search

XifengGuo / DEC-keras

Keras implementation for Deep Embedding Clustering (DEC)
MIT License
477 stars 162 forks source link

Why is evaluation done on the training set #31

Open fulcus opened 2 years ago

fulcus commented 2 years ago

If I understand correctly, the model is evaluated on the same data that it's trained on. Doesn't this lead to a wrong evaluation?

Load data https://github.com/XifengGuo/DEC-keras/blob/2438070110b17b4fb9bc408c11d776fc1bd1bd56/DEC.py#L290 https://github.com/XifengGuo/DEC-keras/blob/2438070110b17b4fb9bc408c11d776fc1bd1bd56/datasets.py#L94-L103

Evaluate https://github.com/XifengGuo/DEC-keras/blob/2438070110b17b4fb9bc408c11d776fc1bd1bd56/DEC.py#L333-L335

Shouldn't x_train and y_trained used to pretrain and fit, and then x_test and y_test used to evaluate the model?