OpenNMT

1 BLEU Score result of quick start documentation

I am following the same exact steps as in the quick start documentation of Opennmt-py Quickstart — OpenNMT-py documentation, I got a result of 1.0674705347651567 BLEU point as a result of the translation using Sacre Bleu. Is that normal?

There is the following note at the end of the quickstart:

The predictions are going to be quite terrible, as the demo dataset is small. Try running on some larger datasets!

It’s normal that the results are bad.