Splitting data to training and testing without validation

2 views (last 30 days)
Nagwa megahed
Nagwa megahed on 21 Apr 2022
Dear all,
my dataset has only 20 sample per class and i don't apply any augmentation in order to build a few shot learning model,
i ask is the results wil be correct ? if i test the data to train and test without validation in orde to i can calculate the generalization error
as i know the generalization error is the disability of the model to recognize new unseen data in training

Answers (0)




Community Treasure Hunt

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

Start Hunting!

Translated by