XiaLiPKU / EMANet

The code for Expectation-Maximization Attention Networks for Semantic Segmentation (ICCV'2019 Oral)

https://xialipku.github.io/publication/expectation-maximization-attention-networks-for-semantic-segmentation/

FLOPs

songqi-github opened this issue 4 years ago · comments

Qi Song commented 4 years ago

How to compute the FLOPs and parameters of your EMA module?
Could you please share the computing details? Thanks!

Xia Li 李夏 commented 4 years ago

We adopt the same rule in torchstat.
That is, for matrix computation mn x np, the FLOPS here is mnp