def load(fname, verbose=0):
data = utils.parse_hvsrpy_output(fname)
std_curve = np.log(data["upper"]) - np.log(data["curve"])
clarity = utils.sesame_clarity(data["frequency"],
data["curve"],
std_curve,
data["std_f0"],
verbose=verbose)
reliability = utils.sesame_reliability(data["windowlength"],
data["accepted_windows"],
data["frequency"],
data["curve"],
std_curve,
verbose=verbose)
return (reliability, clarity)
def test_parse_hvsrpy_output(self):
def compare_data_dict(expected_dict, returned_dict):
for key, value in expected_dict.items():
if isinstance(value, (str, int, bool)):
self.assertEqual(expected_dict[key], returned_dict[key])
elif isinstance(value, float):
self.assertAlmostEqual(expected_dict[key],
returned_dict[key],
places=3)
elif isinstance(value, np.ndarray):
self.assertArrayAlmostEqual(expected_dict[key],
returned_dict[key],
places=3)
else:
raise ValueError
# Ex 0
fname = self.full_path + "data/utils/ex0.hv"
expected = {
"windowlength": 60.0,
"total_windows": 30,
"rejection_bool": True,
"n_for_rejection": 2.0,
"accepted_windows": 29,
"distribution_f0": "normal",
"mean_f0": 0.6976,
"std_f0": 0.1353,
"distribution_mc": "log-normal",
"f0_mc": 0.7116,
"amplitude_f0_mc": 3.8472,
}
names = ["frequency", "curve", "lower", "upper"]
df = pd.read_csv(fname, comment="#", names=names)
for name in names:
expected[name] = df[name].to_numpy()
returned = utils.parse_hvsrpy_output(fname)
compare_data_dict(expected, returned)
# Ex 1
fname = self.full_path + "data/utils/ex1.hv"
expected = {
"windowlength": 60.0,
"total_windows": 60,
"rejection_bool": True,
"n_for_rejection": 2.0,
"accepted_windows": 48,
"distribution_f0": "normal",
"mean_f0": 0.7155,
"std_f0": 0.0759,
"distribution_mc": "log-normal",
"f0_mc": 0.7378,
"amplitude_f0_mc": 3.9661,
}
names = ["frequency", "curve", "lower", "upper"]
df = pd.read_csv(fname, comment="#", names=names)
for name in names:
expected[name] = df[name].to_numpy()
returned = utils.parse_hvsrpy_output(fname)
compare_data_dict(expected, returned)
# Ex 2
fname = self.full_path + "data/utils/ex2.hv"
expected = {
"windowlength": 60.0,
"total_windows": 30,
"rejection_bool": True,
"n_for_rejection": 2.0,
"accepted_windows": 29,
"distribution_f0": "normal",
"mean_f0": 0.7035,
"std_f0": 0.1391,
"distribution_mc": "log-normal",
"f0_mc": 0.7116,
"amplitude_f0_mc": 3.9097,
}
names = ["frequency", "curve", "lower", "upper"]
df = pd.read_csv(fname, comment="#", names=names)
for name in names:
expected[name] = df[name].to_numpy()
returned = utils.parse_hvsrpy_output(fname)
compare_data_dict(expected, returned)
# Ex 3
fname = self.full_path + "data/utils/ex3.hv"
expected = {
"windowlength": 60.0,
"total_windows": 60,
"rejection_bool": True,
"n_for_rejection": 2.0,
"accepted_windows": 33,
"distribution_f0": "normal",
"mean_f0": 0.7994,
"std_f0": 0.0365,
"distribution_mc": "log-normal",
"f0_mc": 0.7933,
"amplitude_f0_mc": 4.8444,
}
names = ["frequency", "curve", "lower", "upper"]
df = pd.read_csv(fname, comment="#", names=names)
for name in names:
expected[name] = df[name].to_numpy()
returned = utils.parse_hvsrpy_output(fname)
compare_data_dict(expected, returned)