trustworthiness#

cuml.metrics.trustworthiness(X, X_embedded, n_neighbors=5, metric='euclidean', convert_dtype=True, 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, optional (default=False)

When set to True, the trustworthiness method will automatically convert the inputs to np.float32.

batch_sizeint (default=512)

The number of samples to use for each batch.

Returns:
trustworthiness scoredouble

Trustworthiness of the low-dimensional embedding