XenonStack Recommends

Cross-Validation

J

K

N

O

R

X

Y

Z

Cross-Validation

What is Cross-Validation?

It is a technique which involves reserving a particular sample of a dataset which is not used to train the model. Later, the model is tested on the sample to evaluate the performance.

There are various methods:

• Leave one out cross validation (LOOCV)

• K-fold cross validation

• Stratified k-fold cross validation

• Adersarial validation