--preprocess full looking for mappingnet_checkpoint file?

[blid11]

I was able to run the following command and get an output:
python inference.py --driven_audio Recording.M4A --source_image face2.png --enhancer gfpgan

However, trying to use the --preprocess full and --still flags gave me the following error using the example image from the instructions:
PS C:\Users\brett\Documents\SadTalker> python inference.py --driven_audio Recording.M4A --source_image full_body_2.png --result_dir results/test_output2 --still --preprocess full WARNING: The new version of the model will be updated by safetensor, you may need to download it mannully. We run the old version of the checkpoint this time! Traceback (most recent call last): File "C:\Users\brett\Documents\SadTalker\inference.py", line 144, in <module> main(args) File "C:\Users\brett\Documents\SadTalker\inference.py", line 40, in main animate_from_coeff = AnimateFromCoeff(sadtalker_paths, device) File "C:\Users\brett\Documents\SadTalker\src\facerender\animate.py", line 70, in __init__ self.load_cpk_mapping(sadtalker_path['mappingnet_checkpoint'], mapping=mapping) File "C:\Users\brett\Documents\SadTalker\src\facerender\animate.py", line 145, in load_cpk_mapping checkpoint = torch.load(checkpoint_path, map_location=torch.device(device)) File "C:\Users\brett\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\serialization.py", line 713, in load return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args) File "C:\Users\brett\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\serialization.py", line 920, in _legacy_load magic_number = pickle_module.load(f, **pickle_load_args) EOFError: Ran out of input

Did anyone else using this option have to do anything extra (e.g. download another checkpoint file)?

[blid11]

somehow erased one of the mapping net files, just redownloaded them!