search

google-research / big_transfer

Official repository for the "Big Transfer (BiT): General Visual Representation Learning" paper.
https://arxiv.org/abs/1912.11370
Apache License 2.0
1.5k stars 175 forks source link

Mobilenetv2 & MobilenetV3 pretrained models ! #3

Closed dragen1860 closed 4 years ago

dragen1860 commented 4 years ago

Dear authors: Thanks for the beneficial research work! everyone doing cv will gain a lot from your pretrained models. AFAK, lots of friends from industry are inteseted in these mobile-friendly models. It would be great if you could publish some edge-device pretrained models, such as mobilenetv2 and mobilenetv3, which is the son of Google! Thank you!

lucasb-eyer commented 4 years ago

Hi, thanks for the interest.

In the paper, we show that unfortunately small models benefit less from big pre-training. We are actively working in the direction of more edge-friendly models, but as this is research, I cannot promise anything :)

dragen1860 commented 4 years ago

Yes, it's helpful! THank you.