search

mlpc-ucsd / CoaT

(ICCV 2021 Oral) CoaT: Co-Scale Conv-Attentional Image Transformers
Apache License 2.0
227 stars 30 forks source link

When will you release the code? #1

Closed morizin closed 3 years ago

morizin commented 3 years ago

I am very interested in your project. i want to know when will your team releases the code with pre-trained model please

xwjabc commented 3 years ago

I am very interested in your project. i want to know when will your team releases the code with pre-trained model please

Thank you for your interest in our project! We are preparing the code repo now and plan to release a version of the code and pretrained models by the end of this week. Stay tuned!

Update: I found a bug in my code just now (an incorrect order in concatenation of cls_token and img_tokens). The release date would be delayed for several days.

lartpang commented 3 years ago

@morizin Based on my own understanding, I implemented the ConvAttention module. I hope it works for you. And if you find a mistake, welcome to point it out!

morizin commented 3 years ago

Sure. I will start to preparing my network with this

liqiokkk commented 3 years ago

I am looking forward to the author's code.

xwjabc commented 3 years ago

Hi all, there is a version of code released today. Feel free to give it a try!