Model Card for Deepakvictor/tan-ta

This model is Finetuned on Facebook's m2m model to convert Tanglish words to Tamil

Model Details

Model is finetuned on facebook/m2m100_418M m2m100_418m page --> https://huggingface.co/facebook/m2m100_418M

Model Description

• Developed by: Deepakvictor
• Language(s) (NLP): Tamil,Tanglish
• Finetuned from model [facebook/m2m100_418M]: [https://huggingface.co/facebook/m2m100_418M]

Model Sources

• Repository: [Need to be uploaded]
• Demo [optional]: [Need to be uploaded]

How to Get Started with the Model

Use the code below to get started with the model.

# Load model directly from transformers library
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("Deepakvictor/tan-ta")
model = AutoModelForSeq2SeqLM.from_pretrained("Deepakvictor/tan-ta")

#pass the input 
inp = tokenizer("Thalaivaru nirantharam",return_tensors="pt")
out= model.generate(**inp)
tokenizer.batch_decode(out,skip_special_tokens=True)
#['தலைவரு நிரந்தரம்']

Repo code --> github.com/devic1 🖤