Closed khawar-islam closed 3 years ago
Hello have you checked the authors' repo https://github.com/whai362/PVT ?
Is the model you reference (ResNet34 with Pyramid Vision Transformer Version 2 ) described or implemented in the paper?
I am also looking on the pyramid vision transformer concept. @khawar512
No, they did not use the ResNet18 model with PVT but I want to use it. do you have any idea? We can work together and discuss PVT concepts.
do you have any intuition why this would help you solve your task? i would like to add some version of pvt but i am currently short on time
Thank you for your work with a clear explanation. As you know, ViT doesn't work on small datasets and I am implementing ResNet34 with Pyramid Vision Transformer Version 2 to make it better. The architecture of ViT and PVT V2 is completely different. Could you provide me some help to implement it? please