Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

MultiGP #274

Merged
merged 3 commits into from
Jun 25, 2018
Merged

MultiGP #274

merged 3 commits into from
Jun 25, 2018

Conversation

costashatz
Copy link
Member

This PR mainly addresses #266 ... It also removes from default optimizer the parallel_repeater, because in most cases it doesn't improve performance and makes things slower.. Of course, one can use it if he wants..

One thing that I thought while coding it is the following. I created the MultiGP class as a template so that one can train the N-GPs as he wishes. On the contrary, I am constraining the querying to the cases where the GPs are independent. Should I provide some functor/functionality to allow for full multi-output GPs? Of course the default case would be the independent version. From an implementation point of you, it is very easy to add it; I am just asking if it makes sense/is worth it? What do you think?

@jbmouret
Copy link
Member

jbmouret commented Jun 25, 2018

I like it! And I think we can focus on independent GPs.

@costashatz
Copy link
Member Author

I like it! And I think we can focus on independent GPs.

OK! I will leave it as is then.. I will merge once the tests pass..

@costashatz
Copy link
Member Author

Merging..

@costashatz costashatz merged commit 6241f7f into master Jun 25, 2018
@costashatz costashatz deleted the multi_gp branch June 25, 2018 17:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants