mahesh27/div-class-roberta

This is the best-performing classfier in Divinity classification of Rig Vedic verses.

Sample usage:

from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained("mahesh27/div-class-roberta")
model = AutoModelForSequenceClassification.from_pretrained("mahesh27/div-class-roberta", device_map="auto")

text = "ahim ahann sindhum pariśiśriyānam valam abhinad yataḥ sindhavo akṣaran prati vatsam na dhenavaḥ"
### "Killed the snake lying in the river Indus, broke the cave entrance whence flowed the rivers like cows towards a calf."
inputs = tokenizer(text, return_tensors='pt')["input_ids"].to(model.device)

model.eval()
pred = model(input_ids= inputs)
prediction = int(pred['logits'].argmax(dim=-1)[0])
print(model.config.id2label[prediction])

> 'indraḥ'

Citation:

@inproceedings{mahesh-bhattacharya-2023-creation,
    title = "Creation of a Digital Rig {V}edic Index (Anukramani) for Computational Linguistic Tasks",
    author = "Akavarapu, V.S.D.S.Mahesh  and
      Bhattacharya, Arnab",
    editor = "Kulkarni, Amba  and
      Hellwig, Oliver",
    booktitle = "Proceedings of the Computational {S}anskrit {\&} Digital Humanities: Selected papers presented at the 18th World {S}anskrit Conference",
    month = jan,
    year = "2023",
    address = "Canberra, Australia (Online mode)",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.wsc-csdh.6/",
    pages = "89--96"
}