GaussianMixture.predict_proba

GaussianMixture.predict_proba(data)[source]

Computes a distribution over the components for each of the provided datapoints.

Parameters

data (Union[ndarray[Any, dtype[float32]], Tensor]) -- The datapoints for which to compute the component assignment probabilities.

Return type

Tensor

Returns

A tensor of shape [num_datapoints, num_components] with the assignment probabilities for each component and datapoint. Note that each row of the vector sums to 1, i.e. the returned tensor provides a proper distribution over the components for each datapoint.

Attention

When calling this function in a multi-process environment, each process receives only a subset of the predictions. If you want to aggregate predictions, make sure to gather the values returned from this method.