def test_trustworthiness(input_type, n_samples, n_features, n_components):
centers = round(n_samples*0.4)
X, y = make_blobs(n_samples=n_samples, centers=centers,
n_features=n_features)
X_embedded = \
UMAP(n_components=n_components).fit_transform(X)
X = X.astype(np.float32)
X_embedded = X_embedded.astype(np.float32)
if input_type == 'dataframe':
gdf = cudf.DataFrame()
for i in range(X.shape[1]):
gdf[str(i)] = np.asarray(X[:, i], dtype=np.float32)
gdf_embedded = cudf.DataFrame()
for i in range(X_embedded.shape[1]):
gdf_embedded[str(i)] = np.asarray(X_embedded[:, i],
dtype=np.float32)
score = cuml_trustworthiness(gdf, gdf_embedded)
else:
score = cuml_trustworthiness(X, X_embedded)
sk_score = sklearn_trustworthiness(X, X_embedded)
eps = 0.001
assert (sk_score * (1 - eps) <= score and
score <= sk_score * (1 + eps))
def test_trustworthiness(input_type, n_samples, n_features, n_components,
batch_size):
centers = round(n_samples*0.4)
X, y = make_blobs(n_samples=n_samples, centers=centers,
n_features=n_features, random_state=32)
X_embedded = \
UMAP(n_components=n_components, random_state=32).fit_transform(X)
X = X.astype(np.float32)
X_embedded = X_embedded.astype(np.float32)
sk_score = sklearn_trustworthiness(X, X_embedded)
if input_type == 'dataframe':
X = cudf.DataFrame(X)
X_embedded = cudf.DataFrame(X_embedded)
score = cuml_trustworthiness(X, X_embedded, batch_size=batch_size)
assert abs(score - sk_score) <= 1e-3
