-
Notifications
You must be signed in to change notification settings - Fork 525
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
Support Vector Machine #912
Conversation
@teju85 and @cjnolet: originally I had an Sklearn like stateful cpp interface #456 in svc.h, and the python bindings are using that for convenience. I have added a stateless layer afterwards. Currently everything is linked into libcuml++.so. Do we have any update on how we should separate the stateless and stateful API? |
…inition error of non template function
Thanks @cjnolet for the review! I will address the issues with the failing test. |
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.
Left couple of comments for mostly code structure. Overall, very well written. Congrats.
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.
Thanks Onur for the review! I have addressed what I could now, and created new issues for those that would prefer to postpone after 0.10.
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.
Just a small change to use RMM
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.
Needed an additional comment, so treat both of my reviews as a single one :P
Hi @dantegd , thanks for the mini-review! I have addressed the issues, could you have a look? |
Support Vector Machine
This branch adds a Support Vector Machine Classifier (C-SVC).
We use prims from PRs #892 and #853, reviewers please focus here on the SVM files in the cpp/src/svm directory.
The Sequential Minimal Optimization method is used to fit the SVM, similarily to ThunderSVM or OHD-SVM.
Currently only binary classification is implemented.