How to use Grid search to find the optimal hyperparameters ?

52 views (last 30 days)
% Train the classifier
knn = ClassificationKNN.fit(Xtrain,Ytrain,'Distance','seuclidean');

Answers (1)

Mohith Kulkarni
Mohith Kulkarni on 9 Dec 2020
Edited: Mohith Kulkarni on 11 Dec 2020
Use the 'OptimizeHyperparameters' Name-Value pair argument to specify the Parameters to optimize when creating the 'ClassificationKNN' model using 'fitcknn'. Use the 'HyperparameterOptimizationOptions' to specify the Optimizer for the model. Refer the code below.
Mdl = fitcknn(X,Y,'OptimizeHyperparameters','auto',...
'HyperparameterOptimizationOptions',...
struct('Optimizer','gridsearch'))
You can use this model for fitting the data with the specified optimization options.
Refer the documention of fitcknn and ClassificationKNN for more information on the Name-Value pair arguments.

Community Treasure Hunt

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

Start Hunting!