Hydraplus_Net issues

2

Thanks for your excellent work to reproduce the original paper.Could you share your tricks for the improvements? Such as: 1.How many epoch do you use in the train code,is it...

xjhjinhui

version

1

First of all, thank you for providing the code, please tell me the pytorch version and your enviroment

568770592

Have you got the same effect as the original paper?

I changed my batch_size to 8 because of the restriction of conditions, and I couldn't run to 1000 epochs. The result of my run was 7 percentage points worse than...

wzzwh110

checkpoint!

1

Hello, can you tell me how to get your checkpoint, or can you provide me with a checkpoint.Thank you very much

he527465038

Details about training init weights~

As you write in README.MD,the first step is to train init weights,but what's the steps to train those Nets(MNet、AF1、AF2、AF3、HP)？ DO I have to train them all for 1000 epochs?That would...

JimKimHome

In the paper, the author said "... , which the **cross-entropy** loss L^attr is applied for pedestrain attribute recognition." However, in your code, the loss function is **BCEWithLogitsLoss**. Could you...

HarminDo