Skip to content
This repository has been archived by the owner on Oct 8, 2019. It is now read-only.

Refactor the internal to be more easy to implement new ML algorithms #116

Open
myui opened this issue Sep 18, 2014 · 0 comments
Open

Refactor the internal to be more easy to implement new ML algorithms #116

myui opened this issue Sep 18, 2014 · 0 comments
Assignees
Milestone

Comments

@myui
Copy link
Owner

myui commented Sep 18, 2014

  • Refactor the base class of learners to be more easy to implement new algorithms for ML developers
    ObjectInspector should be transparent to ML implementation
  • Separate convex optimization algorithms (sgd/mini-batch/cg) from classifier/regression implementation
  • Make LossFunctions selectable from each implementation
@myui myui self-assigned this Sep 18, 2014
@myui myui added this to the v0.3 milestone Sep 18, 2014
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

1 participant