-
Notifications
You must be signed in to change notification settings - Fork 8
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
Implement CRPS for normal distribution. #60
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think the native code is a bit too complicated for this project. What about one of these in pure cunumeric: https://en.wikipedia.org/wiki/Error_function#Approximation_with_elementary_functions?
I think you could probably do something in 5 lines or so. It may not be that fast, but it might be better to implement it as simply as possible, then adopt a possible legate scipy implementation later.
All comments are addressed. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. Trying to get CI back up and running today, then we can merge.
PROPERTIES | ||
CXX_STANDARD 17 | ||
CXX_STANDARD_REQUIRED ON | ||
POSITION_INDEPENDENT_CODE ON) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What is position independant code for?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nothing, it's the default, just a habit of being explicit. Can revert if needed.
erf
function. Thinking about upstreaming it to somewhere like raft-legate. Not sure if legateboost wants to depend on raft-legate.The metric is important for probabilistic forecasting.