Revision history [back]

oh, another problem. the HOGDescriptor applies a 2 class regression"" , not a **multi-class classification,

So I train a svm with 15 classes. everything is fine.

no, nothing's fine. you have to train it on positive / negative samples using some variant of SVM::C_SVR.

please, again take a look at train_HOG.cpp !

oh, another problem. the HOGDescriptor applies a 2 class regression"" , not a **multi-class classification,

So I train a svm with 15 classes. everything is fine.

no, nothing's fine. fine, that's plain wrong.

you have to train it on positive / negative samples using some variant of SVM::C_SVR.

please, again take a look at train_HOG.cpp !

So I train a svm with 15 classes. everything is fine.

no, nothing's fine, that's plain wrong.not fine.

the HOGDescriptor needs a 2 class regression , not a multi-class classification,

you have to train it on positive / negative samples using some variant of SVM::C_SVR.

please, again take a look at train_HOG.cpp !

So I train a svm with 15 classes. everything is fine.

no, not fine.

the HOGDescriptor needs a 2 class regression , not a multi-class classification,

you have to train it on positive / negative samples of a single class using some variant of SVM::C_SVR.

please, again take a look at train_HOG.cpp !