Skip to content

Conversation

@ColCarroll
Copy link
Member

This is just tidying up some of the covariance adaptation code. It should not change any behavior, just makes things a little more readable.

I left a few things in so the QuadPotentialDiagAdaptGrad can be worked on in the future.

@junpenglao
Copy link
Member

I am not sure if we should make this change, IIUC the weight is for potentially adding a decaying weighting like exponential. Although right now weight=1 but I imagine removing it will make doing experiment more difficult (although nobody is actually doing these kind of experiment currently).

@ColCarroll
Copy link
Member Author

Agree -- I'll roll that back, but leave some of the other changes in. I consider this a promise that you'll continue these experiments, though!

@junpenglao
Copy link
Member

I will put it in my OKR for the next quarter 😂

@rpgoldman
Copy link
Contributor

@ColCarroll Any reason not to merge this?

@ColCarroll
Copy link
Member Author

Should be good now!

@ColCarroll ColCarroll merged commit bc027b3 into pymc-devs:master Jun 29, 2019
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.

3 participants