def test_vector(text_corpora: TextCorpora,
corpus_collection: Dict[str, Corpus]) -> None:
vector_cat1 = corpus_collection['cat1'].vectors
vector_cat2 = corpus_collection['cat2'].vectors
vector = np.vstack((vector_cat1, vector_cat2))
assert np.all(np.isclose(text_corpora.vectors('cat1'), vector_cat1))
assert np.all(np.isclose(text_corpora.vectors('cat2'), vector_cat2))
assert np.all(np.isclose(text_corpora.vectors(), vector))
assert np.all(np.isclose(text_corpora.vectors(['cat1', 'cat2']), vector))
def test_add_with_records(
spacy_lang: Language, record_collection: Dict[str, Sequence[str]]
) -> None: # pylint: disable=missing-docstring
corpora = TextCorpora(spacy_lang)
corpora.add('cat1', record_collection['cat1'])
corpora.add('cat2', record_collection['cat2'])
assert len(corpora) == 4
assert corpora.labels == ['cat1', 'cat2']
def test_can_add_docs(
spacy_lang: Language, doc_collection: Dict[str, Sequence[Doc]]
) -> None: # pylint: disable=missing-docstring
corpora = TextCorpora(spacy_lang)
corpora.add_docs('cat1', doc_collection['cat1'])
corpora.add_docs('cat2', doc_collection['cat2'])
assert len(corpora) == 4
assert corpora.labels == ['cat1', 'cat2']
def test_add_labelled_texts(
spacy_lang: Language, text_collection: Dict[str, Sequence[str]]
) -> None: # pylint: disable=missing-docstring
corpora = TextCorpora(spacy_lang)
labelled_collection = to_labelled(text_collection)
corpora.add_labelled(labelled_collection)
assert len(corpora) == 4
assert corpora.labels == ['cat1', 'cat2']
def test_word_counts_attributes(text_corpora: TextCorpora,
corpus_collection: Dict[str, Corpus]) -> None:
wc_cat1 = corpus_collection['cat1']\
.word_counts(as_strings=True)
wc_cat2 = corpus_collection['cat2']\
.word_counts(as_strings=True)
wc = tlz.merge_with(sum, wc_cat1, wc_cat2)
assert text_corpora.word_counts(corpora='cat1', as_strings=True) == wc_cat1
assert text_corpora.word_counts(corpora='cat2', as_strings=True) == wc_cat2
assert text_corpora.word_counts(as_strings=True) == wc
assert text_corpora.word_counts(corpora=['cat2', 'cat1'],
as_strings=True) == wc
def test_can_add_text(
spacy_lang: Language, text_collection: Dict[str, Sequence[str]]
) -> None: # pylint: disable=missing-docstring
corpora = TextCorpora(spacy_lang)
for text in text_collection['cat1']:
corpora.add_text('cat1', text)
for text in text_collection['cat2']:
corpora.add_text('cat2', text)
assert len(corpora) == 4
assert corpora.labels == ['cat1', 'cat2']
def test_can_add_corpus(spacy_lang: Language,
corpus_collection: Dict[str, Corpus]) -> None:
corpora = TextCorpora(spacy_lang)
corpora.add_corpus('cat1', corpus_collection['cat1'])
assert corpora['cat1'] == corpus_collection['cat1']
def test_n_tokens_attributes(text_corpora: TextCorpora) -> None:
assert text_corpora.n_tokens() == 51
assert text_corpora.n_tokens('cat1') == 28
assert text_corpora.n_tokens('cat2') == 23
assert text_corpora.n_tokens(['cat2', 'cat1']) == 51
def test_n_sents_attributes(text_corpora: TextCorpora) -> None:
assert text_corpora.n_sents() == 6
assert text_corpora.n_sents('cat1') == 2
assert text_corpora.n_sents('cat2') == 4
assert text_corpora.n_sents(['cat2', 'cat1']) == 6
def test_can_create_corpora(spacy_lang: Language,
corpus_collection: Dict[str, Corpus]) -> None:
corpora = TextCorpora(spacy_lang, corpus_collection)
assert isinstance(corpora, TextCorpora)
assert corpora.labels == ['cat1', 'cat2']
assert corpora.n_corpora == 2
def test_can_create_empty_corpora(spacy_lang: Language) -> None:
corpora = TextCorpora(spacy_lang)
assert isinstance(corpora, TextCorpora)
def text_corpora(spacy_lang: Language,
corpus_collection: Dict[str, Corpus]) -> TextCorpora:
return TextCorpora(spacy_lang, corpus_collection)