환경에 따라 충돌이 많이 발생하는것을 몸소 체험함. 현재 나는 아래와 같은 스펙으로 진행했음
conda create -n Isaaclab_env python=3.12 -y
conda activate Isaaclab_env
python --version
which python
python -m pip install --upgrade pip
nvidia-smi
conda env list
=> Python 3.12.x
=> /home/ryung/miniconda3/envs/Isaaclab_env/bin/python
cd ~/Desktop
git clone --branch v3.0.0-beta2.patch1 --depth 1 \
<https://github.com/isaac-sim/IsaacLab.git>
cd IsaacLab
git describe --tags --exact-match
python --version
which python
=> v3.0.0-beta2.patch1
=> Python 3.12.13
=> /home/ryung/miniconda3/envs/Isaaclab_env/bin/python
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. pyqtgraph 0.14.0 requires colorama, which is not installed.
python -m pip install \
torch==2.10.0 \
torchvision==0.25.0 \
--index-url <https://download.pytorch.org/whl/cu128>
python - <<'PY'
import torch
print("torch:", torch.__version__)
print("PyTorch CUDA:", torch.version.cuda)
print("CUDA 사용 가능:", torch.cuda.is_available())
print("GPU:", torch.cuda.get_device_name(0))
a = torch.randn(2048, 2048, device="cuda")
b = a @ a
print("행렬 연산:", b.shape)
print("연산 장치:", b.device)
PY
nvidia-smi
=>
(이하부터는 ERROR: pip's dependency re~ 에 대한 에러를 해결하여 ROS2와 연동이 잘된상태로 가정)