garage.np.regressors package

Regressors which use NumPy as a numerical backend.

class ProductRegressor(regressors)[source]

Bases: object

A class for performing MLE regression by fitting a product distribution to the outputs. A separate regressor will be trained for each individual input distribution.

fit(xs, ys)[source]
get_param_values(**tags)[source]
predict(xs)[source]
predict_log_likelihood(xs, ys)[source]
sample_predict(xs)[source]
set_param_values(flattened_params, **tags)[source]