def test_word_embedding_extractor():
pytest.importorskip('gensim')
stims = [TextStim(text='this'), TextStim(text='sentence')]
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True)
result = merge_results(ext.transform(stims), extractor_names='multi',
format='wide')
assert ('WordEmbeddingExtractor', 'embedding_dim99') in result.columns
assert np.allclose(0.0010911,
result[('WordEmbeddingExtractor', 'embedding_dim0')][0])
unk = TextStim(text='nowaythisinvocab')
result = ext.transform(unk).to_df()
assert result['embedding_dim10'][0] == 0.0
ones = np.ones(100)
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True, unk_vector=ones)
result = ext.transform(unk).to_df()
assert result['embedding_dim10'][0] == 1.0
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True, unk_vector='random')
result = ext.transform(unk).to_df()
assert result['embedding_dim10'][0] <= 1.0
assert result['embedding_dim10'][0] >= -1.0
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True, unk_vector='nothing')
result = ext.transform(unk).to_df()
assert result['embedding_dim10'][0] == 0.0
def test_indico_api_text_extractor():
ext = IndicoAPITextExtractor(api_key=os.environ['INDICO_APP_KEY'],
models=['emotion', 'personality'])
# With ComplexTextStim input
srtfile = join(get_test_data_path(), 'text', 'wonderful.srt')
srt_stim = ComplexTextStim(srtfile, onset=4.2)
result = merge_results(ext.transform(srt_stim), extractor_names=False)
outdfKeysCheck = {
'onset', 'duration', 'order', 'object_id', 'emotion_anger',
'emotion_fear', 'emotion_joy', 'emotion_sadness', 'emotion_surprise',
'personality_openness', 'personality_extraversion',
'personality_agreeableness', 'personality_conscientiousness'
}
meta_columns = {'source_file', 'history', 'class', 'filename'}
assert set(result.columns) - set(['stim_name'
]) == outdfKeysCheck | meta_columns
assert result['onset'][1] == 92.622
# With TextStim input
ts = TextStim(text="It's a wonderful life.")
result = ext.transform(ts).to_df(object_id=True)
assert set(result.columns) == outdfKeysCheck
assert len(result) == 1
def test_indico_api_image_extractor():
ext = IndicoAPIImageExtractor(api_key=os.environ['INDICO_APP_KEY'],
models=['fer', 'content_filtering'])
stim1 = ImageStim(join(IMAGE_DIR, 'apple.jpg'))
result1 = merge_results(ext.transform([stim1, stim1]),
extractor_names=False)
outdfKeysCheck = {
'object_id', 'fer_Surprise', 'fer_Neutral', 'fer_Sad', 'fer_Happy',
'fer_Angry', 'fer_Fear', 'content_filtering'
}
meta_columns = {
'source_file', 'history', 'class', 'filename', 'onset', 'duration',
'order'
}
assert set(result1.columns) - set(['stim_name'
]) == outdfKeysCheck | meta_columns
assert result1['content_filtering'][0] < 0.2
stim2 = ImageStim(join(IMAGE_DIR, 'obama.jpg'))
result2 = ext.transform(stim2).to_df(timing=False, object_id=True)
assert set(result2.columns) == outdfKeysCheck
assert result2['fer_Happy'][0] > 0.7
url = 'https://tuition.utexas.edu/sites/all/themes/tuition/logo.png'
stim = ImageStim(url=url)
result = ext.transform(stim).to_df()
assert result['fer_Neutral'][0] > 0.1
def test_indico_api_image_extractor():
ext = IndicoAPIImageExtractor(api_key=os.environ['INDICO_APP_KEY'],
models=['fer', 'content_filtering'])
image_dir = join(get_test_data_path(), 'image')
stim1 = ImageStim(join(image_dir, 'apple.jpg'))
result1 = merge_results(ext.transform([stim1, stim1]),
extractor_names=False)
outdfKeysCheck = {
'object_id', 'fer_Surprise', 'fer_Neutral', 'fer_Sad', 'fer_Happy',
'fer_Angry', 'fer_Fear', 'content_filtering'
}
meta_columns = {
'source_file', 'history', 'class', 'filename', 'onset', 'duration',
'order'
}
assert set(result1.columns) - set(['stim_name'
]) == outdfKeysCheck | meta_columns
assert result1['content_filtering'][0] < 0.2
stim2 = ImageStim(join(image_dir, 'obama.jpg'))
result2 = ext.transform(stim2).to_df(timing=False, object_id=True)
assert set(result2.columns) == outdfKeysCheck
assert result2['fer_Happy'][0] > 0.7
def test_word_embedding_extractor():
pytest.importorskip('gensim')
stims = [TextStim(text='this'), TextStim(text='sentence')]
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True)
result = merge_results(ext.transform(stims),
extractor_names='multi',
format='wide')
assert ('WordEmbeddingExtractor', 'embedding_dim99') in result.columns
assert np.allclose(0.0010911,
result[('WordEmbeddingExtractor', 'embedding_dim0')][0])
unk = TextStim(text='nowaythisinvocab')
result = ext.transform(unk).to_df()
assert result['embedding_dim10'][0] == 0.0
ones = np.ones(100)
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True,
unk_vector=ones)
result = ext.transform(unk).to_df()
assert result['embedding_dim10'][0] == 1.0
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True,
unk_vector='random')
result = ext.transform(unk).to_df()
assert result['embedding_dim10'][0] <= 1.0
assert result['embedding_dim10'][0] >= -1.0
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True,
unk_vector='nothing')
result = ext.transform(unk).to_df()
assert result['embedding_dim10'][0] == 0.0
def test_indico_api_image_extractor():
ext = IndicoAPIImageExtractor(api_key=os.environ['INDICO_APP_KEY'],
models=['fer', 'content_filtering'])
stim1 = ImageStim(join(IMAGE_DIR, 'apple.jpg'))
result1 = merge_results(ext.transform([stim1, stim1]),
extractor_names=False)
outdfKeysCheck = {
'object_id',
'fer_Surprise',
'fer_Neutral',
'fer_Sad',
'fer_Happy',
'fer_Angry',
'fer_Fear',
'content_filtering'}
meta_columns = {'source_file',
'history',
'class',
'filename',
'onset',
'duration',
'order'}
assert set(result1.columns) - set(['stim_name']) == outdfKeysCheck | meta_columns
assert result1['content_filtering'][0] < 0.2
stim2 = ImageStim(join(IMAGE_DIR, 'obama.jpg'))
result2 = ext.transform(stim2).to_df(timing=False, object_id=True)
assert set(result2.columns) == outdfKeysCheck
assert result2['fer_Happy'][0] > 0.7
url = 'https://via.placeholder.com/350x150'
stim = ImageStim(url=url)
result = ext.transform(stim).to_df()
assert result['fer_Neutral'][0] > 0.
def test_clarifai_api_extractor_batch():
stim = ImageStim(join(IMAGE_DIR, 'apple.jpg'))
stim2 = ImageStim(join(IMAGE_DIR, 'obama.jpg'))
ext = ClarifaiAPIImageExtractor()
results = ext.transform([stim, stim2])
results = merge_results(results)
assert results['ClarifaiAPIImageExtractor#apple'][0] > 0.5 or \
results['ClarifaiAPIImageExtractor#apple'][1] > 0.5
def test_word_embedding_extractor():
pytest.importorskip('gensim')
stims = [TextStim(text='this'), TextStim(text='sentence')]
ext = WordEmbeddingExtractor(join(TEXT_DIR, 'simple_vectors.bin'),
binary=True)
result = merge_results(ext.transform(stims))
assert ('WordEmbeddingExtractor', 'embedding_dim99') in result.columns
assert 0.001091 in result[('WordEmbeddingExtractor', 'embedding_dim0')]
def test_part_of_speech_extractor():
import nltk
nltk.download('tagsets')
stim = ComplexTextStim(join(TEXT_DIR, 'complex_stim_with_header.txt'))
result = merge_results(PartOfSpeechExtractor().transform(stim), extractor_names=False)
assert result.shape == (4, 52)
assert result['NN'].sum() == 1
assert result['VBD'][3] == 1
def test_tfhub_image_reshape():
stim = ImageStim(join(IMAGE_DIR, 'apple.jpg'))
stim2 = ImageStim(join(IMAGE_DIR, 'obama.jpg'))
ext = TFHubImageExtractor(MNET_URL,
reshape_input=(224, 224, 3),
features='feature_vector')
df = merge_results(ext.transform([stim, stim2]), extractor_names=False)
assert df.shape[0] == 2
assert all([len(v) == 1280 for v in df['feature_vector']])
def test_clarifai_api_extractor_batch():
image_dir = join(get_test_data_path(), 'image')
stim = ImageStim(join(image_dir, 'apple.jpg'))
stim2 = ImageStim(join(image_dir, 'obama.jpg'))
ext = ClarifaiAPIExtractor()
results = ext.transform([stim, stim2])
results = merge_results(results)
assert results['ClarifaiAPIExtractor#apple'][0] > 0.5 or \
results['ClarifaiAPIExtractor#apple'][1] > 0.5
def test_clarifai_api_extractor_large():
default = config.get_option('allow_large_jobs')
default_large = config.get_option('large_job')
config.set_option('allow_large_jobs', False)
config.set_option('large_job', 1)
ext = ClarifaiAPIExtractor()
images = [ImageStim(join(get_test_data_path(), 'image', 'apple.jpg'))] * 2
with pytest.raises(ValueError):
merge_results(ext.transform(images))
config.set_option('allow_large_jobs', True)
results = merge_results(ext.transform(images))
assert 'ClarifaiAPIExtractor#apple' in results.columns
assert results.shape == (1, 29) # not 2 cause all the same instance
config.set_option('allow_large_jobs', default)
config.set_option('large_job', default_large)
def test_clarifai_api_extractor_large():
default = config.get_option('allow_large_jobs')
default_large = config.get_option('large_job')
config.set_option('allow_large_jobs', False)
config.set_option('large_job', 1)
ext = ClarifaiAPIImageExtractor()
images = [ImageStim(join(IMAGE_DIR, 'apple.jpg')),
ImageStim(join(IMAGE_DIR, 'obama.jpg'))]
with pytest.raises(ValueError):
merge_results(ext.transform(images))
config.set_option('allow_large_jobs', True)
results = merge_results(ext.transform(images))
assert 'ClarifaiAPIImageExtractor#apple' in results.columns
assert results.shape == (2, 49)
config.set_option('allow_large_jobs', default)
config.set_option('large_job', default_large)
def test_part_of_speech_extractor():
import nltk
nltk.download('tagsets')
stim = ComplexTextStim(join(TEXT_DIR, 'complex_stim_with_header.txt'))
result = merge_results(PartOfSpeechExtractor().transform(stim),
format='wide', extractor_names=False)
assert result.shape == (4, 54)
assert result['NN'].sum() == 1
result = result.sort_values('onset')
assert result['VBD'].iloc[3] == 1
def test_tensor_flow_inception_v3_extractor():
image_dir = join(get_test_data_path(), 'image')
imgs = [join(image_dir, f) for f in ['apple.jpg', 'obama.jpg']]
imgs = [ImageStim(im) for im in imgs]
ext = TensorFlowInceptionV3Extractor()
results = ext.transform(imgs)
df = merge_results(results)
assert len(df) == 2
assert df.iloc[0][('TensorFlowInceptionV3Extractor', 'label_1')] == 'Granny Smith'
assert df.iloc[1][('TensorFlowInceptionV3Extractor', 'score_2')] == '0.17326'
def test_tfhub_text_transformer_tokens():
cstim = ComplexTextStim(join(TEXT_DIR, 'wonderful.txt'))
tkn_ext = TFHubTextExtractor(ELECTRA_URL,
features='token_encodings',
output_key='sequence_output',
preprocessor_url_or_path=TOKENIZER_URL)
tkn_df = merge_results(tkn_ext.transform(cstim.elements[:3]),
extractor_names=False)
assert all([tkn_df['token_encodings'][i].shape == (128, 256) \
for i in range(tkn_df.shape[0])])
def test_tfhub_text_one_feature():
stim = TextStim(join(TEXT_DIR, 'scandal.txt'))
cstim = ComplexTextStim(join(TEXT_DIR, 'wonderful.txt'))
ext = TFHubTextExtractor(GNEWS_URL, output_key=None, features='embedding')
df = merge_results(ext.transform(cstim), extractor_names=False)
assert df.shape[0] == len(cstim.elements)
true = hub.KerasLayer(GNEWS_URL)([cstim.elements[3].text])[0, 2].numpy()
assert np.isclose(df['embedding'][3][2], true)
with pytest.raises(ValueError) as err:
TFHubTextExtractor(GNEWS_URL, output_key='key').transform(stim)
assert 'not a dictionary' in str(err.value)
def test_indico_api_extractor_large():
default = config.get_option('allow_large_jobs')
default_large = config.get_option('large_job')
config.set_option('allow_large_jobs', False)
config.set_option('large_job', 1)
ext = IndicoAPIImageExtractor(models=['fer'])
images = [ImageStim(join(IMAGE_DIR, 'apple.jpg'))] * 2
with pytest.raises(ValueError):
merge_results(ext.transform(images))
config.set_option('allow_large_jobs', True)
results = merge_results(ext.transform(images))
assert 'IndicoAPIImageExtractor#fer_Neutral' in results.columns
assert results.shape == (1, 15) # not 2 rows cause all the same instance
config.set_option('allow_large_jobs', default)
config.set_option('large_job', default_large)
def test_tensor_flow_inception_v3_extractor():
imgs = [join(IMAGE_DIR, f) for f in ['apple.jpg', 'obama.jpg']]
imgs = [ImageStim(im, onset=4.2, duration=1) for im in imgs]
ext = TensorFlowInceptionV3Extractor()
results = ext.transform(imgs)
df = merge_results(results, format='wide', extractor_names='multi')
assert len(df) == 2
assert ('TensorFlowInceptionV3Extractor', 'Granny Smith') in df.columns
assert 0.22610 in df[
('TensorFlowInceptionV3Extractor', 'Windsor tie')].values
assert 4.2 in df[('onset', np.nan)].values
assert 1 in df[('duration', np.nan)].values
def test_indico_api_extractor_large():
default = config.get_option('allow_large_jobs')
default_large = config.get_option('large_job')
config.set_option('allow_large_jobs', False)
config.set_option('large_job', 1)
ext = IndicoAPIImageExtractor(models=['fer'])
images = [ImageStim(join(IMAGE_DIR, 'apple.jpg')),
ImageStim(join(IMAGE_DIR, 'obama.jpg'))]
with pytest.raises(ValueError):
merge_results(ext.transform(images))
config.set_option('allow_large_jobs', True)
results = merge_results(ext.transform(images))
assert 'IndicoAPIImageExtractor#fer_Neutral' in results.columns
assert results.shape == (2, 15)
config.set_option('allow_large_jobs', default)
config.set_option('large_job', default_large)
def test_merge_extractor_results():
np.random.seed(100)
image_dir = join(get_test_data_path(), 'image')
stim1 = ImageStim(join(image_dir, 'apple.jpg'))
stim2 = ImageStim(join(image_dir, 'obama.jpg'))
de_names = ['Extractor1', 'Extractor2', 'Extractor3']
des = [DummyExtractor(name=name) for name in de_names]
results = [de.transform(stim1) for de in des]
results += [de.transform(stim2) for de in des]
df = merge_results(results, format='wide')
assert df.shape == (200, 18)
cols = [
'onset', 'duration', 'order', 'class', 'filename', 'history',
'stim_name', 'source_file'
]
assert not set(cols) - set(df.columns)
assert 'Extractor2#feature_3' in df.columns
df = merge_results(results, format='wide', extractor_names='drop')
assert df.shape == (200, 12)
assert not set(cols) - set(df.columns)
assert 'feature_3' in df.columns
df = merge_results(results, format='wide', extractor_names='multi')
assert df.shape == (200, 18)
_cols = [(c, np.nan) for c in cols]
assert not set(_cols) - set(df.columns)
assert ('Extractor2', 'feature_3') in df.columns
with pytest.raises(ValueError):
merge_results(results, format='long', extractor_names='multi')
df = merge_results(results, format='long', extractor_names='column')
assert df.shape == (1800, 12)
_cols = cols + ['feature', 'extractor', 'value']
assert not set(_cols) - set(df.columns)
row = df.iloc[523, :]
assert row['feature'] == 'feature_2'
assert row['value'] == 475
assert row['extractor'] == 'Extractor1'
df = merge_results(results, format='long', extractor_names='drop')
assert df.shape == (1800, 11)
assert set(_cols) - set(df.columns) == {'extractor'}
df = merge_results(results, format='long', extractor_names='prepend')
assert df.shape == (1800, 11)
row = df.iloc[523, :]
assert row['feature'] == 'Extractor1#feature_2'
def test_merge_extractor_results():
np.random.seed(100)
image_dir = join(get_test_data_path(), 'image')
stim1 = ImageStim(join(image_dir, 'apple.jpg'))
stim2 = ImageStim(join(image_dir, 'obama.jpg'))
de_names = ['Extractor1', 'Extractor2', 'Extractor3']
des = [DummyExtractor(name=name) for name in de_names]
results = [de.transform(stim1) for de in des]
results += [de.transform(stim2) for de in des]
df = merge_results(results, format='wide')
assert df.shape == (200, 18)
cols = ['onset', 'duration', 'order', 'class', 'filename', 'history',
'stim_name', 'source_file']
assert not set(cols) - set(df.columns)
assert 'Extractor2#feature_3' in df.columns
df = merge_results(results, format='wide', extractor_names='drop')
assert df.shape == (200, 12)
assert not set(cols) - set(df.columns)
assert 'feature_3' in df.columns
df = merge_results(results, format='wide', extractor_names='multi')
assert df.shape == (200, 18)
_cols = [(c, np.nan) for c in cols]
assert not set(_cols) - set(df.columns)
assert ('Extractor2', 'feature_3') in df.columns
with pytest.raises(ValueError):
merge_results(results, format='long', extractor_names='multi')
df = merge_results(results, format='long', extractor_names='column')
assert df.shape == (1800, 12)
_cols = cols + ['feature', 'extractor', 'value']
assert not set(_cols) - set(df.columns)
row = df.iloc[523, :]
assert row['feature'] == 'feature_2'
assert row['value'] == 475
assert row['extractor'] == 'Extractor1'
df = merge_results(results, format='long', extractor_names='drop')
assert df.shape == (1800, 11)
assert set(_cols) - set(df.columns) == {'extractor'}
df = merge_results(results, format='long', extractor_names='prepend')
assert df.shape == (1800, 11)
row = df.iloc[523, :]
assert row['feature'] == 'Extractor1#feature_2'
def test_tensor_flow_inception_v3_extractor():
image_dir = join(get_test_data_path(), 'image')
imgs = [join(image_dir, f) for f in ['apple.jpg', 'obama.jpg']]
imgs = [ImageStim(im, onset=4.2, duration=1) for im in imgs]
ext = TensorFlowInceptionV3Extractor()
results = ext.transform(imgs)
df = merge_results(results)
assert len(df) == 2
assert 'Granny Smith' in df[('TensorFlowInceptionV3Extractor',
'label_1')].values
assert '0.22610' in df[('TensorFlowInceptionV3Extractor',
'score_2')].values
assert 4.2 in df[('onset', '')].values
assert 1 in df[('duration', '')].values
def test_tensorflow_keras_inception_v3_extractor():
imgs = [join(IMAGE_DIR, f) for f in ['apple.jpg', 'obama.jpg']]
imgs = [ImageStim(im, onset=4.2, duration=1) for im in imgs]
ext = TensorFlowKerasInceptionV3Extractor()
results = ext.transform(imgs)
df = merge_results(results, format='wide', extractor_names='multi')
assert df.shape == (2, 19)
true = 0.9737075
pred = df['TensorFlowKerasInceptionV3Extractor'].loc[0, 'Granny_Smith']
np.isclose(true, pred, 1e-05)
true = 0.64234024
pred = df['TensorFlowKerasInceptionV3Extractor'].loc[1, 'Windsor_tie']
np.isclose(true, pred, 1e-05)
assert 4.2 in df[('onset', np.nan)].values
assert 1 in df[('duration', np.nan)].values
def test_tensorflow_keras_inception_v3_extractor():
imgs = [join(IMAGE_DIR, f) for f in ['apple.jpg', 'obama.jpg']]
imgs = [ImageStim(im, onset=4.2, duration=1) for im in imgs]
ext = TensorFlowKerasInceptionV3Extractor()
results = ext.transform(imgs)
df = merge_results(results, format='wide', extractor_names='multi')
assert df.shape == (2, 19)
true = 0.9737075
pred = df['TensorFlowKerasInceptionV3Extractor'].loc[0, 'Granny_Smith']
np.isclose(true, pred, 1e-05)
true = 0.64234024
pred = df['TensorFlowKerasInceptionV3Extractor'].loc[1, 'Windsor_tie']
np.isclose(true, pred, 1e-05)
assert 4.2 in df[('onset', np.nan)].values
assert 1 in df[('duration', np.nan)].values
def test_merge_extractor_results():
np.random.seed(100)
image_dir = join(get_test_data_path(), 'image')
stim1 = ImageStim(join(image_dir, 'apple.jpg'))
stim2 = ImageStim(join(image_dir, 'obama.jpg'))
de = DummyExtractor()
de_names = ['Extractor1', 'Extractor2', 'Extractor3']
results = [de.transform(stim1, name) for name in de_names]
results += [de.transform(stim2, name) for name in de_names]
df = merge_results(results)
assert df.shape == (355, 14)
cols = ['onset', 'class', 'filename', 'history', 'stim']
assert df.columns.levels[0].unique().tolist() == de_names + cols
assert df.columns.levels[1].unique().tolist() == ['duration', 0, 1, 2, '']
assert set(df['stim'].unique()) == set(['obama.jpg', 'apple.jpg'])
def run(self, stim, merge=True, **merge_kwargs):
''' Executes the graph by calling all Transformers in sequence.
Args:
stim (str, Stim, list): One or more valid inputs to any
Transformer's 'transform' call.
merge (bool): If True, all results are merged into a single pandas
DataFrame before being returned. If False, a list of
ExtractorResult objects is returned (one per Extractor/Stim
combination).
merge_kwargs: Optional keyword arguments to pass onto the
merge_results() call.
'''
results = list(chain(*[self.run_node(n, stim) for n in self.roots]))
results = list(flatten(results))
self._results = results # For use in plotting
return merge_results(results, **merge_kwargs) if merge else results
def run(self, stim, merge=True, **merge_kwargs):
''' Executes the graph by calling all Transformers in sequence.
Args:
stim (str, Stim, list): One or more valid inputs to any
Transformer's 'transform' call.
merge (bool): If True, all results are merged into a single pandas
DataFrame before being returned. If False, a list of
ExtractorResult objects is returned (one per Extractor/Stim
combination).
merge_kwargs: Optional keyword arguments to pass onto the
merge_results() call.
'''
results = list(chain(*[self.run_node(n, stim) for n in self.roots]))
results = list(flatten(results))
self._results = results # For use in plotting
return merge_results(results, **merge_kwargs) if merge else results
def test_merge_extractor_results_flattened():
np.random.seed(100)
image_dir = join(get_test_data_path(), 'image')
stim1 = ImageStim(join(image_dir, 'apple.jpg'))
stim2 = ImageStim(join(image_dir, 'obama.jpg'))
de_names = ['Extractor1', 'Extractor2', 'Extractor3']
des = [DummyExtractor(name=name) for name in de_names]
results = [de.transform(stim1) for de in des]
results += [de.transform(stim2) for de in des]
df = merge_results(results, flatten_columns=True)
de_cols = ['Extractor1_0', 'Extractor1_1', 'Extractor1_2',
'Extractor2_0', 'Extractor2_1', 'Extractor2_2',
'Extractor3_0', 'Extractor3_1', 'Extractor3_2']
assert df.shape == (354, 16)
cols = ['onset', 'class', 'filename', 'history', 'stim_name', 'duration',
'source_file']
assert set(df.columns.unique().tolist()) == set(cols + de_cols)
def test_tfhub_text_transformer_sentence():
stim = TextStim(join(TEXT_DIR, 'scandal.txt'))
cstim = ComplexTextStim(join(TEXT_DIR, 'wonderful.txt'))
ext = TFHubTextExtractor(ELECTRA_URL,
features='sent_encoding',
preprocessor_url_or_path=TOKENIZER_URL)
res = ext.transform(cstim.elements[:6])
df = merge_results(res, extractor_names=False)
pmod = hub.KerasLayer(TOKENIZER_URL)
mmod = hub.KerasLayer(ELECTRA_URL)
true = mmod(pmod([cstim.elements[5].text]))\
['pooled_output'][0,20].numpy()
assert np.isclose(df['sent_encoding'][5][20], true)
with pytest.raises(ValueError) as err:
TFHubTextExtractor(ELECTRA_URL,
preprocessor_url_or_path=TOKENIZER_URL,
output_key='key').transform(stim)
assert 'Check which keys' in str(err.value)
def test_vectorizer_extractor():
pytest.importorskip('sklearn')
stim = TextStim(join(TEXT_DIR, 'scandal.txt'))
result = TextVectorizerExtractor().transform(stim).to_df()
assert 'woman' in result.columns
assert result['woman'][0] == 3
from sklearn.feature_extraction.text import TfidfVectorizer
custom_vectorizer = TfidfVectorizer()
ext = TextVectorizerExtractor(vectorizer=custom_vectorizer)
stim2 = TextStim(join(TEXT_DIR, 'simple_text.txt'))
result = merge_results(ext.transform([stim, stim2]))
assert ('TextVectorizerExtractor', 'woman') in result.columns
assert 0.129568189476 in result[('TextVectorizerExtractor', 'woman')]
ext = TextVectorizerExtractor(vectorizer='CountVectorizer',
analyzer='char_wb',
ngram_range=(2, 2))
result = ext.transform(stim).to_df()
assert 'wo' in result.columns
assert result['wo'][0] == 6
def test_indico_api_text_extractor():
ext = IndicoAPITextExtractor(api_key=os.environ['INDICO_APP_KEY'],
models=['emotion', 'personality'])
assert ext.validate_keys()
# With ComplexTextStim input
srtfile = join(get_test_data_path(), 'text', 'wonderful.srt')
srt_stim = ComplexTextStim(srtfile, onset=4.2)
result = merge_results(ext.transform(srt_stim), extractor_names=False)
outdfKeysCheck = {
'onset',
'duration',
'order',
'object_id',
'emotion_anger',
'emotion_fear',
'emotion_joy',
'emotion_sadness',
'emotion_surprise',
'personality_openness',
'personality_extraversion',
'personality_agreeableness',
'personality_conscientiousness'}
meta_columns = {'source_file',
'history',
'class',
'filename'}
assert set(result.columns) - set(['stim_name']) == outdfKeysCheck | meta_columns
assert result['onset'][1] == 92.622
# With TextStim input
ts = TextStim(text="It's a wonderful life.")
result = ext.transform(ts).to_df(object_id=True)
assert set(result.columns) == outdfKeysCheck
assert len(result) == 1
ext = IndicoAPITextExtractor(api_key='nogood', models=['language'])
assert not ext.validate_keys()
def test_vectorizer_extractor():
pytest.importorskip('sklearn')
stim = TextStim(join(TEXT_DIR, 'scandal.txt'))
result = TextVectorizerExtractor().transform(stim).to_df()
assert 'woman' in result.columns
assert result['woman'][0] == 3
from sklearn.feature_extraction.text import TfidfVectorizer
custom_vectorizer = TfidfVectorizer()
ext = TextVectorizerExtractor(vectorizer=custom_vectorizer)
stim2 = TextStim(join(TEXT_DIR, 'simple_text.txt'))
result = merge_results(ext.transform([stim, stim2]), format='wide',
extractor_names='multi')
assert ('TextVectorizerExtractor', 'woman') in result.columns
assert np.allclose(0.129568189476,
result[('TextVectorizerExtractor', 'woman')][0])
ext = TextVectorizerExtractor(vectorizer='CountVectorizer',
analyzer='char_wb',
ngram_range=(2, 2))
result = ext.transform(stim).to_df()
assert 'wo' in result.columns
assert result['wo'][0] == 6
def test_merge_extractor_results():
np.random.seed(100)
image_dir = join(get_test_data_path(), 'image')
stim1 = ImageStim(join(image_dir, 'apple.jpg'))
stim2 = ImageStim(join(image_dir, 'obama.jpg'))
de_names = ['Extractor1', 'Extractor2', 'Extractor3']
des = [DummyExtractor(name=name) for name in de_names]
not_features = ['object_id']
for de in des:
not_features.append(de._log_attributes)
results = [de.transform(stim1) for de in des]
results += [de.transform(stim2) for de in des]
df = merge_results(results, format='wide')
assert df.shape == (200, 18)
cols = [
'onset', 'duration', 'order', 'class', 'filename', 'history',
'stim_name', 'source_file'
]
assert not set(cols) - set(df.columns)
assert 'Extractor2#feature_3' in df.columns
df = merge_results(results, format='wide', extractor_names='drop')
assert df.shape == (200, 12)
assert not set(cols) - set(df.columns)
assert 'feature_3' in df.columns
df = merge_results(results, format='wide', extractor_names='multi')
assert df.shape == (200, 18)
_cols = [(c, np.nan) for c in cols]
assert not set(_cols) - set(df.columns)
assert ('Extractor2', 'feature_3') in df.columns
with pytest.raises(ValueError):
merge_results(results, format='long', extractor_names='multi')
df = merge_results(results, format='long', extractor_names='column')
assert df.shape == (1800, 12)
_cols = cols + ['feature', 'extractor', 'value']
assert not set(_cols) - set(df.columns)
row = df.iloc[523, :]
assert row['feature'] == 'feature_2'
assert row['value'] == 475
assert row['extractor'] == 'Extractor1'
assert not set(not_features).intersection(set(df['feature']))
df = merge_results(results, format='long', extractor_names='drop')
assert df.shape == (1800, 11)
assert set(_cols) - set(df.columns) == {'extractor'}
assert not set(not_features).intersection(set(df['feature']))
df = merge_results(results, format='long', extractor_names='prepend')
assert df.shape == (1800, 11)
row = df.iloc[523, :]
assert row['feature'] == 'Extractor1#feature_2'
assert not set(not_features).intersection(set(df['feature']))
df = merge_results(results, format='wide', extractor_params=True)
logattr = {}
for de in des:
logattr[
de.
name] = de._log_attributes #stores log attributes to be found for each extractor
for feat in ['feature_1', 'feature_2', 'feature_3']:
idx_str = f'{de.name}#{feat}#extractor_params'
assert idx_str in df.columns
df_log_attr = json.loads(df[idx_str][0])
for l in logattr[de.name]:
assert l in df_log_attr.keys()
df = merge_results(results, format='long', extractor_params=True)
for idx, row in df.iterrows():
de_name = row['feature'].split('#')[0]
logs = logattr[de_name]
df_logs = row['extractor_params']
for l in logs:
assert l in json.loads(df_logs).keys()
def run(self, stim, merge=True):
results = list(chain(*[self.run_node(n, stim) for n in self.roots]))
results = list(flatten(results))
self._results = results # For use in plotting
return merge_results(results) if merge else results
