def test_init(self):
try:
with open(CONFIG_PATH) as config_file:
config = json.loads(config_file.read())
MiniBatchGen.from_config(config)
except Exception as e:
pytest.fail("Unexpected Error: {}".format(e))
def test_consuming_minibatch(self):
with open(CONFIG_PATH) as config_file:
config = json.loads(config_file.read())
mb_gen_dict = MiniBatchGen.from_config(config)
try:
if not os.path.exists(FEATURE_ROOT):
os.makedirs(FEATURE_ROOT)
for _, mb_gen in mb_gen_dict.items():
mb_gen_e = mb_gen.execute()
next(mb_gen_e)
except Exception as e:
pytest.fail("Unexpected Error: {}".format(e))
finally:
shutil.rmtree(FEATURE_ROOT)
def test_data_multiple_features(self):
"""
Compare data from "manual" constructor to config file-based constructor.
"""
# construct minibatch generator "manually"
sample_rate = 22050
win_size = 256
hop_size = 128
energy_threshold = 0.2
spectral_flatness_threshold = 0.3
seg_duration = 0.2
seg_overlap = 0.5
batch_size = 50
num_features = 64
n_time_bins = 34
af_gen = AudioFrameGen(sample_rate=sample_rate,
win_size=win_size,
hop_size=hop_size)
en_ext = EnergyExtractor()
sf_ext = SpectralFlatnessExtractor()
mel_ext = MelSpectrumExtractor(sample_rate=sample_rate,
fft_size=win_size,
n_mels=64,
min_freq=0,
max_freq=sample_rate / 2)
ff_pro = FrameFeatureProcessor(af_gen, [en_ext, sf_ext, mel_ext],
feature_container_root=FEATURE_ROOT)
ffc_ext = FrameFeatureChunkExtractor("mel_spectrum")
act_det = Simple(
energy_threshold=energy_threshold,
spectral_flatness_threshold=spectral_flatness_threshold)
sf_pro = SegmentFeatureProcessor([act_det, ffc_ext],
ff_pro=ff_pro,
audio_root=DATA_PATH)
parser = CSVFileLabelParser(TEST_FILE2LABEL_PATH,
label_file=TEST_LABEL_PATH)
sc_gen = SegmentContainerGenerator(DATA_PATH,
sf_pro,
label_parser=parser,
seg_duration=seg_duration,
seg_overlap=seg_overlap)
mb_gen_1 = MiniBatchGen(
sc_gen, batch_size, {
"mel_spectrum": {
"feature_size": num_features,
"n_time_bins": n_time_bins
}
}, 0)
# parse json file
with open(CONFIG_PATH) as config_file:
config = json.loads(config_file.read())
config["features"] = [{
'name': 'audio_chunk',
'config': {}
}, {
'name': 'mel_spectrum',
'config': {
'n_mels': 64,
'min_freq': 0,
'max_freq': 11025,
'log_amp': 1
}
}]
# construct minibatch generator from config
mb_gen_dict = MiniBatchGen.from_config(config)
mb_gen_2 = mb_gen_dict["default"]
# execute and compare
try:
mb_gen_1_e = mb_gen_1.execute(active_segments_only=True,
with_targets=True,
with_filenames=False)
mb_gen_2_e = mb_gen_2.execute(active_segments_only=True,
with_targets=True,
with_filenames=False)
if not os.path.exists(FEATURE_ROOT):
os.makedirs(FEATURE_ROOT)
for mb1, mb2 in zip(mb_gen_1_e, mb_gen_2_e):
assert np.all(mb1[0]["mel_spectrum"] == mb2[0]["mel_spectrum"])
assert np.all(mb1[1] == mb2[1])
except Exception as e:
pytest.fail("Unexpected Error: {}".format(e))
finally:
shutil.rmtree(FEATURE_ROOT)
def test_audio_data(self):
"""
Compare data from "manual" constructor to config file-based constructor,
with raw audio feature.
"""
# construct minibatch generator "manually"
sample_rate = 22050
seg_duration = 0.2
seg_overlap = 0.5
batch_size = 50
num_features = 1
n_time_bins = 4410
ac_ext = AudioChunkExtractor(DATA_PATH, sample_rate)
sf_pro = SegmentFeatureProcessor([ac_ext],
ff_pro=None,
audio_root=DATA_PATH)
parser = CSVFileLabelParser(TEST_FILE2LABEL_PATH,
label_file=TEST_LABEL_PATH)
sc_gen = SegmentContainerGenerator(DATA_PATH,
sf_pro,
label_parser=parser,
seg_duration=seg_duration,
seg_overlap=seg_overlap)
mb_gen_1 = MiniBatchGen(
sc_gen, batch_size, {
"audio_chunk": {
"feature_size": num_features,
"n_time_bins": n_time_bins
}
}, 0)
# parse json file
with open(CONFIG_PATH) as config_file:
config = json.loads(config_file.read())
# replace feature by audio_chunk and remove activity detection
config["features"] = [{"name": "audio_chunk", "config": {}}]
config.pop("activity_detection")
# construct minibatch generator from config
mb_gen_dict = MiniBatchGen.from_config(config)
mb_gen_2 = mb_gen_dict["default"]
# execute and compare
mb_gen_1_e = mb_gen_1.execute(active_segments_only=False,
with_targets=True,
with_filenames=True)
mb_gen_2_e = mb_gen_2.execute(active_segments_only=False,
with_targets=True,
with_filenames=True)
for mb1, mb2 in zip(mb_gen_1_e, mb_gen_2_e):
assert np.all(mb1[0]["audio_chunk"] == mb2[0]["audio_chunk"])
assert np.all(mb1[1] == mb2[1])
assert np.all(mb1[2] == mb2[2])