search

yxKryptonite / OpenFMNav

Official implementation of OpenFMNav: Towards Open-Set Zero-Shot Object Navigation via Vision-Language Foundation Models
https://yxkryptonite.github.io/OpenFMNav
MIT License
21 stars 1 forks source link

Dimension mismatch problem #5

Closed Rh-Dang closed 3 months ago

Rh-Dang commented 4 months ago

Traceback (most recent call last): File "/home/pai/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap self.run() File "/home/pai/lib/python3.9/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/mnt/workspace/OpenFMNav/envs/habitat/utils/vector_env.py", line 202, in _worker_env observations = env.reset() File "/mnt/workspace/OpenFMNav/agents/sem_exp.py", line 113, in reset obs, info = super().reset() File "/mnt/workspace/OpenFMNav/envs/habitat/objectgoal_env21.py", line 154, in reset state = np.concatenate((rgb, depth, semantic), axis=2).transpose(2, 0, 1) ValueError: all the input arrays must have same number of dimensions, but the array at index 0 has 3 dimension(s) and the array at index 2 has 4 dimension(s)

The above is my error. this looks like the demension is mismatched. So I printed out their dimensions.

rgb shape: , (480, 640, 3) depth shape: , (480, 640, 1) semantic shape: , (480, 640, 1, 1)

It seems that the semantic tensor has one more dimension, so how do I solve this problem?

yxKryptonite commented 4 months ago

I didn't come across this issue in my environment. Maybe can you try to deduct one dimension by yourself and see how things go?

yxKryptonite commented 3 months ago

closed if no further questions