Hi,
I'm new to AutoML, and I'm trying to train a model that translates from domain specific arabic texts to English. My issue is that the English sentences in the data I've scraped don't always align with the corresponding arabic sentences; in many cases the translator has shortened the sentences by adding a period mark, since the arabic sentences tend to be quite long. Is there any way to get around this, by for instance passing in whole paragraphs as input, instead of sentences?
Any help is very much appreciated,
Thea