Cleanup for the environment.

Context manager exit, closes simulation app if created.

output: obs: T, Do action: T, Da

output: obs: key: T, * action: T, Da

This function is used to initialize the configuration.

Initialize the wrapper.

Initialize the environment.

A simple replay buffer that stores transitions in a circular buffer. Supports n-step returns and asymmetric observations. Wh

Initialize EmpiricalNormalization module. Args: shape (int or tuple of int): Shape of input values except batch axis.

Dummy constructor. Use copy_from* and create_from* class methods instead.

@crowsonkb's notes on EMA Warmup: If gamma=1 and power=1, implements a simple average. gamma=1, power=2/3 are good values for mod

Draw a pymunk.Space on a pygame.Surface object. Typical usage:: >>> import pymunk >>> surface = pygame.Surface((10,10))

key_first_k: dict str: int Only take first k data from these keys (to improve perf)

Different from Obs accumulator, the action accumulator allows overwriting previous values.

Valid representations Always use matrix as intermediate representation.

Assumes rgb input: B,C,H,W Assumes low_dim input: B,D

Args: input_shape (tuple, list): shape of input (not including batch dimension) crop_height (int): crop height

Initializes the model. Parameters: backbones: torch module of the backbone to be used. See backbone.py transformer: t