compellit/mt5-scansion-gl-cx

Model Card for mt5-scan-gl-cx

Metrical scansion in Galician (lexical to metrical syllabification). Fine-tuned mT5.

Uses the previous and following input line as context, as in this example from "Á moda" by Filomena Dato.

Input format: PREV: sin / *fe / nin / cre- / *en- / zas | CUR: *ten / *cen- / tos / de / al- / *ta- / res | NEXT: *che- / os / de / ri- / *que- / zas | OUTPUT:

Output for the above: *ten / *cen- / tos / de al- / *ta- / res

Use the code below to get started with the model.

import torch
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

model_name = "compellit/mt5-scan-gl-cx"

device = "cuda" if torch.cuda.is_available() else "cpu"

tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

text = "PREV: sin / *fe / nin / cre- / *en- / zas | CUR: *ten / *cen- / tos / de / al- / *ta- / res | NEXT: *che- / os / de / ri- / *que- / zas | OUTPUT: "

inputs = tokenizer(text, return_tensors="pt")

with torch.no_grad():
    outputs = model.generate(
        **inputs,
        max_length=256,
        num_beams=1,
        do_sample=False
    )

print(tokenizer.decode(outputs[0], skip_special_tokens=True))