app.py

import streamlit as st
from PIL import Image
import easyocr
from transformers import AutoTokenizer, AutoModelForCausalLM

# Define a function to extract text from an image using easyocr
def extract_text(image):
    reader = easyocr.Reader(['en'])
    extracted_text = reader.readtext(image)
    return ' '.join([result[1] for result in extracted_text])

# Define a function to generate an explanation using the "HuggingFaceH4/zephyr-7b-alpha" model
def explain_text(text, tokenizer, model):
    input_ids = tokenizer.encode(text, return_tensors="pt", max_length=256, truncation=True)
    explanation_ids = model.generate(input_ids, max_length=100, num_return_sequences=1)
    explanation = tokenizer.decode(explanation_ids[0], skip_special_tokens=True)
    return explanation

st.title("Text Extraction and Explanation")

# Allow users to upload an image
uploaded_file = st.file_uploader("Upload an image:")

# Load the text generation model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("HuggingFaceH4/zephyr-7b-alpha")
model = AutoModelForCausalLM.from_pretrained("HuggingFaceH4/zephyr-7b-alpha")

if uploaded_file is not None:
    # Read the uploaded image
    image = Image.open(uploaded_file)

    # Extract text from the image using easyocr
    extracted_text = extract_text(image)

    # Explain the extracted text using the "HuggingFaceH4/zephyr-7b-alpha" model
    explanation = explain_text(extracted_text, tokenizer, model)

    # Display the extracted text and explanation
    st.markdown("**Extracted text:**")
    st.markdown(extracted_text)

    st.markdown("**Explanation:**")
    st.markdown(explanation)

else:
    st.markdown("Please upload an image to extract text and get an explanation.")