When obtain_samples is called, the path sampler will generates the minimum number of rollouts such that at least num_steps timesteps

When obtain_samples is called, the path sampler will generates the minimum number of rollouts such that at least num_steps timesteps

Return len(self), which is the number of environments.

The least generic type all column values are convertible to. >>> _column_type(["1", "2"]) is _int_type True >>> _column_type(["1", "2.3"]

Train the discriminator

Implement anything that needs to happen after every rollout.

Return a horizontal line with optional colons to indicate column's alignment (as in `pipe` output format).

Visible width of a printed string. ANSI color codes are removed. >>> _visible_width('\x1b[31mhello\x1b[0m'), _visible_width("world") (5, 5)

Add a transition tuple.

Args: x: torch tensor img (hsv type) Factor: torch tensor with same length as x 0 gives black ima

Args: x: torch tensor img (rgb type) Factor: torch tensor with same length as x 0 gives gray solid image,

Args: x: torch tensor img (hsv type) Factor: torch tensor with same length as x 0 gives black ima

Tie convolutional layers