[Bug Report] Misleading doc on autoresetting

[vmoens]

Describe the bug

The doc on autoresetting states this:

Disabled Mode

No automatic resetting occurs and users need to manually reset the sub-environment through a mask, env.reset(mask=np.array([True, False, ...], dtype=bool)). The easier way of generating this mask is np.logical_or(terminations, truncations). This makes training code closer to single vector training code, however, can be slower is some cases due to running another function.

import gymnasium as gym
import numpy as np
from collections import deque

# Initialize environment, buffer and episode_start
envs = gym.vector.SyncVectorEnv(
    [lambda: gym.make("CartPole-v1") for _ in range(2)],
    autoreset_mode=gym.vector.AutoresetMode.DISABLED
)
replay_buffer = deque(maxlen=100)

observations, _ = envs.reset()
while True:   # Training loop
    actions = policy(observations)
    next_observations, rewards, terminations, truncations, infos = envs.step(actions)

    # Add to replay buffer
    for i in range(envs.num_envs):
        replay_buffer.append((observations[i], actions[i], rewards[i], terminations[i], next_observations[i]))

    # update observation
    autoreset = np.logical_or(terminations, truncations)
    if np.any(autoreset):
        observations = envs.reset(options={"mask": autoreset})
    else:
        observations = next_observations
envs.close()

This is misleading:

env.reset(mask=np.array([True, False, ...], dtype=bool))

It should be

env.reset(options={"reset_mask": np.array([True, False, ...], dtype=bool)})

This is also wrong:

    if np.any(autoreset):
        observations = envs.reset(options={"mask": autoreset})
    else:
        observations = next_observations

It should be

    if np.any(autoreset):
        observations = envs.reset(options={"reset_mask": autoreset})
    else:
        observations = next_observations

happy to propose a patch, IDK where the doc is hosted though.

Code example

System info

No response

Additional context

No response

Checklist

• I have checked that there is no similar issue in the repo

[pseudo-rnd-thoughts]

Thanks for spotting those errors, the documentation is stored in docs and most of the time uses doc strings in the code.

Could you link the page and I might be able to better direct you.

Checking the blog post, that has the same bug in it that I will look to fix as that is stored in a different repo