def test_featurize_files_function(tmpdir):
"""Test featurize function for on-disk time series"""
with sample_ts_files(size=4, labels=['A', 'B']) as ts_paths:
fset, labels = featurize.featurize_ts_files(
ts_paths, features_to_use=["std_err"], scheduler=dask.get)
assert "std_err" in fset
assert fset.shape == (4, 1)
npt.assert_array_equal(labels, ['A', 'B', 'A', 'B'])
def test_featurize_files_function(tmpdir):
"""Test featurize function for on-disk time series"""
with sample_ts_files(size=4) as ts_paths:
fset = featurize.featurize_ts_files(ts_paths,
features_to_use=["std_err"],
scheduler=get_sync)
assert "std_err" in fset
assert fset.shape == (4, 1)
def test_featurize_files_function(tmpdir):
"""Test featurize function for on-disk time series"""
with sample_ts_files(size=4, labels=['A', 'B']) as ts_paths:
fset, labels = featurize.featurize_ts_files(ts_paths,
features_to_use=["std_err"],
scheduler=dask.get)
assert "std_err" in fset
assert fset.shape == (4, 1)
npt.assert_array_equal(labels, ['A', 'B', 'A', 'B'])
def test_featurize_files_function_regression_data():
"""Test featurize function for on-disk time series - regression data"""
fset_path = pjoin(TEMP_DIR, 'output_featureset.nc')
with sample_ts_files(size=4, targets=[1.0, 2.0]) as ts_paths:
fset = featurize.featurize_ts_files(ts_paths,
features_to_use=["std_err"],
output_path=fset_path,
scheduler=get_sync)
assert("std_err" in fset.data_vars)
assert(all(target in [1.0, 2.0] for target in fset['target'].values))
def test_featurize_files_function():
"""Test featurize function for on-disk time series"""
fset_path = pjoin(TEMP_DIR, 'output_featureset.nc')
with sample_ts_files(size=4, targets=['class1', 'class2']) as ts_paths:
fset = featurize.featurize_ts_files(ts_paths,
features_to_use=["std_err"],
output_path=fset_path,
scheduler=get_sync)
assert("std_err" in fset.data_vars)
assert(all(class_name in ['class1', 'class2']
for class_name in fset['target'].values))