Closed Doraemonzm closed 6 months ago
Hi @Doraemonzm, thank you for your attention to our work.
As outlined in sec. 5.5 of our paper, the linear complexity of our module ensures that the Agent-Swin model maintains consistent computation complexity with an increased window size. Therefore, we use a window size of 56 to benefit from a global receptive field. In contrast, for Swin-T, enlarging the window size from 7 to 56 would elevate FLOPs from 4.5G to 8.8G.
Our design principle for stage attention types is using our agent attention in earlier stages, where feature map resolutions are higher, to fully leverage the benefits of enlarged receptive field.
Thanks for your quick reply. Could you please share the code for calculating FLOPs?
We use fvcore to calculate FLOPs.
Thank you for your excellent work! I find that the original window size in Swin-T is 7, whereas in agent_swin, it is 56. I am curious about your design choices regarding the window size and stage attention types in agent-swin-T/S/B. Are there any guiding principles behind these decisions?