New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

example MPT quantization doesn't work #360

Open

havietisov opened this issue Jul 9, 2023 · 1 comment

havietisov commented Jul 9, 2023

Followed this guide : https://github.com/ggerganov/ggml/tree/master/examples/mpt
When tried to cast model to fp-16, got following error :

Loading part: pytorch_model-00001-of-00007.bin
Traceback (most recent call last):
File "/mnt/j/ggml/examples/mpt/convert-h5-to-ggml.py", line 133, in
model_part = torch.load(f"{dir_model}/{part_name}", map_location="cpu")
File "/home/_username/.local/lib/python3.10/site-packages/torch/serialization.py", line 815, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/home/_username/.local/lib/python3.10/site-packages/torch/serialization.py", line 1033, in _legacy_load
magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: invalid load key, 'v'.

Maybe guide/code should be updated

Contributor

klosax commented Jul 20, 2023 •

edited

Loading

_pickle.UnpicklingError: invalid load key, 'v'.

It looks like your model files are corrupted. Try download them again.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment