about code

[will876]

Hello! Thanks for your amazing work.
In test_org.txt, I have the following:
AG8A5398
AG8A5400
AG8A5402,
In test_selected.txt, I have the following:
AG8A5398-2f9cfdf6d4
AG8A5400-d1783082d2
AG8A5402-c04fda61ec
In train_org.txt, I have the following:
AG8A5146
AG8A5150
AG8A5177
Is this the correct way to edit the txt file?
Then when I run the data_prepare_laval.py file on the Linux server, the .png image of test will be generated in the defined from_folder path, and the .exr file of train will be generated in the to_folder path. Is this normal in the preprocessing of the dataset?
Then after running the code, I kept getting stuck at this place: min and max values ​​of dynamic ranges: -12.387324 13.446923
min and max values ​​of dynamic ranges: -12.387324 13.446923
Evaluating metrics...
cache_file: /root/.cache/dnnlib/gan-metrics/train-inception-2015-12-05-727a22bb5ae280d417e8b5ea7683aa63.pkl
The GPU usage is always 0, but the video memory usage size is 5600MB. This phenomenon occurs because I did not change the paths in the two py files: 4. Test Lighting estimation and editing Set paths (stylegan2_ada_ffhq) in PTI_utils/paths_config.py Set options (lighting estimation or lighting editing) in PTI_utils/hyperparameters.py. Does it have anything to do with this? (In addition, I would like to ask why the same command is in the windows system. When I run python data_prepare_laval.py, there are only three empty folders in the generated *****128x256-data-splits2 folder, and there is no generated data. And the .png image of test is generated in my from_folder path. Is it recommended to apply this code in Linux and macOS as you said?) Thank you very much!