The documentation only mentions using the command line to run the translate.py script already given in the OpenNMT-py repo, i.e. use
python translate.py -model model_name.pt -src data/src-test.txt -output pred.txt
and that's fine, but how do I use the onmt library to load the trained model in a jupyter notebook and make translations on the go? I want it to be able to do something like this in jupyter:
model = onmt.module.load(/path/to/model.pt)
text = "hola como estas"
model.translate(text)
Output:
>>>Hello how are you
I can't find any documentation on how to do this. Please help!