def test_encodeSplines():
x = np.arange(100)
assert np.allclose(encodeSplines(x, start=-1, end=50).sum(2), 1)
assert np.allclose(encodeSplines(x, start=-1, end=120).sum(2), 1)
assert np.allclose(encodeSplines(x, start=10, end=120).sum(2), 1)
assert np.allclose(encodeSplines(x).sum(2), 1)
def test_EncodeSplines_comparison():
x = np.random.normal(size=(100, 3, 2))
x1t = encodeSplines(x[..., 0])
x2t = encodeSplines(x[..., 1])
es = EncodeSplines()
es.fit(x)
x_all = es.transform(x)
assert np.allclose(x_all[..., 0, :], x1t)
assert np.allclose(x_all[..., 1, :], x2t)
assert np.allclose(x_all[..., 1, :].sum(-1), 1)
assert es.data_min_[0] == x[..., 0].min()
assert es.data_max_[1] == x[..., 1].max()
def transform(self, x):
# impute missing values and rescale the distances
xnew = self.minmax_scaler.transform(self.funct_transform.transform(self.imp.transform(x)))
# convert distances to spline bases
dist = {"dist_" + k: encodeSplines(xnew[:, i, np.newaxis], start=0, end=1, warn=False)
for i, k in enumerate(self.POS_FEATURES)}
return dist
def transform(self, x):
# impute missing values and rescale the distances
xnew = self.preproc_pipeline.transform(self.imp.transform(x))
# convert distances to spline bases
dist = {
"dist_" + k: encodeSplines(xnew[:, i, np.newaxis], start=0, end=1)
for i, k in enumerate(self.POS_FEATURES)
}
return dist
def encodeSplines_common(x, pos_range, n_bases, spline_order):
x_processed = {
k: fill_nan(
encodeSplines(v,
n_bases=n_bases,
spline_order=spline_order,
start=pos_range[k]["min"],
end=pos_range[k]["max"]))
for k, v in x.items()
}
return x_processed
