What's the difference between train.txt and val.txt?

[kimkyeongnam]

There are four data files: src-train.txt, target-train.txt, src-val.txt, tgt-val.txt in OpenNMT.

I know that the difference between src and target is language.
But I can’t guess what the difference is between a train and val file.

Could you explain about it? Thank you

[guillaumekln (Guillaume Klein)]

This post should help:

Validation Data

Hi, You will find some useful information about training, validation and test sets in this page: tl;dr: the validation data is a set of sentences used to evaluate the convergence of the training. The model that achieves the lowest perplexity on this dataset is considered the best. In practice, you usually start with a single corpus of parallel sentences and cut it in 3 parts: ~2000 validation sentences* ~2000 test sentences* the rest as training sentences * reasonable sizes are between 2…

[kimkyeongnam]

I didn’t know the abbreviation for ‘val’ was validation.
Thank you so much.