frajem

Hi @limingqishi Can you replicate the results in the paper?

I'm interested to know the answer to this question, too. Thank you!

Hi @intuinno Would you share the model file trained on Coco? Also, what are your best validation/test cost for Flickr8k and Coco? Thanks.

@kyunghyuncho , @kelvinxu Hello, By using default parameters for Coco, soft attention, I get much lower results on the test set than what was published: BLEU-1=0.545, METEOR=0.164, CIDer=0.274. The only...

Yes, the dictionary has IDs in descending frequency order. Any idea about why I'm getting so much lower metrics on Coco (see my comment above)? Thanks.

Hello and thanks for this repository! Why do you need to specify whether it's the "first_train_for_this_task"? Cannot you define the faster_params before the loop, and then just use the following...