nizhenliang

What is the relationship between GMACs and GFLOPs?

19

Why use the GMACs？GMACs is different with GFLOPs.

ImageNet Pretraining config

3

您在文中并没有提到在imagenet上训练跑了多少epoch以及用了几块GPU，batch size是多大，还有优化器和学习率等参数。是否方便公布一下这些参数或者将预训练的log放出来？这会极大的帮助我们复现您的工作，非常感谢！

An error is reported during the fpn_efficientformerv2_s0 test.

2

q = self.q(x).flatten(2).reshape(B, self.num_heads, -1, H * W).permute(0, 1, 3, 2) RuntimeError: shape '[1, 8, -1, 336]' is invalid for input of size 90112

Hello, I found that you used 32 A100s. Would you like to ask how much memory is the A100? 80 G? If we use 8 cards 32G V100, how long...

Dear author: Hello. There are some questions to consult. 1. What is the batch size of the segmentation model training? 2. In addition, can you release the results of b1,...