Multi-Dataset Training

[alanakbik]

Some tasks have multiple datasets available over which we might want to train at the same time.

For instance, there are currently 6 English treebanks listed on the UD web page over which we might want to train a PoS tagger for English. Or we might want to train a multilingual PoS tagger for several languages, in which case we would want to simultaneously train over UD corpora for English, German etc.

Task: Add data and training methods to allow for training of models over multiple datasets.