Skip to content

A tool to plot ROC curve in multi-class classification models.

Notifications You must be signed in to change notification settings

ttbond/mulRocPlayer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

BEFORE:
make sure you have the following softwares
python3 numpy sklearn
R ggplot2 viridis cowplot


Steps of using mulROCplayer
1 prepare source file.(The format and details of the file are showed in the following)
2 run mulRocPlayer.py


(The only file you have to prepare, making clear of the format is essential to use this tool)
Source file format:
file name should be 'source.dat'
line 1: sample number(represented by n)
line 2: name of the columns of probability matrix(no blank in in the name and each of them is seperated by one blank)
line 3->n+2:probability matrix(seperated by blank)
line n+2->2*n+1:the real classification in probability matrix

(If only images are requested, it is unnecessary for you to read the format of Rel file.)
Rel file(the python code give out) format:
file name rel.dat
line 1: AUC of each ROC curve
line 2: name of each class (the order is same as the following lines)
line 3->2*m+2: FPR and TPR for each ROC in turn(that means one line is FPR and the next line is TPR.)

Details:
The id of class should start from 0.
m represents the number of class.

About

A tool to plot ROC curve in multi-class classification models.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published