You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
I tried to train the model with Flickr30K dataset, but when I tested, I got Bleu-4 score of 0.16137 which is away lower than the results that reported in the paper. Based in your comments in other post, you got the same Bleu-4 score as the original paper, so it seems like I am doing something wrong. I would like to know if you are changing any parameters in the code when working with Flickr30K or you keep the same parameters of COCO dateset.
The text was updated successfully, but these errors were encountered:
Hi. The previous comment was based on the older version of the code, which was trained on Flickr30k. Let me check the older version and get back to you.
Hi,
I tried to train the model with Flickr30K dataset, but when I tested, I got Bleu-4 score of 0.16137 which is away lower than the results that reported in the paper. Based in your comments in other post, you got the same Bleu-4 score as the original paper, so it seems like I am doing something wrong. I would like to know if you are changing any parameters in the code when working with Flickr30K or you keep the same parameters of COCO dateset.
The text was updated successfully, but these errors were encountered: