Submit a single or batch tensor to refine the streaming mean.

Read the duration since the last PING was written. Returns: Tuple of worker who wrote the last ping and duration until then. Raises:

Clips the current episode.

See predict() above for documentation.

Compute the divergence measure between two states.

We expand a node using the value, reward and policy prediction obtained from the neural network.

Finalizes the network. Arguments: optimizer: (tf.train.Optimizer) An optimizer class from those available at tf.train.Optimizer.

Returns list of observations for each agent.

Get the (transformed) mode of the distribution. Borrowed from https://github.com/HorizonRobotics/alf/blob/0f8d0ec5d60ef6f30307c6a66ba388852e8c

Returns the all attributes. This method returns the all attributes including ones in subclasses. Some of scikit-learn utilities will

Returns the variables of this layer.

Checks if the target_pos is in the firing range(5x5) :param source_pos: Coordinates of the source :param target_pos: Coordin

Registers scaler class. Args: cls: scaler class inheriting ``Scaler``.

Run example locally, potentially parallelizing across cpus/gpus.

Saves all information required to recreate this model in two files in savedir (or self.model_dir if savedir is None), one containing the model

Apply action to the game. Args: action : action of the action_space to take. Returns: The n