StanfordVL / GibsonEnv

Gibson Environments: Real-World Perception for Embodied Agents
http://gibsonenv.stanford.edu/
MIT License
857 stars 146 forks source link

Problems on running the example of env_example.py #124

Closed KARAMKIM88 closed 4 years ago

KARAMKIM88 commented 4 years ago

Hi I'm really interested in your package

Error message is

File "env_example.py", line 27, in main() File "env_example.py", line 17, in main nav_env.reset() File "/iGibson/gibson2/envs/locomotor_env.py", line 780, in reset state = super(NavigateRandomEnv, self).reset() File "/iGibson/gibson2/envs/locomotor_env.py", line 697, in reset self.reset_agent() File "/iGibson/gibson2/envs/locomotor_env.py", line 579, in reset_agent self.reset_initial_and_target_pos() File "/iGibson/gibson2/envs/locomotor_env.py", line 758, in reset_initial_and_targetpos , dist = self.get_shortest_path(from_initial_pos=True) File "/iGibson/gibson2/envs/locomotor_env.py", line 438, in get_shortest_path return self.scene.get_shortest_path(self.floor_num, source, target, entire_path=entire_path) File "/iGibson/gibson2/core/physics/scene.py", line 337, in get_shortest_path source_map = tuple(self.world_to_map(source_world)) File "/iGibson/gibson2/core/physics/scene.py", line 328, in world_to_map return np.flip((xy / self.trav_map_resolution + self.trav_map_size / 2.0)).astype(np.int) TypeError: flip() missing 1 required positional argument: 'axis'

I think that my numpy package's version is different with yours

mine is '1.14.3'

What version do you use?

KARAMKIM88 commented 4 years ago

I solved the problem I upgraded numpy- 1.19 it didn't happen