Hello! I encountered an issue while reproducing the training part code of your paper: I trained on multiple tasks, but it seems that I could only achieve the segmentation of task A in the end. The segmentation results on other tasks are also the segmentation of A. This may be because the model can only learn A, But the difference in data volume is not significant, and data augmentation has also been performed. My concern is - how do you implement the learning of multiple tasks in the code? Because the structure of universeg is mainly aimed at single tasks and single tags, I would like to know how you handle it
Hello! I encountered an issue while reproducing the training part code of your paper: I trained on multiple tasks, but it seems that I could only achieve the segmentation of task A in the end. The segmentation results on other tasks are also the segmentation of A. This may be because the model can only learn A, But the difference in data volume is not significant, and data augmentation has also been performed. My concern is - how do you implement the learning of multiple tasks in the code? Because the structure of universeg is mainly aimed at single tasks and single tags, I would like to know how you handle it