how to get the alignments between tokens and node? - Githubissues

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

245 stars 48 forks source link

how to get the alignments between tokens and node? #20

Closed LucyFann closed 3 years ago

LucyFann commented 3 years ago

how to get the alignments between tokens and node?

ramon-astudillo commented 3 years ago

can you be more concrete? alignments for training are obtained by installing and running the aligner tools, as explained in the README. Once a model is trained, it produces AMR files with alignments in JAMR notation (at right-most side of the nodes)

LucyFann commented 3 years ago

sorry for not clear my problem...I want to get the alignments as you explain in README by JAMR tools. The bash preprocess/install_alignment_tools.sh for JAMR's default setup script in JAMR repos is set to a version no longer available. Now I have succeed get the alignments after use new versions of scala and sbt. Refer to this issue for more details.