Evaluating test data while training, not other process

[SSUHan]

Hello jingxil,

Thank you for sharing this wonderful codes :) I'm trying to use this code in "Sentence Reordering Task" now, then I got some troubles..

You use the forward_only option to separate "train" scope and "test" (or inference) scope into process units. However, I want to do test during training to see if the model overfits.. but it's not easy because of attention mechanism(-> train and test run differently..) :(

Is there any good way to solve this problem??

Even if you don't code it, I'd really appreciate if you could share a reference sites or hints I could refer to

Thank you

[jingxil]

Hi, SSUHan. Currently, I am using the PyTorch framework which is easy to switch between 'test' and 'train' mode. Since I am quite rusty at the Tensorflow, I do not think I can help here. Sorry for that.