Can I use cross-validation k-fold for making FIS and optimization it with ANFIS or ANFIS itself has cross-validation?
First my data was divided into two group randomly. One for making model and other for testing best model. Then first group was used for k-fold cross-validation. FIS and ANFIS were performed and best model was selected. Then second group was used for test best model.
Is that method true?
my suggestion, it will better if you test the model these way, two tests. one is train-test, another cross-validation. then, you can compare the results of two tests, is their any significant differences? if no significant difference then we can say the model is good, working perfectly for each data.
The way you decided, one for making model and other for testing best model; it is suitable as train-test model (like as 40%train, 40% test, 20%validation). otherwise it will be complicated.
Data set also fact, is time-series data?
Thanks, no, my data is not time series data.
In my opinion if I want to measure the FIS/ANFIS performances on the test set
my method is fine. However the split should be pretty much 70% for training & cross-validation and 30% for testing. I perform it then cross-validation on 70% of the data, select the "best results", train again the model with such "best results" and then use such model on the validation set to gather the final performances.
yup... idea is great. After training the model by test set, cross-validation rather splitting data several sets is good idea. Good luck.
- Badges
- Science topic
- Similar topics
- Organizational Psychology
- Training