app.py

import streamlit as st
import pandas as pd
from transformers import MarianMTModel, MarianTokenizer

model_name = 'Helsinki-NLP/opus-mt-en-ur'

"""
MODEL_NAME = os.environ.get("model_name")

if not MODEL_NAME:
    raise ValueError("MODEL_NAME is not set. Please add it in the Hugging Face Secrets.")
"""


model = MarianMTModel.from_pretrained(model_name)
tokenizer = MarianTokenizer.from_pretrained(model_name)

# Function to translate text from English to Urdu
def translate_text(text):
    inputs = tokenizer(text, return_tensors="pt", padding=True)
    translated = model.generate(**inputs)
    return tokenizer.decode(translated[0], skip_special_tokens=True)

# Streamlit app
st.title("Dataset Translator From English to Urdu For Chatbot")
st.title("Upload Csv file for translation into Urdu. Remember Csv file must contain Only Question and Answer Column")

# Upload CSV file
uploaded_file = st.file_uploader("Upload your CSV file", type=["csv"])
if uploaded_file:
    # Read the file into a pandas DataFrame
    data = pd.read_csv(uploaded_file)

    # Translate questions and answers
    if 'Question' in data.columns and 'Answer' in data.columns:
        data['Question_Urdu'] = data['Question'].apply(translate_text)
        data['Answer_Urdu'] = data['Answer'].apply(translate_text)

        # Display the translated dataframe
        st.write(data)

        # Provide option to download the translated CSV
        translated_file = data.to_csv(index=False)
        st.download_button("Download Translated CSV", translated_file, "Diabetes_Translated_Urdu.csv")
    else:
        st.error("CSV file must contain 'Question' and 'Answer' columns")