Bo Sun

Feature request

1

Hi Author, I really appreciate this Hugo theme, I wonder would it be possible for you to add a map features showing the "the site visits and click demographics to...

what has been changed in your commit fix isort?

2

Dear Author, What change has been made in the commit fix isort to prepare voc few shot data? Thanks

**Is your feature request related to a problem? Please describe.** The implementation of Relative Positional Encoding are all included 0 paddings, which turns out to be not necessary. **Describe the...

Hi Author, You mentioned in pytorch zero-padding is used to replace symmetric padding. I think [replication padding](https://pytorch.org/docs/stable/nn.html#torch.nn.ReplicationPad2d) in pytorch is the same as symmetric padding in tf. Would you take...

Models seem not converging.

1

Hi, I tried to train CoAtNet_0 with tiny image net from cs231n (200 classes). Seems the model does not converge. Could it be that the implementation is not 100% correct?...

Question about mssim

2

李博你好, I'm reading your source code and I have two questions: 1. I think there is a little bug in `net.py`, correct me if I am wrong. In `net.py`, I...

Quick Question: what is the actual training dataset volume?

1

Hi Author, Thanks for the great work. I've been curious in the "ID-Oriented Human Data Construction" described in paper, after ID verification and other filtering, what is the final training...

pip install decord fails.

7

❯ pip install decord ERROR: Could not find a version that satisfies the requirement decord (from versions: none) ERROR: No matching distribution found for decord  Hi author, can you...

Question regarding color constancy loss

1

Hi Author, In the `MyLoss.py` you uploaded, ```python class L_color(nn.Module): def __init__(self): super(L_color, self).__init__() def forward(self, x ): b,c,h,w = x.shape mean_rgb = torch.mean(x,[2,3],keepdim=True) mr,mg, mb = torch.split(mean_rgb, 1, dim=1)...

{placeholder_token} vs *s in inference

1

In the fine tune code, there is an assert hard code the special token as {placeholder_tokne}. ```assert ( "{placeholder_token}" in args.prompt_template ), "You must specify the location of placeholder token...