I tried running ddppo baseline for objectnav task using ddppo_objectnav_v2_hm3d_stretch.yaml, and got the following error:
File "/habitat-lab/habitat-baselines/habitat_baselines/rl/ppo/ppo_trainer.py", line 723, in train
self._init_train(resume_state)
File "/habitat-lab/habitat-baselines/habitat_baselines/rl/ppo/ppo_trainer.py", line 318, in _init_train
self._setup_actor_critic_agent(ppo_cfg)
File "/habitat-lab/habitat-baselines/habitat_baselines/rl/ppo/ppo_trainer.py", line 159, in _setup_actor_critic_agent
self.actor_critic = policy.from_config(
File "/habitat-lab/habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py", line 111, in from_config
return cls(
File "/habitat-lab/habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py", line 72, in __init__
PointNavResNetNet(
File "/habitat-lab/habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py", line 265, in __init__
action_space.n + 1, self._n_prev_action
AttributeError: 'Box' object has no attribute 'n'
Exception ignored in: <function VectorEnv.__del__ at 0x7f2d8f3dca60>
Traceback (most recent call last):
File "/habitat-lab/habitat-lab/habitat/core/vector_env.py", line 589, in __del__
File "/habitat-lab/habitat-lab/habitat/core/vector_env.py", line 460, in close
File "/habitat-lab/habitat-lab/habitat/core/vector_env.py", line 129, in __call__
File "/habitat-lab/habitat-lab/habitat/utils/pickle5_multiprocessing.py", line 63, in send
File "/opt/conda/envs/habitat/lib/python3.8/multiprocessing/connection.py", line 200, in send_bytes
File "/opt/conda/envs/habitat/lib/python3.8/multiprocessing/connection.py", line 411, in _send_bytes
File "/opt/conda/envs/habitat/lib/python3.8/multiprocessing/connection.py", line 368, in _send
BrokenPipeError: [Errno 32] Broken pipe
Turns out I mistakenly put _self_ at the start of the default list in ddppo_objectnav_v2_hm3d_stretch.yaml. Putting _self_ back at the last position fixed the problem.
I tried running ddppo baseline for objectnav task using
ddppo_objectnav_v2_hm3d_stretch.yaml
, and got the following error:I've piped stdout and stderr messages to this log file: habitat.log According to https://github.com/openai/gym/issues/839,
Perhaps this is due to version conflicts among habitat repos and openai gym? This is a list of related packages in my conda env: