy.Tok2VecListener.v1"
width = ${components.tok2vec.model.encode.width}
[components.tok2vec]
factory = "tok2vec"
[components.tok2vec.model]
@architectures = "spacy.Tok2Vec.v1"
[components.tok2vec.model.embed]
@architectures = "spacy.MultiHashEmbed.v1"
width = ${components.tok2vec.model.encode.width}
rows = 2000
also_embed_subwords = true
also_use_static_vectors = false
[components.tok2vec.model.encode]
@architectures = "spacy.MaxoutWindowEncoder.v1"
width = 96
depth = 4
window_size = 1
maxout_pieces = 3
"""
TRAIN_DATA = [
("I like green eggs", {"tags": ["N", "V", "J", "N"]}),
("Eat blue ham", {"tags": ["V", "J", "N"]}),
]
2020-09-04 14:42:33 +03:00
def test_tok2vec_listener():
orig_config = Config().from_str(cfg_string)
nlp, config = util.load_model_from_config(orig_config, auto_fill=True, validate=True)
assert nlp.pipe_names == ["tok2vec", "tagger"]
tagger = nlp.get_pipe("tagger")
tok2vec = nlp.get_pipe("tok2vec")
tagger_tok2vec = tagger.model.get_ref("tok2vec")
assert isinstance(tok2vec, Tok2Vec)
assert isinstance(tagger_tok2vec, Tok2VecListener)
train_examples = []
for t in TRAIN_DATA:
train_examples.append(Example.from_dict(nlp.make_doc(t[0]), t[1]))
for tag in t[1]["tags"]:
tagger.add_label(tag)
# Check that the Tok2Vec component finds it listeners
assert tok2vec.listeners == []
optimizer = nlp.begin_training(lambda: train_examples)
assert tok2vec.listeners == [tagger_tok2vec]
for i in range(5):
losses = {}
nlp.update(train_examples, sgd=optimizer, losses=losses)
doc = nlp("Running the pipeline as a whole.")
doc_tensor = tagger_tok2vec.predict([doc])[0]
assert_equal(doc.tensor, doc_tensor)
# TODO: should this warn or error?
nlp.select_pipes(disable="tok2vec")
assert nlp.pipe_names == ["tagger"]
nlp("Running the pipeline with the Tok2Vec component disabled.")