discrim

This is version 0.3 of the Discriminant Analysis Toolbox with major bug fixes.

Description

This is version 0.3 of the Discriminant Analysis Toolbox with major bug fixes and additional functionality. A new class hierarchy is introduced with a single parent class for all classification objects. In addition to linear, quadratic and logistic discriminant analysis, limited feed-forward neural network classification is also made possible using the SOFTMAX multinomial deviance critereon for error minimisation. Additional functions include k-fold cross-validation, confustion matrices and several plotting functions. Many of these algorithms are described in B. D. Ripley (1996) Pattern Recognition and Neural Networks.

• MATLAB release: MATLAB 5.2 (R10)

[Wilmer Henao]

The graphics are great, it´s quite the best you can get free on the net. There is plenty of tests and the files are easy to understand and change.

[Carl Nelson]

This is a thorough and well-designed implementation of the procedures in Ripley's "Pattern Recognition and Neural Networks."

[ganael bascoul]

 

[fatemeh emdad]

very good

[Antonio Ferreira]

 

[asahraf fad]

 

[Kris De Gussem]

Good built quality. The toolbox does exactly what you can expect.

[Matthieu Bosiger]

really helpfull!

[Neos Neo]

normal

[sherly novita]

good

[Yaju Fan]

 

[Tim Yee]

needs more documentation

[sophie sue higgs]

i need to know more !

[Bhushan Thekedar]

 

[dafna hirschfeld]

In the file @lda/classify.m in row 49 there is the command:
classify(f.classifier, X);

I don't see what this line is for, it seems that it doesn't do anything .

[Felix Franke]

>In the file @lda/classify.m in row 49 there is the command:
>classify(f.classifier, X);
>I don't see what this line is for, it seems that it doesn't do anything .

This is just for error checking using the underlying classifier.classify() function.

Nice toolbox by the way! Thanks

[Xiaoying Tang]

Why I can't download these files?