Hello author, thank you for your excellent work. I'm a graduate student and want to replicate your UniFormer-L+Layer Scale to train my own data. But I didn't figure out exactly how to use this code. My task is to train a dichotomous image network that currently only has images and corresponding 0/1 labels. How do I prepare my data? Do I need to generate token label data for my training data, and if so, how? Looking forward to your answer.
Hello author, thank you for your excellent work. I'm a graduate student and want to replicate your UniFormer-L+Layer Scale to train my own data. But I didn't figure out exactly how to use this code. My task is to train a dichotomous image network that currently only has images and corresponding 0/1 labels. How do I prepare my data? Do I need to generate token label data for my training data, and if so, how? Looking forward to your answer.