Spaces:
Sleeping
Sleeping
| import pandas as pd | |
| from sentence_transformers import SentenceTransformer | |
| import faiss | |
| import numpy as np | |
| import streamlit as st | |
| import pickle | |
| import os | |
| # Load CSV file | |
| csv_file = "Hydra-Movie-Scrape.csv" | |
| df = pd.read_csv(csv_file) | |
| # Load the precomputed embeddings | |
| with open("embeddings.pkl", "rb") as f: | |
| doc_embeddings = pickle.load(f) | |
| # Convert the embeddings into a NumPy array (FAISS requires float32) | |
| embedding_matrix = np.array(doc_embeddings).astype("float32") | |
| # Build a FAISS index for efficient similarity search | |
| index = faiss.IndexFlatL2(embedding_matrix.shape[1]) | |
| index.add(embedding_matrix) | |
| # Load the SentenceTransformer model for encoding the query | |
| model = SentenceTransformer('all-MiniLM-L6-v2') | |
| # Function to retrieve the most relevant movies based on a query | |
| def retrieve(query, top_k=10): | |
| query_embedding = model.encode(query) | |
| query_vector = np.array(query_embedding).astype("float32") | |
| distances, indices = index.search(np.array([query_vector]), top_k) | |
| return indices[0] | |
| # Streamlit app layout | |
| st.title("Movie Dataset RAG Application") | |
| query = st.text_input("Ask a question about movies:") | |
| if st.button("Submit"): | |
| if query: | |
| indices = retrieve(query) | |
| response = "" | |
| for idx in indices: | |
| if idx != -1: | |
| movie_details = df.iloc[idx] | |
| response += f"*Title*: {movie_details['Title']}\n" | |
| response += f"*Year*: {movie_details['Year']}\n" | |
| response += f"*Director*: {movie_details['Director']}\n" | |
| response += f"*Cast*: {movie_details['Cast']}\n" | |
| response += f"*Summary*: {movie_details['Summary']}\n\n" | |
| if response: | |
| st.write("Here are some movies that match your query:") | |
| st.markdown(response) | |
| else: | |
| st.write("No relevant documents found.") | |
| else: | |
| st.write("Please enter a query.") | |