(self, embed_dim, patch_size=8, channels=3, H=128, W=128)

(self, frame_size=(128, 128), patch_size=8, embed_dim=128, num_heads=8, hidden_dim=256, num_b

(self, params, lr: float, momentum: float = 0.95, backend_steps: int = 5, nesterov: bool = Tr

(self, latent_dim=5, num_bins=4)

(self, frame_size=(128, 128), patch_size=8, embed_dim=128, num_heads=8, hidden_dim=256, num_

(self, frame_size=(128, 128), patch_size=8, embed_dim=128, num_heads=8, hidden_dim=256, num_b

(self, embed_dim, eps=1e-5)

(self, embed_dim, conditioning_dim=None)

(self, embed_dim, num_heads, conditioning_dim=None)

(self, embed_dim, num_heads, causal=True, conditioning_dim=None)

(self, embed_dim, hidden_dim)

(self, embed_dim, hidden_dim, num_experts=4, top_k=2, aux_loss_coeff=0.01, conditioning_dim=N

(self, embed_dim, num_heads, hidden_dim, causal=True, conditioning_dim=None, use_moe=False, n

(self, embed_dim, num_heads, hidden_dim, num_blocks, causal=True, conditioning_dim=None, use_

(self, frame_size=(128, 128), patch_size=4, embed_dim=128, num_heads=8, hidden_dim=128, num_b

(self, frame_size=(128, 128), patch_size=8, embed_dim=128)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(args)

(args)

(args, model_config: Dict[str, Any])

(self)

(self, frames)

(self, tokens)

(self, latents)

(self, frames)

(self, z)

(self, frames)

(self, frames, actions, training=True)

(self, frames)

(self, x)

(self, x)

(self, x, conditioning=None)

(self, x, conditioning=None)

(self, x, conditioning=None)

(self, x, conditioning=None)

(self, x)

(self, x, conditioning=None)

(self, x, conditioning=None)

(self, x, conditioning=None)

(self, frames)

(self, quantized_z)

(idx)

(codebook_usage: float, step: int, model_name: str = "model")

(self)

(self)

(self)

()

(dataloader, num_batches=3, save_dir="batch_visualizations")