#Index error after loading custom pretrain on RVC Disconnected

1 messages · Page 1 of 1 (latest)

autumn leaf
#

I just downloaded the 32k sample rate Titan Medium pretrain on RVC Disconnected. I had the links for the g and d path files in the correct boxes when downloading them and I got an index error when I started training
Here's the code:

INFO:JohnWettonTitan:loaded pretrained pretrained_v2/f0G32k_custom.pth
Process Process-1:
Traceback (most recent call last):
File "/content/Mangio-RVC-Fork/train_nsf_sim_cache_sid_load_pretrain.py", line 210, in run
utils.latest_checkpoint_path(hps.model_dir, "D_*.pth"), net_d, optim_d
File "/content/Mangio-RVC-Fork/train/utils.py", line 206, in latest_checkpoint_path
x = f_list[-1]
IndexError: list index out of range

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/usr/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap
self.run()
File "/usr/lib/python3.10/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "/content/Mangio-RVC-Fork/train_nsf_sim_cache_sid_load_pretrain.py", line 230, in run
torch.load(hps.pretrainG, map_location="cpu")["model"]
File "/usr/local/lib/python3.10/dist-packages/torch/serialization.py", line 1040, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/usr/local/lib/python3.10/dist-packages/torch/serialization.py", line 1258, in _legacy_load
magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: invalid load key, '<'.

These are also the links I inserted to download:
G path: https://huggingface.co/blaise-tk/TITAN/blob/main/models/medium/32k/pretrained/G-f032k-TITAN-Medium.pth
D path: https://huggingface.co/blaise-tk/TITAN/blob/main/models/medium/32k/pretrained/D-f032k-TITAN-Medium.pth

novel junco
autumn leaf
#

Nevermind, I already managed to resolve it. I saw it on #🔥│model-maker-chat on how to load the custom pretrain and now it's working. Thanks for the help though.

abstract venture
autumn leaf
abstract venture
#

I'm trying to use the finetuned Rigel 32k model

#

I already tried that switching out "blob" with "resolve" in the links for d path and g path like cauthess did but that yielded the same result

autumn leaf
abstract venture