SoTA Abstract Meaning Representation (AMR) parsing with word-node alignments in Pytorch. Includes checkpoints and other tools such as statistical significance Smatch.
Thank you for releasing your code and checkpoints! It's very helpful. However, there may be a config name typo in the Trained checkpoints section in README.md
Thank you for releasing your code and checkpoints! It's very helpful. However, there may be a config name typo in the
Trained checkpointssection inREADME.mdfix typo
existing problems
The config is the same for those two lines:
possible fix
Change the word in the first line from
amr3jointtoamr2joint, according to your aws bucket and paper.Best wishes!