File size: 1,032 Bytes
49c09fe eb0342c 49c09fe eb0342c 49c09fe |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 |
import gradio as gr
import subprocess
subprocess.check_call(["pip", "install", "--upgrade", "huggingface-hub"])
subprocess.check_call(["pip", "install", "transformers"])
subprocess.check_call(["pip", "install", "torch"])
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("balaramas/mbart-sahitrans_new_data")
model = AutoModelForSeq2SeqLM.from_pretrained("balaramas/mbart-sahitrans_new_data")
def sanmt(txt):
tokenizer.src_lang = "hi_IN"
encoded_ar = tokenizer(txt, return_tensors="pt")
generated_tokens = model.generate(
**encoded_ar,
forced_bos_token_id=tokenizer.lang_code_to_id["hi_IN"]
)
output = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)[0]
return output
iface = gr.Interface(
fn=sanmt,
inputs=gr.Textbox(label="Enter text in Sanskrit", placeholder="Type here..."),
outputs=gr.Textbox(label="Translated Hindi Text"),
title="Sanskrit to Hindi Translator"
)
iface.launch() |