Shoufa Chen comments

Results 121 comments of


                                            Shoufa Chen

nan occurs when training ImageNet 128x128

I am now at 430180 steps and don't meet `NaN`.

nan occurs when training ImageNet 128x128

I am training a 128*128 ImageNet model.

code?

Hi, Thanks for your interest. Would you please which part you need most? I will push this part of the code with high priority since my limited bandwidth. Thanks for...

We are glad to provide core code here: https://gist.github.com/ShoufaChen/263eaf55599c6e884584d7fce445af45, but unfortunately don't have the bandwidth to support full code at this time. We hope you understand. You are welcome to...

Low Linear-Prob accuracy

Hi, @launchauto, @michuanhaohao , @mts42000 Thanks for your efforts in reproducing the linear probe results. I noticed that the official MAE repo has released the linear probe code. Thus, it...

Code

Hi, thanks for your interest. Please see https://github.com/ShoufaChen/WOO/issues/5#issuecomment-1296731745.

Some questions.

We're closing this issue since no further questions. Please feel free to reopen it if.

When I run train.py for about 40 iters, it got an error and the program break

Because when then **iter** run over one whole epoch, it can't go on iterating from the start point automatically. [here](https://github.com/amdegroot/ssd.pytorch/issues) solved this problem

Overlapping Segments

Hi, @MoritzNekolla May I ask how do you get the number 70543? Thanks in advance.

Overlapping Segments

Thanks for your reply @MoritzNekolla .