Pretrained-Pix2Seq issues

Will you try to align the accuracy to the paper based on the official tensorflow code ?

The offical code: https://github.com/google-research/pix2seq

Could you update the pytorch code based on the official tf implementation?

Thanks for your work. The official code is in https://github.com/google-research/pix2seq.

Why much slower than Stable-Pix2Seq

I found that it takes about 75 minutes per epoch during I was training 'Pre-trained Pix2Seq', while only takes 50 minutes per epoch in 'Stable Pix2Seq'. Why? Where's the differences...

yshMars

About the gap between Training loss and Inference loss

I found that the loss gap between training and inference stage is very large. The loss in inference stage is 10 times that in training. Even making inference on the...

T-FOUNTAIN

About LargeScaleJitter

3

hi, great work! We also try to reimplement the Pix2Seq, we find the absolute coordinate is useful, which is similar to your LargeScaleJitter (pad or crop the image to the...

ShuaiBai623

NUCLEUS SAMPLING

2

Great work! Have you implemented NUCLEUS SAMPLING mentioned in paper? Thank you.

hust-nj

Some problems about your great work

Hi! Thank you for your great work. And I have some problems about your work: in the paper,there are some dropouts for class when building input seq,but in your work...

hughwcq