search

xiaolai-sqlai / mobilenetv3

mobilenetv3 with pytorch，provide pre-train model
MIT License
1.6k stars 340 forks source link

Why you didn't use avg pool in SeModule? #3

Closed PistonY closed 5 years ago

PistonY commented 5 years ago

https://github.com/xiaolai-sqlai/mobilenetv3/blob/5ec8ae3386c3f8268e2d79a61017fc13db051885/mobilenetv3.py#L28

PistonY commented 5 years ago

You just define it but never use. It's that the reason you results higher than proposed?

BowieHsu commented 5 years ago

maybe it's a mistake

PistonY commented 5 years ago

LOL, you implement a pixel level attention by mistake.

BowieHsu commented 5 years ago

@PistonY hey, easy, man, i just mean maybe it's legacy code, after all there is no static check for pytorch code.

PistonY commented 5 years ago

HaHa