How to decide value of 'ndim' when using ,[residual​s,reconstr​ucted] = pcares(X,ndim) , for Feature size reduction using PCA?

1 view (last 30 days)
ipwork
ipwork on 27 Jun 2015
I am working on feature classification using KNN and SVM. Data size is 2000 images and features are histograms (of various bin sizes ) of large size for each image. I read above function can be used but my query is how to decide value of ndim which will ensure that the best features are retained ?

Sign in to answer this question.

Answers (0)

Categories

Find more on Dimensionality Reduction and Feature Extraction in Help Center and File Exchange

Tags

Community Treasure Hunt

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

Start Hunting!