diff --git a/examples/training/train_ner.py b/examples/training/train_ner.py index bcc087d07..e50e36756 100644 --- a/examples/training/train_ner.py +++ b/examples/training/train_ner.py @@ -3,66 +3,26 @@ import json import pathlib import random -import spacy -from spacy.pipeline import EntityRecognizer -from spacy.gold import GoldParse -from spacy.tagger import Tagger - - -try: - unicode -except: - unicode = str +import spacy.lang.en +from spacy.gold import GoldParse, biluo_tags_from_offsets -def train_ner(nlp, train_data, entity_types): - # Add new words to vocab. - for raw_text, _ in train_data: - doc = nlp.make_doc(raw_text) - for word in doc: - _ = nlp.vocab[word.orth] - - # Train NER. - ner = EntityRecognizer(nlp.vocab, entity_types=entity_types) - for itn in range(5): - random.shuffle(train_data) - for raw_text, entity_offsets in train_data: - doc = nlp.make_doc(raw_text) - gold = GoldParse(doc, entities=entity_offsets) - ner.update(doc, gold) - return ner - -def save_model(ner, model_dir): - model_dir = pathlib.Path(model_dir) - if not model_dir.exists(): - model_dir.mkdir() - assert model_dir.is_dir() - - with (model_dir / 'config.json').open('wb') as file_: - data = json.dumps(ner.cfg) - if isinstance(data, unicode): - data = data.encode('utf8') - file_.write(data) - ner.model.dump(str(model_dir / 'model')) - if not (model_dir / 'vocab').exists(): - (model_dir / 'vocab').mkdir() - ner.vocab.dump(str(model_dir / 'vocab' / 'lexemes.bin')) - with (model_dir / 'vocab' / 'strings.json').open('w', encoding='utf8') as file_: - ner.vocab.strings.dump(file_) +def reformat_train_data(tokenizer, examples): + """Reformat data to match JSON format""" + output = [] + for i, (text, entity_offsets) in enumerate(examples): + doc = tokenizer(text) + ner_tags = biluo_tags_from_offsets(tokenizer(text), entity_offsets) + words = [w.text for w in doc] + tags = ['-'] * len(doc) + heads = [0] * len(doc) + deps = [''] * len(doc) + sentence = (range(len(doc)), words, tags, heads, deps, ner_tags) + output.append((text, [(sentence, [])])) + return output def main(model_dir=None): - nlp = spacy.load('en', parser=False, entity=False, add_vectors=False) - - # v1.1.2 onwards - if nlp.tagger is None: - print('---- WARNING ----') - print('Data directory not found') - print('please run: `python -m spacy.en.download --force all` for better performance') - print('Using feature templates for tagging') - print('-----------------') - nlp.tagger = Tagger(nlp.vocab, features=Tagger.feature_templates) - train_data = [ ( 'Who is Shaka Khan?', @@ -74,23 +34,35 @@ def main(model_dir=None): (len('I like London and '), len('I like London and Berlin'), 'LOC')] ) ] - ner = train_ner(nlp, train_data, ['PERSON', 'LOC']) - - doc = nlp.make_doc('Who is Shaka Khan?') - nlp.tagger(doc) - ner(doc) - for word in doc: - print(word.text, word.orth, word.lower, word.tag_, word.ent_type_, word.ent_iob) - - if model_dir is not None: - save_model(ner, model_dir) - - - - + nlp = spacy.lang.en.English(pipeline=['tensorizer', 'ner']) + get_data = lambda: reformat_train_data(nlp.tokenizer, train_data) + optimizer = nlp.begin_training(get_data) + for itn in range(100): + random.shuffle(train_data) + losses = {} + for raw_text, entity_offsets in train_data: + doc = nlp.make_doc(raw_text) + gold = GoldParse(doc, entities=entity_offsets) + nlp.update( + [doc], # Batch of Doc objects + [gold], # Batch of GoldParse objects + drop=0.5, # Dropout -- make it harder to memorise data + sgd=optimizer, # Callable to update weights + losses=losses) + print(losses) + print("Save to", model_dir) + nlp.to_disk(model_dir) + print("Load from", model_dir) + nlp = spacy.lang.en.English(pipeline=['tensorizer', 'ner']) + nlp.from_disk(model_dir) + for raw_text, _ in train_data: + doc = nlp(raw_text) + for word in doc: + print(word.text, word.ent_type_, word.ent_iob_) if __name__ == '__main__': - main('ner') + import plac + plac.call(main) # Who "" 2 # is "" 2 # Shaka "" PERSON 3 diff --git a/spacy/lang/hu/punctuation.py b/spacy/lang/hu/punctuation.py index 27a2912e2..b758e0104 100644 --- a/spacy/lang/hu/punctuation.py +++ b/spacy/lang/hu/punctuation.py @@ -1,18 +1,17 @@ # coding: utf8 from __future__ import unicode_literals -from ..punctuation import TOKENIZER_INFIXES -from ..char_classes import LIST_PUNCT, LIST_ELLIPSES, LIST_QUOTES, CURRENCY +from ..char_classes import LIST_PUNCT, LIST_ELLIPSES, LIST_QUOTES from ..char_classes import QUOTES, UNITS, ALPHA, ALPHA_LOWER, ALPHA_UPPER +LIST_ICONS = [r'[\p{So}--[°]]'] _currency = r'\$|¢|£|€|¥|฿' _quotes = QUOTES.replace("'", '') +_prefixes = ([r'\+'] + LIST_PUNCT + LIST_ELLIPSES + LIST_QUOTES + LIST_ICONS) -_prefixes = ([r'\+'] + LIST_PUNCT + LIST_ELLIPSES + LIST_QUOTES) - -_suffixes = (LIST_PUNCT + LIST_ELLIPSES + LIST_QUOTES + +_suffixes = (LIST_PUNCT + LIST_ELLIPSES + LIST_QUOTES + LIST_ICONS + [r'(?<=[0-9])\+', r'(?<=°[FfCcKk])\.', r'(?<=[0-9])(?:{})'.format(_currency), @@ -20,8 +19,7 @@ _suffixes = (LIST_PUNCT + LIST_ELLIPSES + LIST_QUOTES + r'(?<=[{}{}{}(?:{})])\.'.format(ALPHA_LOWER, r'%²\-\)\]\+', QUOTES, _currency), r'(?<=[{})])-e'.format(ALPHA_LOWER)]) - -_infixes = (LIST_ELLIPSES + +_infixes = (LIST_ELLIPSES + LIST_ICONS + [r'(?<=[{}])\.(?=[{}])'.format(ALPHA_LOWER, ALPHA_UPPER), r'(?<=[{a}]),(?=[{a}])'.format(a=ALPHA), r'(?<=[{a}"])[:<>=](?=[{a}])'.format(a=ALPHA), @@ -29,7 +27,6 @@ _infixes = (LIST_ELLIPSES + r'(?<=[{a}]),(?=[{a}])'.format(a=ALPHA), r'(?<=[{a}])([{q}\)\]\(\[])(?=[\-{a}])'.format(a=ALPHA, q=_quotes)]) - TOKENIZER_PREFIXES = _prefixes TOKENIZER_SUFFIXES = _suffixes TOKENIZER_INFIXES = _infixes diff --git a/spacy/language.py b/spacy/language.py index d9a888507..324d78622 100644 --- a/spacy/language.py +++ b/spacy/language.py @@ -96,6 +96,13 @@ class BaseDefaults(object): factories = { 'make_doc': create_tokenizer, + 'tensorizer': lambda nlp, **cfg: [TokenVectorEncoder(nlp.vocab, **cfg)], + 'tagger': lambda nlp, **cfg: [NeuralTagger(nlp.vocab, **cfg)], + 'parser': lambda nlp, **cfg: [ + NeuralDependencyParser(nlp.vocab, **cfg), + nonproj.deprojectivize], + 'ner': lambda nlp, **cfg: [NeuralEntityRecognizer(nlp.vocab, **cfg)], + # Temporary compatibility -- delete after pivot 'token_vectors': lambda nlp, **cfg: [TokenVectorEncoder(nlp.vocab, **cfg)], 'tags': lambda nlp, **cfg: [NeuralTagger(nlp.vocab, **cfg)], 'dependencies': lambda nlp, **cfg: [ @@ -358,37 +365,35 @@ class Language(object): for doc in docs: yield doc - def to_disk(self, path, disable=[]): + def to_disk(self, path, disable=tuple()): """Save the current state to a directory. If a model is loaded, this will include the model. path (unicode or Path): A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. - disable (list): Nameds of pipeline components to disable and prevent + disable (list): Names of pipeline components to disable and prevent from being saved. EXAMPLE: >>> nlp.to_disk('/path/to/models') """ path = util.ensure_path(path) - with path.open('wb') as file_: - file_.write(self.to_bytes(disable)) - #serializers = { - # 'vocab': lambda p: self.vocab.to_disk(p), - # 'tokenizer': lambda p: self.tokenizer.to_disk(p, vocab=False), - # 'meta.json': lambda p: ujson.dump(p.open('w'), self.meta) - #} - #for proc in self.pipeline: - # if not hasattr(proc, 'name'): - # continue - # if proc.name in disable: - # continue - # if not hasattr(proc, 'to_disk'): - # continue - # serializers[proc.name] = lambda p: proc.to_disk(p, vocab=False) - #util.to_disk(serializers, path) + serializers = OrderedDict(( + ('vocab', lambda p: self.vocab.to_disk(p)), + ('tokenizer', lambda p: self.tokenizer.to_disk(p, vocab=False)), + ('meta.json', lambda p: p.open('w').write(json_dumps(self.meta))) + )) + for proc in self.pipeline: + if not hasattr(proc, 'name'): + continue + if proc.name in disable: + continue + if not hasattr(proc, 'to_disk'): + continue + serializers[proc.name] = lambda p, proc=proc: proc.to_disk(p, vocab=False) + util.to_disk(path, serializers, {p: False for p in disable}) - def from_disk(self, path, disable=[]): + def from_disk(self, path, disable=tuple()): """Loads state from a directory. Modifies the object in place and returns it. If the saved `Language` object contains a model, the model will be loaded. @@ -403,24 +408,21 @@ class Language(object): >>> nlp = Language().from_disk('/path/to/models') """ path = util.ensure_path(path) - with path.open('rb') as file_: - bytes_data = file_.read() - return self.from_bytes(bytes_data, disable) - #deserializers = { - # 'vocab': lambda p: self.vocab.from_disk(p), - # 'tokenizer': lambda p: self.tokenizer.from_disk(p, vocab=False), - # 'meta.json': lambda p: ujson.dump(p.open('w'), self.meta) - #} - #for proc in self.pipeline: - # if not hasattr(proc, 'name'): - # continue - # if proc.name in disable: - # continue - # if not hasattr(proc, 'to_disk'): - # continue - # deserializers[proc.name] = lambda p: proc.from_disk(p, vocab=False) - #util.from_disk(deserializers, path) - #return self + deserializers = OrderedDict(( + ('vocab', lambda p: self.vocab.from_disk(p)), + ('tokenizer', lambda p: self.tokenizer.from_disk(p, vocab=False)), + ('meta.json', lambda p: p.open('w').write(json_dumps(self.meta))) + )) + for proc in self.pipeline: + if not hasattr(proc, 'name'): + continue + if proc.name in disable: + continue + if not hasattr(proc, 'to_disk'): + continue + deserializers[proc.name] = lambda p, proc=proc: proc.from_disk(p, vocab=False) + util.from_disk(path, deserializers, {p: False for p in disable}) + return self def to_bytes(self, disable=[]): """Serialize the current state to a binary string. diff --git a/spacy/pipeline.pyx b/spacy/pipeline.pyx index 963dd2faa..ff7098439 100644 --- a/spacy/pipeline.pyx +++ b/spacy/pipeline.pyx @@ -41,7 +41,7 @@ from .parts_of_speech import X class TokenVectorEncoder(object): """Assign position-sensitive vectors to tokens, using a CNN or RNN.""" - name = 'tok2vec' + name = 'tensorizer' @classmethod def Model(cls, width=128, embed_size=7500, **cfg): @@ -176,17 +176,19 @@ class TokenVectorEncoder(object): return self def to_disk(self, path, **exclude): - serialize = { - 'model': lambda p: p.open('w').write(util.model_to_bytes(self.model)), - 'vocab': lambda p: self.vocab.to_disk(p) - } + serialize = OrderedDict(( + ('model', lambda p: p.open('wb').write(util.model_to_bytes(self.model))), + ('vocab', lambda p: self.vocab.to_disk(p)) + )) util.to_disk(path, serialize, exclude) def from_disk(self, path, **exclude): - deserialize = { - 'model': lambda p: util.model_from_bytes(self.model, p.open('rb').read()), - 'vocab': lambda p: self.vocab.from_disk(p) - } + if self.model is True: + self.model = self.Model() + deserialize = OrderedDict(( + ('model', lambda p: util.model_from_bytes(self.model, p.open('rb').read())), + ('vocab', lambda p: self.vocab.from_disk(p)) + )) util.from_disk(path, deserialize, exclude) return self @@ -315,7 +317,7 @@ class NeuralTagger(object): def to_disk(self, path, **exclude): serialize = { - 'model': lambda p: p.open('w').write(util.model_to_bytes(self.model)), + 'model': lambda p: p.open('wb').write(util.model_to_bytes(self.model)), 'vocab': lambda p: self.vocab.to_disk(p) } util.to_disk(path, serialize, exclude) @@ -420,7 +422,7 @@ cdef class NeuralDependencyParser(NeuralParser): cdef class NeuralEntityRecognizer(NeuralParser): - name = 'entity' + name = 'ner' TransitionSystem = BiluoPushDown nr_feature = 6 diff --git a/spacy/strings.pyx b/spacy/strings.pyx index e255dbb48..2e42b9667 100644 --- a/spacy/strings.pyx +++ b/spacy/strings.pyx @@ -16,6 +16,7 @@ from .symbols import NAMES as SYMBOLS_BY_INT from .typedefs cimport hash_t from . import util +from .compat import json_dumps cpdef hash_t hash_string(unicode string) except 0: @@ -201,7 +202,7 @@ cdef class StringStore: path = util.ensure_path(path) strings = list(self) with path.open('w') as file_: - ujson.dump(strings, file_) + file_.write(json_dumps(strings)) def from_disk(self, path): """Loads state from a directory. Modifies the object in place and diff --git a/spacy/syntax/nn_parser.pyx b/spacy/syntax/nn_parser.pyx index d49e9cdef..d156156d6 100644 --- a/spacy/syntax/nn_parser.pyx +++ b/spacy/syntax/nn_parser.pyx @@ -44,6 +44,7 @@ from .. import util from ..util import get_async, get_cuda_stream from .._ml import zero_init, PrecomputableAffine, PrecomputableMaxouts from .._ml import Tok2Vec, doc2feats, rebatch +from ..compat import json_dumps from . import _parse_features from ._parse_features cimport CONTEXT_SIZE @@ -633,11 +634,13 @@ cdef class Parser: def to_disk(self, path, **exclude): serializers = { - 'model': lambda p: p.open('wb').write( - util.model_to_bytes(self.model)), + 'lower_model': lambda p: p.open('wb').write( + util.model_to_bytes(self.model[0])), + 'upper_model': lambda p: p.open('wb').write( + util.model_to_bytes(self.model[1])), 'vocab': lambda p: self.vocab.to_disk(p), 'moves': lambda p: self.moves.to_disk(p, strings=False), - 'cfg': lambda p: ujson.dumps(p.open('w'), self.cfg) + 'cfg': lambda p: p.open('w').write(json_dumps(self.cfg)) } util.to_disk(path, serializers, exclude) @@ -645,7 +648,7 @@ cdef class Parser: deserializers = { 'vocab': lambda p: self.vocab.from_disk(p), 'moves': lambda p: self.moves.from_disk(p, strings=False), - 'cfg': lambda p: self.cfg.update(ujson.load((path/'cfg.json').open())), + 'cfg': lambda p: self.cfg.update(ujson.load(p.open())), 'model': lambda p: None } util.from_disk(path, deserializers, exclude) @@ -653,7 +656,14 @@ cdef class Parser: path = util.ensure_path(path) if self.model is True: self.model, cfg = self.Model(**self.cfg) - util.model_from_disk(self.model, path / 'model') + else: + cfg = {} + with (path / 'lower_model').open('rb') as file_: + bytes_data = file_.read() + util.model_from_bytes(self.model[0], bytes_data) + with (path / 'upper_model').open('rb') as file_: + bytes_data = file_.read() + util.model_from_bytes(self.model[1], bytes_data) self.cfg.update(cfg) return self diff --git a/spacy/syntax/transition_system.pyx b/spacy/syntax/transition_system.pyx index 42ec7318b..e33a29ac2 100644 --- a/spacy/syntax/transition_system.pyx +++ b/spacy/syntax/transition_system.pyx @@ -157,22 +157,13 @@ cdef class TransitionSystem: return 1 def to_disk(self, path, **exclude): - actions = list(self.move_names) - deserializers = { - 'actions': lambda p: ujson.dump(p.open('w'), actions), - 'strings': lambda p: self.strings.to_disk(p) - } - util.to_disk(path, deserializers, exclude) + with path.open('wb') as file_: + file_.write(self.to_bytes(**exclude)) def from_disk(self, path, **exclude): - actions = [] - deserializers = { - 'strings': lambda p: self.strings.from_disk(p), - 'actions': lambda p: actions.extend(ujson.load(p.open())) - } - util.from_disk(path, deserializers, exclude) - for move, label in actions: - self.add_action(move, label) + with path.open('rb') as file_: + byte_data = file_.read() + self.from_bytes(byte_data, **exclude) return self def to_bytes(self, **exclude): diff --git a/spacy/tests/README.md b/spacy/tests/README.md index a7699ed54..fd47ae579 100644 --- a/spacy/tests/README.md +++ b/spacy/tests/README.md @@ -13,21 +13,32 @@ Tests for spaCy modules and classes live in their own directories of the same na 2. [Dos and don'ts](#dos-and-donts) 3. [Parameters](#parameters) 4. [Fixtures](#fixtures) -5. [Helpers and utilities](#helpers-and-utilities) -6. [Contributing to the tests](#contributing-to-the-tests) +5. [Testing models](#testing-models) +6. [Helpers and utilities](#helpers-and-utilities) +7. [Contributing to the tests](#contributing-to-the-tests) ## Running the tests +To show print statements, run the tests with `py.test -s`. To abort after the +first failure, run them with `py.test -x`. + ```bash -py.test spacy # run basic tests -py.test spacy --models # run basic and model tests -py.test spacy --slow # run basic and slow tests -py.test spacy --models --slow # run all tests +py.test spacy # run basic tests +py.test spacy --models --en # run basic and English model tests +py.test spacy --models --all # run basic and all model tests +py.test spacy --slow # run basic and slow tests +py.test spacy --models --all --slow # run all tests ``` -To show print statements, run the tests with `py.test -s`. To abort after the first failure, run them with `py.test -x`. +You can also run tests in a specific file or directory, or even only one +specific test: +```bash +py.test spacy/tests/tokenizer # run all tests in directory +py.test spacy/tests/tokenizer/test_exceptions.py # run all tests in file +py.test spacy/tests/tokenizer/test_exceptions.py::test_tokenizer_handles_emoji # run specific test +``` ## Dos and don'ts @@ -83,14 +94,9 @@ These are the main fixtures that are currently available: | Fixture | Description | | --- | --- | | `tokenizer` | Creates **all available** language tokenizers and runs the test for **each of them**. | -| `en_tokenizer` | Creates an English `Tokenizer` object. | -| `de_tokenizer` | Creates a German `Tokenizer` object. | -| `hu_tokenizer` | Creates a Hungarian `Tokenizer` object. | -| `en_vocab` | Creates an English `Vocab` object. | -| `en_entityrecognizer` | Creates an English `EntityRecognizer` object. | -| `lemmatizer` | Creates a `Lemmatizer` object from the installed language data (`None` if no data is found). -| `EN` | Creates an instance of `English`. Only use for tests that require the models. | -| `DE` | Creates an instance of `German`. Only use for tests that require the models. | +| `en_tokenizer`, `de_tokenizer`, ... | Creates an English, German etc. tokenizer. | +| `en_vocab`, `en_entityrecognizer`, ... | Creates an instance of the English `Vocab`, `EntityRecognizer` object etc. | +| `EN`, `DE`, ... | Creates a language class with a loaded model. For more info, see [Testing models](#testing-models). | | `text_file` | Creates an instance of `StringIO` to simulate reading from and writing to files. | | `text_file_b` | Creates an instance of `ByteIO` to simulate reading from and writing to files. | @@ -103,6 +109,48 @@ def test_module_do_something(en_tokenizer): If all tests in a file require a specific configuration, or use the same complex example, it can be helpful to create a separate fixture. This fixture should be added at the top of each file. Make sure to use descriptive names for these fixtures and don't override any of the global fixtures listed above. **From looking at a test, it should immediately be clear which fixtures are used, and where they are coming from.** +## Testing models + +Models should only be loaded and tested **if absolutely necessary** – for example, if you're specifically testing a model's performance, or if your test is related to model loading. If you only need an annotated `Doc`, you should use the `get_doc()` helper function to create it manually instead. + +To specify which language models a test is related to, set the language ID as an argument of `@pytest.mark.models`. This allows you to later run the tests with `--models --en`. You can then use the `EN` [fixture](#fixtures) to get a language +class with a loaded model. + +```python +@pytest.mark.models('en') +def test_english_model(EN): + doc = EN(u'This is a test') +``` + +> ⚠️ **Important note:** In order to test models, they need to be installed as a packge. The [conftest.py](conftest.py) includes a list of all available models, mapped to their IDs, e.g. `en`. Unless otherwise specified, each model that's installed in your environment will be imported and tested. If you don't have a model installed, **the test will be skipped**. + +Under the hood, `pytest.importorskip` is used to import a model package and skip the test if the package is not installed. The `EN` fixture for example gets all +available models for `en`, [parametrizes](#parameters) them to run the test for *each of them*, and uses `load_test_model()` to import the model and run the test, or skip it if the model is not installed. + +### Testing specific models + +Using the `load_test_model()` helper function, you can also write tests for specific models, or combinations of them: + +```python +from .util import load_test_model + +@pytest.mark.models('en') +def test_en_md_only(): + nlp = load_test_model('en_core_web_md') + # test something specific to en_core_web_md + +@pytest.mark.models('en', 'fr') +@pytest.mark.parametrize('model', ['en_core_web_md', 'fr_depvec_web_lg']) +def test_different_models(model): + nlp = load_test_model(model) + # test something specific to the parametrized models +``` + +### Known issues and future improvements + +Using `importorskip` on a list of model packages is not ideal and we're looking to improve this in the future. But at the moment, it's the best way to ensure that tests are performed on specific model packages only, and that you'll always be able to run the tests, even if you don't have *all available models* installed. (If the tests made a call to `spacy.load('en')` instead, this would load whichever model you've created an `en` shortcut for. This may be one of spaCy's default models, but it could just as easily be your own custom English model.) + +The current setup also doesn't provide an easy way to only run tests on specific model versions. The `minversion` keyword argument on `pytest.importorskip` can take care of this, but it currently only checks for the package's `__version__` attribute. An alternative solution would be to load a model package's meta.json and skip if the model's version does not match the one specified in the test. ## Helpers and utilities @@ -152,11 +200,11 @@ print([token.dep_ for token in doc]) **Note:** There's currently no way of setting the serializer data for the parser without loading the models. If this is relevant to your test, constructing the `Doc` via `get_doc()` won't work. - ### Other utilities | Name | Description | | --- | --- | +| `load_test_model` | Load a model if it's installed as a package, otherwise skip test. | | `apply_transition_sequence(parser, doc, sequence)` | Perform a series of pre-specified transitions, to put the parser in a desired state. | | `add_vecs_to_vocab(vocab, vectors)` | Add list of vector tuples (`[("text", [1, 2, 3])]`) to given vocab. All vectors need to have the same length. | | `get_cosine(vec1, vec2)` | Get cosine for two given vectors. | diff --git a/spacy/tests/doc/test_doc_api.py b/spacy/tests/doc/test_doc_api.py index 4281193dd..cbe1bbc66 100644 --- a/spacy/tests/doc/test_doc_api.py +++ b/spacy/tests/doc/test_doc_api.py @@ -102,7 +102,7 @@ def test_doc_api_getitem(en_tokenizer): def test_doc_api_serialize(en_tokenizer, text): tokens = en_tokenizer(text) new_tokens = get_doc(tokens.vocab).from_bytes(tokens.to_bytes()) - assert tokens.string == new_tokens.string + assert tokens.text == new_tokens.text assert [t.text for t in tokens] == [t.text for t in new_tokens] assert [t.orth for t in tokens] == [t.orth for t in new_tokens] diff --git a/spacy/tests/tokenizer/test_exceptions.py b/spacy/tests/tokenizer/test_exceptions.py index 70fb103dc..57281b998 100644 --- a/spacy/tests/tokenizer/test_exceptions.py +++ b/spacy/tests/tokenizer/test_exceptions.py @@ -41,7 +41,5 @@ def test_tokenizer_excludes_false_pos_emoticons(tokenizer, text, length): @pytest.mark.parametrize('text,length', [('can you still dunk?🍕🍔😵LOL', 8), ('i💙you', 3), ('🤘🤘yay!', 4)]) def test_tokenizer_handles_emoji(tokenizer, text, length): - exceptions = ["hu"] tokens = tokenizer(text) - if tokens[0].lang_ not in exceptions: - assert len(tokens) == length + assert len(tokens) == length diff --git a/spacy/tests/util.py b/spacy/tests/util.py index 385ff414b..476ddb993 100644 --- a/spacy/tests/util.py +++ b/spacy/tests/util.py @@ -12,6 +12,7 @@ MODELS = {} def load_test_model(model): + """Load a model if it's installed as a package, otherwise skip.""" if model not in MODELS: module = pytest.importorskip(model) MODELS[model] = module.load() diff --git a/spacy/tokenizer.pyx b/spacy/tokenizer.pyx index c2671d785..44a9a3bae 100644 --- a/spacy/tokenizer.pyx +++ b/spacy/tokenizer.pyx @@ -6,8 +6,8 @@ from cython.operator cimport dereference as deref from cython.operator cimport preincrement as preinc from cymem.cymem cimport Pool from preshed.maps cimport PreshMap +import regex as re -import dill from .strings cimport hash_string from . import util cimport cython @@ -344,8 +344,8 @@ cdef class Tokenizer: strings or `Path`-like objects. RETURNS (Tokenizer): The modified `Tokenizer` object. """ - with path.open('wb') as file_: - bytes_data = file_.read(path) + with path.open('rb') as file_: + bytes_data = file_.read() self.from_bytes(bytes_data, **exclude) return self @@ -355,14 +355,13 @@ cdef class Tokenizer: **exclude: Named attributes to prevent from being serialized. RETURNS (bytes): The serialized form of the `Tokenizer` object. """ - # TODO: Improve this so it doesn't need pickle serializers = { 'vocab': lambda: self.vocab.to_bytes(), - 'prefix': lambda: dill.dumps(self.prefix_search), - 'suffix_search': lambda: dill.dumps(self.suffix_search), - 'infix_finditer': lambda: dill.dumps(self.infix_finditer), - 'token_match': lambda: dill.dumps(self.token_match), - 'exceptions': lambda: dill.dumps(self._rules) + 'prefix': lambda: self.prefix_search.__self__.pattern, + 'suffix_search': lambda: self.suffix_search.__self__.pattern, + 'infix_finditer': lambda: self.infix_finditer.__self__.pattern, + 'token_match': lambda: self.token_match.__self__.pattern, + 'exceptions': lambda: self._rules } return util.to_bytes(serializers, exclude) @@ -373,26 +372,23 @@ cdef class Tokenizer: **exclude: Named attributes to prevent from being loaded. RETURNS (Tokenizer): The `Tokenizer` object. """ - # TODO: Improve this so it doesn't need pickle data = {} deserializers = { 'vocab': lambda b: self.vocab.from_bytes(b), - 'prefix': lambda b: data.setdefault('prefix', dill.loads(b)), - 'suffix_search': lambda b: data.setdefault('suffix_search', dill.loads(b)), - 'infix_finditer': lambda b: data.setdefault('infix_finditer', dill.loads(b)), - 'token_match': lambda b: data.setdefault('token_match', dill.loads(b)), - 'exceptions': lambda b: data.setdefault('rules', dill.loads(b)) + 'prefix': lambda b: data.setdefault('prefix', b), + 'suffix_search': lambda b: data.setdefault('suffix_search', b), + 'infix_finditer': lambda b: data.setdefault('infix_finditer', b), + 'token_match': lambda b: data.setdefault('token_match', b), + 'exceptions': lambda b: data.setdefault('rules', b) } msg = util.from_bytes(bytes_data, deserializers, exclude) if 'prefix' in data: - self.prefix_search = data['prefix'] + self.prefix_search = re.compile(data['prefix']).search if 'suffix' in data: - self.suffix_search = data['suffix'] + self.suffix_search = re.compile(data['suffix']).search if 'infix' in data: - self.infix_finditer = data['infix'] + self.infix_finditer = re.compile(data['infix']).finditer if 'token_match' in data: - self.token_match = data['token_match'] + self.token_match = re.compile(data['token_match']).search for string, substrings in data.get('rules', {}).items(): self.add_special_case(string, substrings) - - diff --git a/spacy/tokens/doc.pyx b/spacy/tokens/doc.pyx index e9d23c568..84b39d454 100644 --- a/spacy/tokens/doc.pyx +++ b/spacy/tokens/doc.pyx @@ -30,6 +30,7 @@ from ..syntax.iterators import CHUNKERS from ..util import normalize_slice from ..compat import is_config from .. import about +from .. import util DEF PADDING = 5 @@ -252,8 +253,12 @@ cdef class Doc: def __get__(self): if 'has_vector' in self.user_hooks: return self.user_hooks['has_vector'](self) - - return any(token.has_vector for token in self) + elif any(token.has_vector for token in self): + return True + elif self.tensor: + return True + else: + return False property vector: """A real-valued meaning representation. Defaults to an average of the @@ -265,12 +270,16 @@ cdef class Doc: def __get__(self): if 'vector' in self.user_hooks: return self.user_hooks['vector'](self) - if self._vector is None: - if len(self): - self._vector = sum(t.vector for t in self) / len(self) - else: - return numpy.zeros((self.vocab.vectors_length,), dtype='float32') - return self._vector + if self._vector is not None: + return self._vector + elif self.has_vector and len(self): + self._vector = sum(t.vector for t in self) / len(self) + return self._vector + elif self.tensor: + self._vector = self.tensor.mean(axis=0) + return self._vector + else: + return numpy.zeros((self.vocab.vectors_length,), dtype='float32') def __set__(self, value): self._vector = value @@ -295,10 +304,6 @@ cdef class Doc: def __set__(self, value): self._vector_norm = value - @property - def string(self): - return self.text - property text: """A unicode representation of the document text. @@ -598,15 +603,16 @@ cdef class Doc: self.is_tagged = bool(TAG in attrs or POS in attrs) return self - def to_disk(self, path): + def to_disk(self, path, **exclude): """Save the current state to a directory. path (unicode or Path): A path to a directory, which will be created if it doesn't exist. Paths may be either strings or `Path`-like objects. """ - raise NotImplementedError() + with path.open('wb') as file_: + file_.write(self.to_bytes(**exclude)) - def from_disk(self, path): + def from_disk(self, path, **exclude): """Loads state from a directory. Modifies the object in place and returns it. @@ -614,25 +620,28 @@ cdef class Doc: strings or `Path`-like objects. RETURNS (Doc): The modified `Doc` object. """ - raise NotImplementedError() + with path.open('rb') as file_: + bytes_data = file_.read() + self.from_bytes(bytes_data, **exclude) - def to_bytes(self): + def to_bytes(self, **exclude): """Serialize, i.e. export the document contents to a binary string. RETURNS (bytes): A losslessly serialized copy of the `Doc`, including all annotations. """ - return dill.dumps( - (self.text, - self.to_array([LENGTH,SPACY,TAG,LEMMA,HEAD,DEP,ENT_IOB,ENT_TYPE]), - self.sentiment, - self.tensor, - self.noun_chunks_iterator, - self.user_data, - (self.user_hooks, self.user_token_hooks, self.user_span_hooks)), - protocol=-1) + array_head = [LENGTH,SPACY,TAG,LEMMA,HEAD,DEP,ENT_IOB,ENT_TYPE] + serializers = { + 'text': lambda: self.text, + 'array_head': lambda: array_head, + 'array_body': lambda: self.to_array(array_head), + 'sentiment': lambda: self.sentiment, + 'tensor': lambda: self.tensor, + 'user_data': lambda: self.user_data + } + return util.to_bytes(serializers, exclude) - def from_bytes(self, data): + def from_bytes(self, bytes_data, **exclude): """Deserialize, i.e. import the document contents from a binary string. data (bytes): The string to load from. @@ -640,27 +649,36 @@ cdef class Doc: """ if self.length != 0: raise ValueError("Cannot load into non-empty Doc") + deserializers = { + 'text': lambda b: None, + 'array_head': lambda b: None, + 'array_body': lambda b: None, + 'sentiment': lambda b: None, + 'tensor': lambda b: None, + 'user_data': lambda user_data: self.user_data.update(user_data) + } + + msg = util.from_bytes(bytes_data, deserializers, exclude) + cdef attr_t[:, :] attrs cdef int i, start, end, has_space - fields = dill.loads(data) - text, attrs = fields[:2] - self.sentiment, self.tensor = fields[2:4] - self.noun_chunks_iterator, self.user_data = fields[4:6] - self.user_hooks, self.user_token_hooks, self.user_span_hooks = fields[6] + self.sentiment = msg['sentiment'] + self.tensor = msg['tensor'] start = 0 cdef const LexemeC* lex cdef unicode orth_ + text = msg['text'] + attrs = msg['array_body'] for i in range(attrs.shape[0]): end = start + attrs[i, 0] has_space = attrs[i, 1] orth_ = text[start:end] lex = self.vocab.get(self.mem, orth_) self.push_back(lex, has_space) - start = end + has_space - self.from_array([TAG,LEMMA,HEAD,DEP,ENT_IOB,ENT_TYPE], - attrs[:, 2:]) + self.from_array(msg['array_head'][2:], + attrs[:, 2:]) return self def merge(self, int start_idx, int end_idx, *args, **attributes): diff --git a/spacy/tokens/token.pyx b/spacy/tokens/token.pyx index ee98a7244..74610e25e 100644 --- a/spacy/tokens/token.pyx +++ b/spacy/tokens/token.pyx @@ -111,7 +111,7 @@ cdef class Token: RETURNS (float): A scalar similarity score. Higher is more similar. """ if 'similarity' in self.doc.user_token_hooks: - return self.doc.user_token_hooks['similarity'](self) + return self.doc.user_token_hooks['similarity'](self) if self.vector_norm == 0 or other.vector_norm == 0: return 0.0 return numpy.dot(self.vector, other.vector) / (self.vector_norm * other.vector_norm) @@ -245,7 +245,10 @@ cdef class Token: def __get__(self): if 'vector' in self.doc.user_token_hooks: return self.doc.user_token_hooks['vector'](self) - return self.vocab.get_vector(self.c.lex.orth) + if self.has_vector: + return self.vocab.get_vector(self.c.lex.orth) + else: + return self.doc.tensor[self.i] property vector_norm: """The L2 norm of the token's vector representation. diff --git a/spacy/util.py b/spacy/util.py index ff5c33aab..dabceb4a8 100644 --- a/spacy/util.py +++ b/spacy/util.py @@ -13,6 +13,7 @@ import random import numpy import io import dill +from collections import OrderedDict import msgpack import msgpack_numpy @@ -418,7 +419,7 @@ def get_raw_input(description, default=False): def to_bytes(getters, exclude): - serialized = {} + serialized = OrderedDict() for key, getter in getters.items(): if key not in exclude: serialized[key] = getter() @@ -433,6 +434,24 @@ def from_bytes(bytes_data, setters, exclude): return msg +def to_disk(path, writers, exclude): + path = ensure_path(path) + if not path.exists(): + path.mkdir() + for key, writer in writers.items(): + if key not in exclude: + writer(path / key) + return path + + +def from_disk(path, readers, exclude): + path = ensure_path(path) + for key, reader in readers.items(): + if key not in exclude: + reader(path / key) + return path + + # This stuff really belongs in thinc -- but I expect # to refactor how all this works in thinc anyway. # What a mess! diff --git a/spacy/vocab.pxd b/spacy/vocab.pxd index 3c31a8f8f..8005cbf06 100644 --- a/spacy/vocab.pxd +++ b/spacy/vocab.pxd @@ -27,7 +27,8 @@ cdef struct _Cached: cdef class Vocab: cdef Pool mem cpdef readonly StringStore strings - cpdef readonly Morphology morphology + cpdef public Morphology morphology + cpdef public object vectors cdef readonly int length cdef public object data_dir cdef public object lex_attr_getters @@ -35,11 +36,10 @@ cdef class Vocab: cdef const LexemeC* get(self, Pool mem, unicode string) except NULL cdef const LexemeC* get_by_orth(self, Pool mem, attr_t orth) except NULL cdef const TokenC* make_fused_token(self, substrings) except NULL - + cdef const LexemeC* _new_lexeme(self, Pool mem, unicode string) except NULL cdef int _add_lex_to_vocab(self, hash_t key, const LexemeC* lex) except -1 cdef const LexemeC* _new_lexeme(self, Pool mem, unicode string) except NULL cdef PreshMap _by_hash cdef PreshMap _by_orth - cdef readonly int vectors_length diff --git a/spacy/vocab.pyx b/spacy/vocab.pyx index 45c9e1a07..f11d3a6ef 100644 --- a/spacy/vocab.pyx +++ b/spacy/vocab.pyx @@ -239,6 +239,16 @@ cdef class Vocab: Token.set_struct_attr(token, attr_id, value) return tokens + @property + def vectors_length(self): + raise NotImplementedError + + def clear_vectors(self): + """Drop the current vector table. Because all vectors must be the same + width, you have to call this to change the size of the vectors. + """ + raise NotImplementedError + def get_vector(self, orth): """Retrieve a vector for a word in the vocabulary. @@ -253,6 +263,16 @@ cdef class Vocab: """ raise NotImplementedError + def set_vector(self, orth, vector): + """Set a vector for a word in the vocabulary. + + Words can be referenced by string or int ID. + + RETURNS: + None + """ + raise NotImplementedError + def has_vector(self, orth): """Check whether a word has a vector. Returns False if no vectors have been loaded. Words can be looked up by string diff --git a/website/_includes/_mixins-base.jade b/website/_includes/_mixins-base.jade index 484f29afc..0c44ce5e2 100644 --- a/website/_includes/_mixins-base.jade +++ b/website/_includes/_mixins-base.jade @@ -108,8 +108,8 @@ mixin quickstart(groups, headline, description, hide_results) | #[+help(group.help)] .c-quickstart__fields for option in group.options - input.c-quickstart__input(class="c-quickstart__input--" + (group.input_style ? group.input_style : group.multiple ? "check" : "radio") type=group.multiple ? "checkbox" : "radio" name=group.id id=option.id value=option.id checked=option.checked) - label.c-quickstart__label(for=option.id)!=option.title + input.c-quickstart__input(class="c-quickstart__input--" + (group.input_style ? group.input_style : group.multiple ? "check" : "radio") type=group.multiple ? "checkbox" : "radio" name=group.id id="qs-#{option.id}" value=option.id checked=option.checked) + label.c-quickstart__label(for="qs-#{option.id}")!=option.title if option.meta | #[span.c-quickstart__label__meta (#{option.meta})] if option.help diff --git a/website/docs/usage/index.jade b/website/docs/usage/index.jade index b12fb0c9d..c79c689a4 100644 --- a/website/docs/usage/index.jade +++ b/website/docs/usage/index.jade @@ -354,12 +354,14 @@ p python -c "import os; import spacy; print(os.path.dirname(spacy.__file__))" p - | Then run #[code pytest] on that directory. The flags #[code --vectors], - | #[code --slow] and #[code --model] are optional and enable additional - | tests: + | Then run #[code pytest] on that directory. The flags #[code --slow] and + | #[code --model] are optional and enable additional tests. +code(false, "bash"). # make sure you are using recent pytest version python -m pip install -U pytest - python -m pytest <spacy-directory> --vectors --models --slow + python -m pytest <spacy-directory> # basic tests + python -m pytest <spacy-directory> --slow # basic and slow tests + python -m pytest <spacy-directory> --models --all # basic and all model tests + python -m pytest <spacy-directory> --models --en # basic and English model tests diff --git a/website/docs/usage/rule-based-matching.jade b/website/docs/usage/rule-based-matching.jade index 8588729b6..71400ea55 100644 --- a/website/docs/usage/rule-based-matching.jade +++ b/website/docs/usage/rule-based-matching.jade @@ -408,7 +408,7 @@ p | To label the hashtags, we first need to add a new custom flag. | #[code IS_HASHTAG] will be the flag's ID, which you can use to assign it | to the hashtag's span, and check its value via a token's - | #[+api("token#check_flag") #[code code check_flag()]] method. On each + | #[+api("token#check_flag") #[code check_flag()]] method. On each | match, we merge the hashtag and assign the flag. +code. diff --git a/website/docs/usage/spacy-101.jade b/website/docs/usage/spacy-101.jade index 052942672..e1300b5b0 100644 --- a/website/docs/usage/spacy-101.jade +++ b/website/docs/usage/spacy-101.jade @@ -2,6 +2,13 @@ include ../../_includes/_mixins +p + | Whether you're new to spaCy, or just want to brush up on some + | NLP basics and implementation details – this page should have you covered. + | Each section will explain one of spaCy's features in simple terms and + | with examples or illustrations. Some sections will also reappear across + | the usage guides as a quick introcution. + +aside("Help us improve the docs") | Did you spot a mistake or come across explanations that | are unclear? We always appreciate improvement @@ -13,6 +20,23 @@ include ../../_includes/_mixins +grid.o-no-block +grid-col("half") + p + | spaCy is a #[strong free, open-source library] for advanced + | #[strong Natural Language Processing] (NLP) in Python. + + p + | If you're working with a lot of text, you'll eventually want to + | know more about it. For example, what's it about? What do the + | words mean in context? Who is doing what to whom? What companies + | and products are mentioned? Which texts are similar to each other? + + p + | spaCy is designed specifically for #[strong production use] and + | helps you build applications that process and "understand" + | large volumes of text. It can be used to build + | #[strong information extraction] or + | #[strong natural language understanding] systems, or to + | pre-process text for #[strong deep learning]. +grid-col("half") +infobox @@ -31,6 +55,29 @@ include ../../_includes/_mixins +item #[+a("#architecture") Architecture] +item #[+a("#community") Community & FAQ] ++h(3, "what-spacy-isnt") What spaCy isn't + ++list + +item #[strong spaCy is not a platform or "an API"]. + | Unlike a platform, spaCy does not provide a software as a service, or + | a web application. It's an open-source library designed to help you + | build NLP applications, not a consumable service. + +item #[strong spaCy is not an out-of-the-box chat bot engine]. + | While spaCy can be used to power conversational applications, it's + | not designed specifically for chat bots, and only provides the + | underlying text processing capabilities. + +item #[strong spaCy is not research software]. + | It's is built on the latest research, but unlike + | #[+a("https://github./nltk/nltk") NLTK], which is intended for + | teaching and research, spaCy follows a more opinionated approach and + | focuses on production usage. Its aim is to provide you with the best + | possible general-purpose solution for text processing and machine learning + | with text input – but this also means that there's only one implementation + | of each component. + +item #[strong spaCy is not a company]. + | It's an open-source library. Our company publishing spaCy and other + | software is called #[+a(COMPANY_URL, true) Explosion AI]. + +h(2, "features") Features p