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.

Sign up for GitHub

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

Jump to bottom

Multiclass Support #2

Open

khdlr opened this issue Sep 13, 2021 · 1 comment · May be fixed by #7

Labels

enhancement high priority

Collaborator

khdlr commented Sep 13, 2021

The introduction of multiclass support would be great

potential overlap with [TODO]

requirements:

Training data/vectors attribute label has numerical values, which can be > 1

preprocessing:

mask as uint8 with label ids as values

data output:

class labels as uint8 as uint8 with label ids as values
class probabilities
- option 1: n bands with p values, n=number of classes (my preferred option)
- option 2: 1 band, p value of majority class

numbers output:

class metrics need to be adapted to multiclass (precision, recall, f-score)
addition of new output: e.g. error matrix

khdlr added enhancement high priority labels

Collaborator Author

khdlr commented Sep 13, 2021 •

edited

Loading

This is essentially ready to merge at #7

khdlr mentioned this issue

Water Masking #4

Open

khdlr linked a pull request

that will close this issue

[WIP] Multiclass Support #7

Open

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