def test_text_model_params():
assert isinstance(text2mat(data, semantic={
'model' : 'LatentDirichletAllocation',
'params' : {
'learning_method' : 'batch'
}}
, corpus=data)[0], np.ndarray)
def test_transform_text():
assert isinstance(text2mat(data)[0], np.ndarray)
def test_corpus():
assert text2mat(data, corpus=data)[0].shape[1]==20
def test_LDA_class_instance():
user_model = LatentDirichletAllocation(n_components=15)
assert text2mat(data, semantic=user_model, corpus=data)[0].shape[1]==15
def test_LDA_class():
assert text2mat(data, semantic=LatentDirichletAllocation, corpus=data)[0].shape[1]==10
def test_vectorizer_params():
assert text2mat(data, vectorizer={
'model' : 'CountVectorizer',
'params': {
'max_features' : 2
}}, corpus=data)[0].shape[1]==20
def test_transform_no_text_model():
assert isinstance(text2mat(data, semantic=None, corpus=data)[0], np.ndarray)
def test_tfidf_NMF():
isinstance(text2mat(data, vectorizer='TfidfVectorizer', semantic='NMF', corpus=data)[0], np.ndarray)
def test_count_NMF():
isinstance(text2mat(data, vectorizer='CountVectorizer', semantic='NMF', corpus=data)[0], np.ndarray)
def test_tfidf_LDA():
isinstance(text2mat(data, vectorizer='TfidfVectorizer',
semantic='LatentDirichletAllocation', corpus=data)[0], np.ndarray)