best performing pretrained weight

[KyonP]

To author,

First of all, on behalf of our lab member who published the original Pororo dataset (K.M. Kim), we very much appreciated that this dataset is still alive and being researched.

We are trying to reproduce your achievement. However, struggling to find mimicking your best performing setting.

Is there any way to receive your best-performing checkpoint file or detailed settings?

Of course, If you don't mind. I hope it is possible.

Your work inspired us to a generative model for story visualization on a realistic drama dataset.

Best wishes,

[adymaharana]

Thank you for your interest in the repository! Your lab's work with the Pororo dataset has been invaluable for story visualization.
I uploaded the checkpoints and inference scripts to the GitHub repository recently.
Could you please try those checkpoints with your setup and see if that works? Let me know.

[KyonP]

I am trying to conduct benchmark experiments comparing them with your reported evaluation results.

For correct comparison, I think it is required that pretrained weight of your evaluation models for Proro-SV.

is it available?

BTW, congratulation on your EMNLP paper.

[adymaharana]

Hi,

Thank you for bringing this to my attention. I just added the pre-trained model for character classification and modified the eval script for more usability. I am adding the other models asap as well. Thanks for noticing the EMNLP paper! And I wish you luck on your paper :) I am excited to read it!