* Constructors */

* Constructors */

Split a node with m elements into an active node

* Return an always valid pointer to an static empty bucket_entry with * last_bucket() == true. */

Update the parameters with the give gradient Args: grads (List[torch.Tensor]): list of gradients

Decorator to add step information to a function. Args: noise_params (dict): Parameters of the noise distribution.

Returns the total variance of the given data Args: x (torch.Tensor): input data Returns: float: total variance of the given

Send given tensor to gpu if it is available Args: e: (torch.Tensor) Returns: e: (torch.Tensor)

Update the global model.