JanineCHEN

Hi, I'm wondering if ClusterBeliefPropagation is supported by now or not??

> What was the solution? Hi, it was the residual batch that caused the problem, you can either drop_last when constructing the dataloader or increase the number of epochs to...

Having the same doubles here

> Hello guys, I trained the code with two classes and each of them consists of ~20000 images. I got an fid of 45 with a batch_size of 26, dimensions...

> Hello, > I'm trying to finetune BigGAN together with @VincieD. We use the weights at 100k iterations provided by the author and try to finetune it on the [INRIA...

> @JanineCHEN Neither fixing the number of classes, training from scratch or fine tuning BigGAN worked in my case. At some point it is collapsing. Also training it for a...

I am encountering the same issue. It would be more than helpful if you could release some trained checkpoint model, and provide some instructions on running the trained model on...

> You can refer to lines 80 and 81 in `interpolate.py`. Thank you! I will take a look. :)

> You can refer to lines 80 and 81 in `interpolate.py`. Hi @zhujiapeng , it turns out that getting the intermediate tensor output is not that straightforward (I am not...

> We use the proposed re-scoring technique to conduct a layer-wise analysis. In this way, the layers with the highest response are treated as the most relevant layers. Thank you...