def test_selection_by_mask(self):
""" Test mask-based Mappable selection.
Only one peak in the test dataset (in study5) should be within the sgACC. """
ids = self.dataset.get_ids_by_mask(
get_test_data_path() + 'sgacc_mask.nii.gz')
self.assertEquals(len(ids), 1)
self.assertEquals('study5', ids[0])
def test_selection_by_mask(self):
""" Test mask-based Mappable selection.
Only one peak in the test dataset (in study5) should be within the sgACC. """
ids = self.dataset.get_ids_by_mask(get_test_data_path() +
'sgacc_mask.nii.gz')
self.assertEquals(len(ids), 1)
self.assertEquals('study5', ids[0])
def test_audio_stim(self):
audio_dir = join(get_test_data_path(), 'audio')
stim = AudioStim(join(audio_dir, 'barber.wav'))
self.assertEquals(stim.sampling_rate, 11025)
ann = STFTAnnotator(frame_size=1., spectrogram=False,
bins=[(100, 300), (300, 3000), (3000, 20000)])
timeline = stim.annotate([ann])
df = timeline.to_df('long')
self.assertEquals(df.shape, (1671, 4))
def test_complex_text_stim(self):
text_dir = join(get_test_data_path(), 'text')
stim = ComplexTextStim(join(text_dir, 'complex_stim_no_header.txt'),
columns='ot', default_duration=0.2)
self.assertEquals(len(stim.elements), 4)
self.assertEquals(stim.elements[2].onset, 34)
self.assertEquals(stim.elements[2].duration, 0.2)
stim = ComplexTextStim(join(text_dir, 'complex_stim_with_header.txt'))
self.assertEquals(len(stim.elements), 4)
self.assertEquals(stim.elements[2].duration, 0.1)
def test_text_annotation(self):
text_dir = join(get_test_data_path(), 'text')
stim = ComplexTextStim(join(text_dir, 'sample_text.txt'),
columns='to', default_duration=1)
td = TextDictionaryAnnotator(join(text_dir,
'test_lexical_dictionary.txt'),
variables=['length', 'frequency'])
self.assertEquals(td.data.shape, (7, 2))
timeline = stim.annotate([td])
df = TimelineExporter.timeline_to_df(timeline)
self.assertEquals(df.shape, (12, 4))
self.assertEquals(df.iloc[9, 3], 10.6)
def test_full_pipeline(self):
''' Smoke test of entire pipeline, from stimulus loading to
event file export. '''
stim = VideoStim(join(get_test_data_path(), 'video', 'small.mp4'))
annotators = [DenseOpticalFlowAnnotator(), FaceDetectionAnnotator()]
timeline = stim.annotate(annotators, show=False)
exp = FSLExporter()
tmpdir = tempfile.mkdtemp()
exp.export(timeline, tmpdir)
from glob import glob
files = glob(join(tmpdir, '*.txt'))
self.assertEquals(len(files), 2)
shutil.rmtree(tmpdir)
def test_video_stim(self):
''' Test VideoStim functionality. '''
filename = join(get_test_data_path(), 'video', 'small.mp4')
video = VideoStim(filename)
self.assertEquals(video.fps, 30)
self.assertEquals(video.n_frames, 166)
self.assertEquals(video.width, 560)
# Test frame iterator
frames = [f for f in video]
self.assertEquals(len(frames), 166)
f = frames[100]
self.assertIsInstance(f, VideoFrameStim)
self.assertIsInstance(f.onset, float)
self.assertEquals(f.data.shape, (320, 560, 3))
#-- Load params
split_name = str(args.split)
test_iters = args.test_iters
test_model = args.test_model
snapshot = args.snapshot
experiment = args.experiment
gauss_var = args.gauss_var
#--- GPU
caffe.set_mode_gpu()
#--- LOAD SMOTHED POSITION MAPS
position_maps = utils.load_position_maps(split_name, gauss_var)
#--- LOAD TEST DATA
test_data = utils.get_test_data_path(split_name)
#--- GET TEST RESULTS PATH
test_res_path = utils.get_result_path(experiment, split_name)
###--- TEST
print 'Testing'
sys.stdout.flush()
net_results, position_results = test_net(test_model, snapshot, test_data, test_iters, position_maps)
im_acc, price_acc, name_acc = net_results
print 'NET: image accuracy:', im_acc
print 'NET: price accuracy:', price_acc
print 'NET: name accuracy:', name_acc
import pandas as pd
from utils import get_test_data_path
import os
from os.path import join
data_dir = get_test_data_path()
zero_csv_file = join(data_dir, "csv_files_zero")
if not os.path.exists(zero_csv_file):
os.makedirs(zero_csv_file)
files = join(data_dir, "csv_files")
print(files)
for filename in os.listdir(files):
if "presentation" in filename:
csv_file = join(files, filename)
df = pd.read_csv(csv_file)
lst = [0]
for row in range(len(df["ImageDisplay1.OffsetTime"])):
try:
time_difference = df["ImageDisplay1.OffsetTime"][row] - df[
"ImageDisplay1.OffsetTime"][row - 2]
if time_difference > 7500 or row == 0:
lst += [row]
print(lst)
except:
continue
