trustworthiness#

cuml.metrics.trustworthiness(
X,
X_embedded,
n_neighbors=5,
metric='euclidean',
convert_dtype='deprecated',
batch_size=512,
) float[source]#

Expresses to what extent the local structure is retained in embedding. The score is defined in the range [0, 1].

Parameters:
Xarray-like (device or host) shape = (n_samples, n_features)

Acceptable formats: cuDF DataFrame, NumPy ndarray, Numba device ndarray, cuda array interface compliant array like CuPy

X_embeddedarray-like (device or host) shape= (n_samples, n_features)

Acceptable formats: cuDF DataFrame, NumPy ndarray, Numba device ndarray, cuda array interface compliant array like CuPy

n_neighborsint, optional (default=5)

Number of neighbors considered

metricstr in [‘euclidean’] (default=’euclidean’)

Metric used to compute the trustworthiness. For the moment only ‘euclidean’ is supported.

convert_dtypebool, default=”deprecated”

Deprecated since version 26.08: convert_dtype was deprecated in version 26.08 and will be removed in version 26.10. cuML only copies input arrays when necessary (e.g. to unify dtypes), there is no reason to provide this keyword going forward.

batch_sizeint (default=512)

The number of samples to use for each batch.

Returns:
trustworthiness scoredouble

Trustworthiness of the low-dimensional embedding