No module named ‘omni.isaac.core’

Omniverse Isaac Sim
1

Hi,
I’m new to Isaaac. I try to import parse_env_cfg by from omni.isaac.lab_tasks.utils import parse_env_cfg. However, I get No module named ‘omni.isaac.core’. Here is the full error:

from omni.isaac.lab_tasks.utils import parse_env_cfg
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/__init__.py", line 25, in <module>
    from .utils import import_packages
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/utils/__init__.py", line 9, in <module>
    from .parse_cfg import get_checkpoint_path, load_cfg_from_registry, parse_env_cfg
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/utils/parse_cfg.py", line 16, in <module>
    from omni.isaac.lab.envs import DirectRLEnvCfg, ManagerBasedRLEnvCfg
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/envs/__init__.py", line 45, in <module>
    from . import mdp, ui
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/envs/mdp/__init__.py", line 18, in <module>
    from .actions import *  # noqa: F401, F403
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/envs/mdp/actions/__init__.py", line 8, in <module>
    from .actions_cfg import *
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/envs/mdp/actions/actions_cfg.py", line 9, in <module>
    from omni.isaac.lab.managers.action_manager import ActionTerm, ActionTermCfg
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/managers/__init__.py", line 13, in <module>
    from .action_manager import ActionManager, ActionTerm
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/managers/action_manager.py", line 20, in <module>
    from omni.isaac.lab.assets import AssetBase
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/assets/__init__.py", line 41, in <module>
    from .articulation import Articulation, ArticulationCfg, ArticulationData
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/assets/articulation/__init__.py", line 8, in <module>
    from .articulation import Articulation
  File "/home/uchi/IsaacLab/source/extensions/omni.isaac.lab/omni/isaac/lab/assets/articulation/articulation.py", line 17, in <module>
    import omni.isaac.core.utils.stage as stage_utils
ModuleNotFoundError: No module named 'omni.isaac.core'

I’m not sure how I’m getting this error if I install it correctly. I also from isaacsim import SimulationApp and simulation_app = SimulationApp({"headless": True}) before env_cfg = parse_env_cfg(env_name, use_gpu=True, num_envs=num_envs).

Does anyone have any idea how to debug it?

2

Are you using the python in the isaac sim installation to run your program and not the regular python installed on your computer?

like

path_to_isaac_sim/python.sh simulation.py

3

I have my py file under tutorials file in Isaaclab and run ./isaaclab.sh -p source/standalone/tutorials/test.py . I also setup my vscode with isaac lab and try to run it on vscode, following this https://isaac-sim.github.io/IsaacLab/source/overview/developer-guide/vs_code.html. I’m not sure that is a correct way

4

Hi @UchiN

Could you please share your test script?

5 
import torch
from torchrl.modules import LSTMModule
from torchrl.envs import ParallelEnv, DoubleToFloat, Compose, ObservationNorm, StepCounter, TransformedEnv, InitTracker
from torchrl.envs.libs.orbit import OrbitEnv
from omni.isaac.kit import SimulationApp
from omni.isaac.lab_tasks.utils import parse_env_cfg

env_name = "Isaac-Cartpole-v0"
device = "cpu" if not torch.has_cuda else "cuda:0"
num_envs = 2
simulation_app = SimulationApp({"headless": False}
)
env_cfg = parse_env_cfg(env_name, use_gpu=True, num_envs=num_envs)

def base_env_maker(env_name, env_cfg, device):
    base_env = OrbitEnv(
        env_name,
        device=device,
        cfg=env_cfg,
    )

    return base_env

def create_rhs_transform(in_size, num_cells, device):
    lstm = LSTMModule(
        input_size=in_size,
        hidden_size=num_cells,
        device=device,
        in_key="observation",
        out_key="rnn_out",
    )
    return lstm.make_tensordict_primer()

def env_maker():
    env = TransformedEnv(
        env=base_env_maker(env_name, env_cfg, device),
        transform=
            Compose(
                InitTracker(),
                DoubleToFloat(
                    in_keys=["observation"],
                ),
                StepCounter(),
                ObservationNorm(in_keys=["observation"]),
            )   
        )
    env.transform[-1].init_stats(10, reduce_dim=0)
    env.append_transform(create_rhs_transform(
                    in_size=11,
                    num_cells=5,
                    device=device,
                ))
    return env

env = env_maker()
td = env.reset()