search

whluo / active_tracking_drl

3 stars 2 forks source link

Errors when run gym_tvizdoom #3

Open Tie-1 opened 2 years ago

Tie-1 commented 2 years ago

I followed install instructions from the [README] After running python train.py --num-workers=2 --env-id=TrackObjSmallMazeRandFlip-v2 --log-dir=save/to --max-global-steps=29999 --val-model-secs=30 --max-val-episodes=2 some errors appeared:

I tensorflow/core/common_runtime/simple_placer.cc:827] global/RNN/while/BasicLSTMCell/split/split_dim: (Const)/job:worker/replica:0/task:1/cpu:0 global/RNN/while/BasicLSTMCell/Linear/concat/concat_dim: (Const): /job:worker/replica:0/task:1/cpu:0 I tensorflow/core/common_runtime/simple_placer.cc:827] global/RNN/while/BasicLSTMCell/Linear/concat/concat_dim: (Const)/job:worker/replica:0/task:1/cpu:0 [2022-04-03 20:38:01,889] Called reset on <Unvectorize<VectorizeFilter[<class 'envs.DiagnosticsInfoI'>]<VizdoomRescale84x84x3<Vectorize<TimeLimit>>>>> before configuring. Configuring automatically with default arguments [2022-04-03 20:38:01,910] Starting training at step=30175 [2022-04-03 20:38:01,923] reached 30175 steps. worker stopped. Error in atexit._run_exitfuncs: Traceback (most recent call last): File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/utils/closer.py", line 67, in close closeable.close() File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/core.py", line 188, in close self.render(close=True) File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/core.py", line 165, in render return self._render(close=close) File "/home/dell/hzt/tracker/gym-tvizdoom/gym_tvizdoom/envs/track_obj_small_maze_rand_flip_v2_env.py", line 122, in _render return self._flip_screen(ret) File "/home/dell/hzt/tracker/gym-tvizdoom/gym_tvizdoom/envs/track_obj_small_maze_rand_flip_v2_env.py", line 103, in _flip_screen return np.fliplr(screen) File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/numpy/lib/twodim_base.py", line 81, in fliplr raise ValueError("Input must be >= 2-d.") ValueError: Input must be >= 2-d. terminate called without an active exception Aborted (core dumped)

Ubuntu 20.04 Python 3.7 tensorflow 0.12 I spent too much time on it, but still not fix it out, can you give me some advice? Thank you!

whluo commented 2 years ago

I followed install instructions from the [README] After running python train.py --num-workers=2 --env-id=TrackObjSmallMazeRandFlip-v2 --log-dir=save/to --max-global-steps=29999 --val-model-secs=30 --max-val-episodes=2 some errors appeared:

I tensorflow/core/common_runtime/simple_placer.cc:827] global/RNN/while/BasicLSTMCell/split/split_dim: (Const)/job:worker/replica:0/task:1/cpu:0 global/RNN/while/BasicLSTMCell/Linear/concat/concat_dim: (Const): /job:worker/replica:0/task:1/cpu:0 I tensorflow/core/common_runtime/simple_placer.cc:827] global/RNN/while/BasicLSTMCell/Linear/concat/concat_dim: (Const)/job:worker/replica:0/task:1/cpu:0 [2022-04-03 20:38:01,889] Called reset on <Unvectorize<VectorizeFilter[<class 'envs.DiagnosticsInfoI'>]<VizdoomRescale84x84x3<Vectorize>>>> before configuring. Configuring automatically with default arguments [2022-04-03 20:38:01,910] Starting training at step=30175 [2022-04-03 20:38:01,923] reached 30175 steps. worker stopped. Error in atexit._run_exitfuncs: Traceback (most recent call last): File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/utils/closer.py", line 67, in close closeable.close() File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/core.py", line 188, in close self.render(close=True) File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/gym/core.py", line 165, in render return self._render(close=close) File "/home/dell/hzt/tracker/gym-tvizdoom/gym_tvizdoom/envs/track_obj_small_maze_rand_flip_v2_env.py", line 122, in _render return self._flip_screen(ret) File "/home/dell/hzt/tracker/gym-tvizdoom/gym_tvizdoom/envs/track_obj_small_maze_rand_flip_v2_env.py", line 103, in _flip_screen return np.fliplr(screen) File "/home/dell/anaconda3/envs/tracker/lib/python3.5/site-packages/numpy/lib/twodim_base.py", line 81, in fliplr raise ValueError("Input must be >= 2-d.") ValueError: Input must be >= 2-d. terminate called without an active exception Aborted (core dumped)

Ubuntu 20.04 Python 3.7 tensorflow 0.12 I spent too much time on it, but still not fix it out, can you give me some advice? Thank you!

Thanks for your interest. Have you strictly followed the instructions in this repo to set up the gym-tvizdoom?

https://bitbucket.org/pengsun000/gym-tvizdoom/src/master/