bingrao / ctin Goto Github PK

hello,may I ask you to share the division of the test, training, and validation sets in the RONIN dataset, RIDI dataset, and OXIOD dataset, so that I can compare my network with yours more fairly,think you.

Hello, CTIN has an amazing performance. When will you release your code?
It is very useful for me.

Hi bingrao, thank you for the amazing work on the CTIN model. I have a question about model implementation. When I build spatial encode blocks, I don't know exactly what number of filters is used in each block so the number of parameters of my model differs from yours. Could you give me that information in order to implement your model correctly?

Can you ask the company if it agrees to open source code? Would love to learn your work.

How is the final MLP layer designed? The decoding generates a tensor of [batch,200,dim], do you use the view function to change it linearly after it becomes [batch,200×dim]? Or do you only make a linear change for the vector of that 200th dim dimension? Because it is the speed of generating the ith moment.