"The paging file is too small for the operation to complete"

[NathanDreamweaver]

OSError: [WinError 1455] The paging file is too small for this operation to complete. Error loading "C:\Users\Nathan\Desktop\enunu_training_kit-1.0.0\python-3.8.10-embed-amd64\lib\site-packages\torch\lib\cudnn_adv_infer64_8.dll" or one of its dependencies. from nnsvs.gen import get_windows File "C:\Users\Nathan\Desktop\enunu_training_kit-1.0.0\python-3.8.10-embed-amd64\lib\site-packages\nnsvs\gen.py", line 4, in <module> import torch File "C:\Users\Nathan\Desktop\enunu_training_kit-1.0.0\python-3.8.10-embed-amd64\lib\site-packages\torch\__init__.py", line 124, in <module> raise err

I have 16 GB of RAM, my paging file is 32 GBs, could it be that my GPU only has 4 GB of VRAM? I have AMD Ryzen 3800X NVidia GeoForce GTX 1650 SUPER 16 GBs of DDR4 RAM 1 TB SSD Windows 11 Pro

[NathanDreamweaver]

I have fixed that through https://stackoverflow.com/a/69489193 , but now getting RuntimeError: CUDA out of memory. Tried to allocate 36.00 MiB (GPU 0; 4.00 GiB total capacity; 1.90 GiB already allocated; 0 bytes free; 2.35 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

[oatsu-gh]

Try decrease batch_size in conf/train/*/data/*.yaml and middle_frequency in config.yaml (or config_rmdn.yaml) to supress required VRAM size.