[BUG] Minihack-MultiRoom map generation bug (map too large)

[Chocological45]

🐛 Bug

Using any of the Minihack-MultiRoom environments ported from Minigrid results in the level not being created because the map is too large.

To Reproduce

Create and reset the environment

import gym
import minihack
from nle import nethack

MOVE_ACTIONS = tuple(nethack.CompassDirection)
NAVIGATE_ACTIONS = MOVE_ACTIONS + (
            nethack.Command.OPEN,
            nethack.Command.KICK,
)

env = gym.make('MiniHack-MultiRoom-N6-v0, observation_keys=("pixel_crop",), actions=NAVIGATE_ACTIONS)
env.reset()

Occasionally, this error will pop up.

mylevel.des: line 29, pos 0: Map too large at (25 x 22), max is (76 x 21) at "ENDMAP"
mylevel.des: 1 errors detected for level "mylevel". No output created!

I've also come across this error too.

mylevel.des: line 30, pos 0: Map too large at (25 x 22), max is (76 x 21) at "ENDMAP"
mylevel.des: line 32, pos 12: Coordinates (7,21) out of map range!
mylevel.des: 2 errors detected for level "mylevel". No output created!

Expected behavior

The error should not occur and the level is generated for the episode.

Environment

MiniHack version: 0.1.6
NLE version: 0.9.0
Gym version: 0.23.0
PyTorch version: 2.4.0
Is debug build: No
CUDA used to build PyTorch: 12.1

OS: Ubuntu 20.04.6 LTS
GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0
CMake version: version 3.16.3

Python version: 3.9
Is CUDA available: Yes
CUDA runtime version: Could not collect
GPU models and configuration:
GPU 0: NVIDIA A100-PCIE-40GB
MIG 1g.5gb Device 0:
MIG 1g.5gb Device 1:
MIG 1g.5gb Device 2:
MIG 1g.5gb Device 3:
MIG 1g.5gb Device 4:
MIG 1g.5gb Device 5:
MIG 1g.5gb Device 6:
GPU 1: NVIDIA A100-PCIE-40GB
MIG 1g.5gb Device 0:
MIG 1g.5gb Device 1:
MIG 1g.5gb Device 2:
MIG 1g.5gb Device 3:
MIG 1g.5gb Device 4:
MIG 1g.5gb Device 5:
MIG 1g.5gb Device 6:

Nvidia driver version: 535.161.08
cuDNN version: Could not collect

Versions of relevant libraries:
[pip3] numpy==1.24.3
[pip3] torch==2.4.0
[pip3] torchaudio==2.4.0
[pip3] torchvision==0.19.0
[conda] blas 1.0 mkl
[conda] ffmpeg 4.3 hf484d3e_0 pytorch
[conda] libjpeg-turbo 2.0.0 h9bf148f_0 pytorch
[conda] mkl 2021.4.0 h06a4308_640
[conda] mkl-service 2.4.0 py39h7f8727e_0
[conda] mkl_fft 1.3.1 py39hd3c417c_0
[conda] mkl_random 1.2.2 py39h51133e4_0
[conda] pytorch 2.4.0 py3.9_cuda12.1_cudnn9.1.0_0 pytorch
[conda] pytorch-cuda 12.1 ha16c6d3_5 pytorch
[conda] pytorch-mutex 1.0 cuda pytorch
[conda] torchaudio 2.4.0 py39_cu121 pytorch
[conda] torchtriton 3.0.0 py39 pytorch
[conda] torchvision 0.19.0 py39_cu121 pytorch

Additional context

[mahnerak]

This issue is being closed as we're transitioning maintenance - please track the updated status of this issue at fork: samvelyan#5