| import streamlit as st | |
| import requests | |
| import pickle | |
| import pandas as pd | |
| movies_dict = pickle.load(open('pickle_files\movies_dict.pkl', 'rb')) | |
| movies = pd.DataFrame(movies_dict) | |
| movies_list = movies['title'].tolist() | |
| movies_cos_sim = pickle.load(open('pickle_files\movies_cos_sim.pkl', 'rb')) | |
| print(movies) | |
| def get_recommendations(movie): | |
| if movie in movies['title'].tolist(): | |
| index = movies[movies['title']==movie].index[0] | |
| ascending_indices = movies_cos_sim[index].argsort() | |
| descending_indices = ascending_indices[::-1] | |
| return movies.iloc[descending_indices[1:21]]['title'].tolist() | |
| else: | |
| return 0 | |
| TMDB_API_KEY = 'd86b9111d56c0b1da25f8f37a94c9a2b' | |
| def fetch_movie_poster(movie_name): | |
| search_url = f"https://api.themoviedb.org/3/search/movie?api_key={TMDB_API_KEY}&query={movie_name}" | |
| response = requests.get(search_url).json() | |
| if response['results']: | |
| poster_path = response['results'][0]['poster_path'] | |
| full_poster_url = f"https://image.tmdb.org/t/p/w500{poster_path}" | |
| return full_poster_url | |
| else: | |
| return None | |
| st.title("Movie Recommendation System") | |
| movie_name = st.selectbox("Enter a movie name:", movies_list) | |
| if st.button("Get Recommendations"): | |
| if movie_name: | |
| recommendations = get_recommendations(movie_name) | |
| if recommendations: | |
| st.write("Here are some movies you might like:") | |
| cols = st.columns(5) | |
| for idx, movie in enumerate(recommendations): | |
| col = cols[idx % 5] | |
| with col: | |
| st.write(movie) | |
| poster_url = fetch_movie_poster(movie) | |
| if poster_url: | |
| st.image(poster_url, width=250) | |
| else: | |
| st.write("Poster not found.") | |
| else: | |
| st.write("No recommendations found. Please check the movie name and try again.") | |
| else: | |
| st.write("Please enter a movie name.") |