8000 Unpickling error when running generation script with pretrained weights · Issue #11 · allenai/scitldr · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Unpickling error when running generation script with pretrained weights #11

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

Jump to bottom
Open
malits opened this issue Jan 21, 2021 · 0 comments
Open

Unpickling error when running generation script with pretrained weights #11

malits opened this issue Jan 21, 2021 · 0 comments

Comments

@malits
Copy link
malits commented Jan 21, 2021

I renamed the bart.tldr-ao.pt model to checkpoint_best.pt and tried running the generation script as python scripts/generate.py model/ SciTLDR-Data/SciTLDR-A/ctrl ./ --beam 2 --lenpen 0.4 --test_fname test.hypo as shown in the github instructors, but meet the following error:

Traceback (most recent call last):
  File "scripts/generate.py", line 99, in <module>
    generate_TLDRs(**vars(args))
  File "scripts/generate.py", line 17, in generate_TLDRs
    task='translation'
  File "/research/home/maxlitster/scitldr/repo/fairseq/models/bart/model.py", line 136, in from_pretrained
    **kwargs,
  File "/research/home/maxlitster/scitldr/repo/fairseq/hub_utils.py", line 75, in from_pretrained
    arg_overrides=kwargs,
  File "/research/home/maxlitster/scitldr/repo/fairseq/checkpoint_utils.py", line 339, in load_model_ensemble_and_task
    state = load_checkpoint_to_cpu(filename, arg_overrides)
  File "/research/home/maxlitster/scitldr/repo/fairseq/checkpoint_utils.py", line 263, in load_checkpoint_to_cpu
    state = torch.load(f, map_location=torch.device("cpu"))
  File "/home/maxlitster/miniconda3/envs/tldr/lib/python3.7/site-packages/torch/serialization.py", line 529, in load
    return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
  File "/home/maxlitster/miniconda3/envs/tldr/lib/python3.7/site-packages/torch/serialization.py", line 692, in _legacy_load
    magic_number = pickle_module.load(f, **pickle_load_args)
_pickle.UnpicklingError: invalid load key, '\x0a'.

I'm curious if anyone else has any advice as to how to generate text with the pretrained weights. Thanks!
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@malits
0