jeffheaton / encog-dotnet-core

http://www.heatonresearch.com/encog
Other
430 stars 150 forks source link

why using IMLRegression in classification? #119

Open jack1232 opened 6 years ago

jack1232 commented 6 years ago

Thank you, Jeff, for your great work.

I have a question regarding to the "Encog 3.3: Quick Start Guide". In the "Using Encog for Classification" example of this Guide, you define the best method using the statement: var bestMethod = (IMLRegression) model.Crossvalidate(5, true);

why IMLRegression is used here? should we need to use IMLClassification instead (because this is a classification problem)?

Another question: when performing cross-validation, is it possible to extract the epoch vs error (or loss) result -- i.e., retrieve the intermediate results within each iteration?

Thanks. Jack