IBM / transition-amr-parser

SoTA Abstract Meaning Representation (AMR) parsing with word-node alignments in Pytorch. Includes checkpoints and other tools such as statistical significance Smatch.
Apache License 2.0
246 stars 48 forks source link

APT Checkpoints? #53

Open Zoher15 opened 1 year ago

Zoher15 commented 1 year ago

Hi @jzhou316!

The instructions say:

from transition_amr_parser.parse import AMRParser
parser = AMRParser.from_checkpoint(in_checkpoint)
annotations = parser.parse_sentences([['The', 'boy', 'travels'], ['He', 'visits', 'places']])
# Penman notation
print(''.join(annotations[0][0]))

Which 'checkpoint' am I supposed to here?

Thanks!

Zoher

jzhou316 commented 1 year ago

Hi Zoher, you can download our trained checkpoints from online. I believe @ramon-astudillo has better instructions about it. Can you shoot an email requesting the information?

ramon-astudillo commented 1 year ago

I think he is looking for APT ones, which we do not offer

jzhou316 commented 1 year ago

I see. I have some of the original APT checkpoints. Maybe I can share if he asks?

Jiawei (Joe) Zhou PhD candidate, SEAS, Harvard Email: @.***

On Mon, Jun 26, 2023 at 5:03 PM Ramón Fernandez Astudillo < @.***> wrote:

I think he is looking for APT ones, which we do not offer

— Reply to this email directly, view it on GitHub https://github.com/IBM/transition-amr-parser/issues/53#issuecomment-1608257237, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHSGBYSGDNIQYSM3NJYL7ELXNH2J7ANCNFSM6AAAAAAZMLRQHQ . You are receiving this because you were mentioned.Message ID: @.***>

ramon-astudillo commented 1 year ago

Perfect, just remember to use the corresponding branch