list of models to evaluate

[jankounchained]

• [x] stanza ancient greek models (Jan)
• [x] udpipe
• [x] cltk
• [ ] spark-nlp ancient greek models (tokenizer, sentencer, POS + lemmas)
• [x] homercy (Marton)
• [x] GreCy (Jan)
• [ ] TranKit (Jan)

Not all components included

• [ ] #8 (Marton)
• [ ] AGILe lemmatizer (Marton)
• [ ] Handling stree in grc morpohologizer
• [ ] xlm roberta pos tagger (Jan)

[jankounchained]

todo: only speed is saved using spacy benchmark accuracy the external models were without a scorer. fixed.