No CUDA GPUs are available - environment.yaml

[rashmi-ss]

Hello,
I am trying to set up my anaconda environment to run the SLA code but I am facing some challenges-

1. I have tried with requirements.txt earlier but there was error - RuntimeError: No CUDA GPUs are available
2. I have tried with environment.yaml but still there is error - RuntimeError: No CUDA GPUs are available
3. I do have Nvidia GPU available in my computer and it works fine for other anaconda projects where GPU is needed. When I run these commands in the newly created environment, I get these results-

torch.version
'2.0.0'
print(torch.cuda.is_available())
True
torch.cuda.get_device_properties(0)
_CudaDeviceProperties(name='NVIDIA GeForce RTX 2060', major=7, minor=5, total_memory=6143MB, multi_processor_count=30)

4. Operating System is Windows 11
5. Since, some of the dependencies were not resolved from environment.yaml, I had to comment those before creating environment. Please refer the attached file 'environment_yaml.txt' (changed the name and type as .yaml was not getting attached here).
   environment_yaml.txt

Could you please help to resolve this issue?

[rashmi-ss]

Thanks for the awesome work. The code provides good result on OfficeHome dataset.

The environment issue is finally resolved after following these-

1. conda create -n sla_env python=3.10.10
   conda activate sla_env
2. pip install -r requirements.txt
   requirements.txt
3. Installed torch and cuda
   pip install torch==1.12.0+cu116 torchvision==0.13.0+cu116 torchaudio==0.12.0 --extra-index-url https://download.pytorch.org/whl/cu116

[chu0802]

Thanks for your help to figure out the best way to build the environment! I will update the instruction for installation accordingly.