GaoLeiA 26ca7c2c03 fix: use torch.no_grad() in inference to prevent excessive memory usage (~30GB) with inference (#349)

2025-07-02 16:38:56 +08:00

..

__init__.py

Create __init__.py

2024-11-15 14:19:02 +08:00

inference.py

feat: windows infer & gradio (#312)

2025-04-12 23:22:22 +08:00

preprocess.py

feat: windows infer & gradio (#312)

2025-04-12 23:22:22 +08:00

realtime_inference.py

fix: use torch.no_grad() in inference to prevent excessive memory usage (~30GB) with inference (#349)

2025-07-02 16:38:56 +08:00