Replies: 1 comment
-
|
Hi @rohanvarm, This is a very good question. We trained and tested NodeCoder on a group of human proteins, for a limited number of tasks as mentioned in the paper. So I am afraid my answer cannot be generalized, and more comprehensive training and testing experiments for various tasks are required to find out about the behaviour of trained models for single-tasks vs multi-task learning. What I can share with you is that, for the conducted experiments, we did not observe improvements when performing multitask learning. Many thanks for your feedback! |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi,
Did you see any improvement when using Multi-task learning as opposed to single task learning? In addition, do you think protein language embeddings would help with the existing framework?
Thanks and great work!
Beta Was this translation helpful? Give feedback.
All reactions