File size: 1,039 Bytes
5dc48e3 e623ffc b12d2a9 5dc48e3 3a74906 b12d2a9 5dc48e3 3a74906 5dc48e3 3a74906 5dc48e3 3a74906 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 |
import streamlit as st
from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
from transformers import AutoModelForMaskedLM, AutoTokenizer
# Load the models and tokenizers
model_translation = MBartForConditionalGeneration.from_pretrained("facebook/mbart-large-50-one-to-many-mmt")
model_masked_lm = AutoModelForMaskedLM.from_pretrained("alabnii/jmedroberta-base-sentencepiece")
model_translation.eval()
tokenizer_translation = MBart50TokenizerFast.from_pretrained("facebook/mbart-large-50-one-to-many-mmt", src_lang="en_XX")
tokenizer_masked_lm = AutoTokenizer.from_pretrained("alabnii/jmedroberta-base-sentencepiece")
text = st.text_area('Enter the text:')
if text:
model_inputs = tokenizer_translation(text, return_tensors="pt")
generated_tokens = model_translation.generate(
**model_inputs,
forced_bos_token_id=tokenizer_translation.lang_code_to_id["hi_IN"]
)
translation = tokenizer_translation.batch_decode(generated_tokens, skip_special_tokens=True)
st.json(translation)
|