how to fine correct classification for each class in NN

2 次查看(过去 30 天)
Dear All
I have 2 questions:
1- What does false/ negative, positive rate and correct neagtive, positive rate means in NN.
2- How to find total correct classification for each class.

采纳的回答

Greg Heath
Greg Heath 2014-10-18
positive/negative == class 1/class 0
target output label
0 0 true negative
0 1 false positive
1 0 false negative
1 1 true positive
The target matrix contains 0-1 column unit vectors
trueclassindices = vec2ind(target)
assignedclassindices = vec2ind(output)
err = assignedclassindices~=trueclassindices
Nerr = sum(err)
individual class errors are ontained via
ind1 = trueclassindices==1'
etc
Thank you for formally accepting my answer
Greg

更多回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Sequence and Numeric Feature Data Workflows 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by