I/O format of the predict function. #16

Open

Labels

opened

[Need discussion]
What should be the format for the I/O of the predict() function in the Python interface? It seems that the current I/O format is not easy enough to use especially for 1 vs k evaluation.

to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Assignees

No one assigned

Labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests