Training dataset using SVM

7 views (last 30 days)
Aya Fathi
Aya Fathi on 30 Nov 2020
Answered: Raunak Gupta on 5 Dec 2020
Hi everyone,
i have a training set that i want to train it using SVM, i have already extracted the features using HOG, but i don't know how to proceed with the training, what should i do next and what function to use (i found out that there's more than 1 function :fitcecoc / fitcsvm / ... )
Also, i would like to know , if i changed HOG to LBP, can i use the same steps and same function ?
Thank you in advance.

Answers (1)

Raunak Gupta
Raunak Gupta on 5 Dec 2020
Hi,
From the dataset, while extracting the features you must be knowing that if this is a binary class problem or a multi-class problem. If it’s a binary class problem, you can use fitcsvm otherwise, you can use fitecoc which fits a multiclass model for SVM classifiers. You can use any features to fit a SVM model whether HOG or LBP but the performance for both the feature sets will be different. You may find this example useful to get started.

Community Treasure Hunt

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

Start Hunting!

Translated by