alasdairtran / transform-and-tell

[CVPR 2020] Transform and Tell: Entity-Aware News Image Captioning
https://transform-and-tell.ml/
89 stars 14 forks source link

Can you provide a trained model checkpoint? #14

Closed zmykevin closed 2 years ago

zmykevin commented 3 years ago

Hi Alasdair, Thank you for sharing the code about your great work! I am interested to play with your a trained news caption model for my own project and I wonder if you can share a trained checkpoint. Thanks in advance and I am looking forward to hearing your reply. Best, Mingyang

alasdairtran commented 3 years ago

If you just need the checkpoints, here are the all the files inside the expt directory:

https://cloudstor.aarnet.edu.au/plus/s/vsN51Y9Dh4eww39

See the README file for more info on the different model variants.