zlyin

decode.acc_seg drops to 0.0 quickly and leads to 0s evaluation scores on custom dataset

2

Hi there, I'm pretty new to this framework and came across a wired situation that my `decode.acc_seg` drops to 0 quickly and stays there. As a result, my evaluation results...

'EfficientNetEncoder' object has no attribute 'act1' when loading back timm-efficient-b5 weights

2

I trained a model of `timm-efficientnet-b5` with `imagenet` pretrained weights on my custom dataset. The train & val process works good. Then when I loaded it back in another script...

Hello Javier, I'm trying to understand your WHD forward function, but fail to understand the dimension of `normalized_y`. ```python # One by one prob_map_b = prob_map[b, :, :] gt_b =...