Skip to content

Releases: nikitakit/self-attentive-parser

Parser models

27 Apr 07:57
@nikitakit nikitakit
models
37c8e66
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.
Compare
Choose a tag to compare
Parser models Latest
Latest

Attached is a selection of the pre-trained models:

  • en_charlstm_dev.93.61.pt: The best English single-system parser from our ACL 2018 paper that does not rely on external word representations
  • en_elmo_dev.95.21.pt: The best English single-system parser from our ACL 2018 paper. Using this parser requires ELMo weights, which must be downloaded separately.
  • benepar_*.zip: These files are downloaded using benepar.download, so you should not need to download them manually.
  • en_elmo_ensemble_tfgraph.pb.gz: This file is downloaded using benepar.download('benepar_en_ensemble'), so you should not need to download it manually.
  • en_elmo_small_tfgraph.pb.gz: This file is downloaded using benepar.download('benepar_en_small'), so you should not need to download it manually.
  • en_elmo_tfgraph.pb.gz: This file is downloaded using benepar.download('benepar_en'), so you should not need to download it manually.

Don't use any of the "Source code" links here; refer to the README instead for instructions on how to run the code.

Assets 32