core.utils.backward_compatibility.torch_load_with_fallback

core.utils.backward_compatibility.torch_load_with_fallback(
    filename,
    preferred_map_location='cuda',
)

Loads a PyTorch file with fallback behavior: 1. Tries preferred_map_location (default: cuda) 2. Falls back to CUDA/MPS/HIP if available 3. Finally falls back to CPU

Returns

Name	Type	Description
	Loaded model and torch device on which it has been loaded