app.py

import streamlit as st
import pickle
from sentence_transformers import CrossEncoder
with open('eiga_docs.pkl', 'rb') as file:
    documents=pickle.load( file)
#with open('index.pkl', 'rb') as file:
#    index=pickle.load( file)
#documents =[ i["title"]+" "+i["text"] for i in documentt]
#bi_encoder = SentenceTransformer('all-mpnet-base-v2',device="cpu",)

# Load a cross-encoder model for re-ranking
cross_encoder = CrossEncoder('cross-encoder/ms-marco-MiniLM-L-6-v2')

# Streamlit app
#st.title("Information Retrieval System")

#query = st.text_input("Enter your query:")


#if query:
    # Encode the query using the bi-encoder
    #query_embedding = bi_encoder.encode([query])

    # Retrieve top-k documents using FAISS
    #k = 30  # Number of documents to retrieve
    #D, I = index.search(query_embedding, k)
    
    #Collect the retrieved documents
    #retrieved_docs = [documents[i] for i in I[0]]
    
    #Re-rank the retrieved documents using the cross-encoder
    #cross_inp = [[query, doc["title"]+" "+doc["text"]] for doc in documents]
    #scores = cross_encoder.predict(cross_inp)
    
    # Sort the documents by score
    #sorted_docs = [documents[doc] for _, doc in sorted(zip(scores, range(len(documents))), reverse=True)][:10]

    # Display the results
    #st.write("Top documents:")
    #for i, doc in enumerate(sorted_docs):
    #    st.write(f"Result {i+1}. Title: {doc['title']}")
    #    st.write(f"Text : {doc['text']}")
    #    st.write(f"Lien de telechargment : {doc['doc_url']}")
    #    st.markdown("""---""")



# Function to display results for a given page
def display_results(sorted_docs, page_number, results_per_page):
    start_index = (page_number - 1) * results_per_page
    end_index = min(page_number * results_per_page, len(sorted_docs))
    for i, doc in enumerate(sorted_docs[start_index:end_index], start=start_index):
        st.write(f"**Result {i+1}.** Title: {doc['title']}")
        st.write(f"Text : {doc['text']}")
        st.write(f"Link : {doc['doc_url']}")
        st.markdown("""---""")



# Function to perform search and return sorted documents
def perform_search(query):
    if query:
        cross_inp = [[query, doc["title"] + " " + doc["text"]] for doc in documents]
        scores = cross_encoder.predict(cross_inp)

        # Sort the documents by score
        sorted_docs = [documents[doc] for _, doc in sorted(zip(scores, range(len(documents))), reverse=True)]
        return sorted_docs
    else:
        return []

# Main Streamlit app
def main():
    st.sidebar.title("Pagination")
    results_per_page = st.sidebar.number_input("Results per page", min_value=1, max_value=10, value=10)
    total_pages = (len(documents) + results_per_page - 1) // results_per_page
    page_number = st.sidebar.number_input("Page Number", min_value=1, max_value=total_pages, value=1)
    
    st.sidebar.write( f"Totale Pages :{total_pages}")
    #print(results_per_page,total_pages,page_number)
   
    st.title("Information Retrieval System")
    query = st.text_input("Enter your question:")

   

    if st.button("Search") or query:
        sorted_docs = perform_search(query)
        st.session_state.sorted_docs = sorted_docs
        
    else:
        sorted_docs = st.session_state.get("sorted_docs", [])



    if sorted_docs:
        display_results(st.session_state.sorted_docs, page_number, results_per_page)
        st.write(f"Current Page Number: {page_number}")
        



# Run the app
if __name__ == "__main__":
    main()