search

zhmiao / OpenLongTailRecognition-OLTR

Pytorch implementation for "Large-Scale Long-Tailed Recognition in an Open World" (CVPR 2019 ORAL)
BSD 3-Clause "New" or "Revised" License
839 stars 128 forks source link

Why the input dimension of the `fc_spatial` layer in `ModulatedAttLayer` is 7*7*in_channel? #63

Closed lixucuhk closed 4 years ago

lixucuhk commented 4 years ago

Dear author, thank you for the great work, but I have one question. Why the input dimension of the fc_spatial layer in ModulatedAttLayer is 77in_channel? Why it is 7?

Personally, I think the dimension should equal the dimension of the input image, e.g. CHW in an image, then the input dimension of the fc_spatial layer should be HWC, right?

Thank you for your advice!

lixucuhk commented 4 years ago

Oh! I undertood now. Thank you!