How to recall Classification Learner models in the command window
4 views (last 30 days)
Hi there, I'm trying to estimate mean and standard deviation of a model on different training and testing combination. Manually it is so bad and I wonder if it is possible to call models from CL like "weighted KNN" "Ensemble KNN" etc in the command window in order to implement them in an iterative algorithm to store results and compute statistics. I saw it's only possible to call simple classifier not from CL. Any idea?
Asvin Kumar on 17 Mar 2021
All the options that you see in the Classification Learner App have functional equivalents as well. All those functions are tucked under this page: https://www.mathworks.com/help/stats/classification.html You will see hyperlinks to different approaches: Nearest Neighbour, Naive Bayes, Trees, SVMs, etc. Each page will have a list of all the functions available for that method.
A simple approach to quickly figure out the functional equivalent of an option in the app is to use the "Generate Function" option from the tool strip on the Classification Learner App. This generates a MATLAB function file with a call to the function for the classification method of your choice. Hope this helps you get started.