Skip to content

sithu31296/coco-caption

 
 

Repository files navigation

Microsoft COCO Caption Evaluation

Evaluation codes for MS COCO caption generation.

This is a new fork from ruotianluo/coco-caption which fixes the following:

  • METEOR evaluation error
  • SPICE evaluation error (Caching error in Windows)
  • Windows Installation setup

Other details can be seen in the forked repo.

Requirements

  • java 1.8.0
  • python 3
  • gensim
  • pyemd

For Windows, download java from here.

Installation

For Linux:

$ ./get_stanford_models.sh
$ ./get_google_word2vec_model.sh

For Windows:

  • Download Stanford CoreNLP model from here.
  • Extract the zip file.
  • Copy stanford-corenlp-3.6.0.jar and stanford-corenlp-3.6.0-models.jar to pycocoevalcap/spice/lib.
  • Download Google News negatie 300 word2vec model from here.
  • Extract the zip file.
  • Copy GoogleNews-vectors-negative300.bin to pycocoevalcap/wmd/data.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 91.5%
  • Python 8.3%
  • Shell 0.2%