def test_groupwise():
from sklearn.preprocessing import scale
coord_1 = ["a"] * 51 + ["b"] * 49
coord_2 = list(range(10)) * 10
X_ds = xr.Dataset(
{"var_1": (["sample", "feature"], np.random.random((100, 10)))},
coords={
"sample": range(100),
"feature": range(10),
"coord_1": (["sample"], coord_1),
"coord_2": (["sample"], coord_2),
},
)
# test wrapped sklearn estimator
preprocess(X_ds, scale, groupby="coord_1")
# test newly defined estimator
Xt_ds2, estimator = split(
X_ds,
new_dim="split_sample",
new_len=5,
groupby="coord_1",
keep_coords_as="initial_sample",
return_estimator=True,
)
assert Xt_ds2.var_1.shape == (19, 10, 5)
Xt_ds2 = estimator.inverse_transform(Xt_ds2)
assert Xt_ds2.var_1.shape == (95, 10)
def test_preprocess():
from sklearn.preprocessing import scale
X_da = xr.DataArray(
np.random.random((100, 10)),
coords={'sample': range(100), 'feature': range(10)},
dims=('sample', 'feature')
)
Xt_da_gt = X_da
Xt_da_gt.data = scale(X_da)
Xt_da = preprocess(X_da, scale)
xrt.assert_allclose(Xt_da, Xt_da_gt)
X_ds = xr.Dataset(
{'var_1' : (['sample', 'feature'], np.random.random((100, 10)))},
coords={'sample': range(100), 'feature': range(10)}
)
Xt_ds = preprocess(X_ds, scale)
xrt.assert_allclose(Xt_ds, X_ds.apply(scale))
def test_preprocess():
from sklearn.preprocessing import scale
X_da = xr.DataArray(
np.random.random((100, 10)),
coords={
"sample": range(100),
"feature": range(10)
},
dims=("sample", "feature"),
)
Xt_da_gt = X_da
Xt_da_gt.data = scale(X_da)
Xt_da = preprocess(X_da, scale)
xrt.assert_allclose(Xt_da, Xt_da_gt)
X_ds = xr.Dataset(
{"var_1": (["sample", "feature"], np.random.random((100, 10)))},
coords={
"sample": range(100),
"feature": range(10)
},
)
Xt_ds = preprocess(X_ds, scale)
xrt.assert_allclose(Xt_ds, X_ds.apply(scale))
def test_groupwise():
from sklearn.preprocessing import scale
coord_1 = ['a'] * 51 + ['b'] * 49
coord_2 = list(range(10)) * 10
X_ds = xr.Dataset(
{'var_1': (['sample', 'feature'], np.random.random((100, 10)))},
coords={
'sample': range(100),
'feature': range(10),
'coord_1': (['sample'], coord_1),
'coord_2': (['sample'], coord_2)
})
# test wrapped sklearn estimator
preprocess(X_ds, scale, groupby='coord_1')
# test newly defined estimator
Xt_ds2, estimator = split(X_ds,
new_dim='split_sample',
new_len=5,
groupby='coord_1',
keep_coords_as='initial_sample',
return_estimator=True)
assert Xt_ds2.var_1.shape == (19, 10, 5)
Xt_ds2 = estimator.inverse_transform(Xt_ds2)
assert Xt_ds2.var_1.shape == (95, 10)