Why not divide the data by training set, validation set, test set, but only the training and validation set

CUHK-AIM-Group / U-KAN

[ArXiv' 24] U-KAN Makes Strong Backbone for Medical Image Segmentation and Generation

https://yes-u-kan.github.io/

283 stars 39 forks source link

Open HUCONGG opened 4 months ago

HUCONGG commented 4 months ago

Why not divide the data by training set, validation set, test set, but only the training and validation set

Docnoah commented 2 months ago

good question! I have the same confusion