app.py

import torch
import gradio as gr
import pandas as pd
from sentence_transformers import SentenceTransformer, util

df = pd.read_csv("movies_dataset.csv")
df['full_text'] = (
    df['title'].fillna('') + " | " +
    df['genres'].fillna('') + " | " +
    df['overview'].fillna('') + " | " +
    df['tagline'].fillna('')
)
df = df.sort_values(by="popularity", ascending=False).head(5000).reset_index(drop=True)
model = SentenceTransformer('all-MiniLM-L6-v2')
df['full_text'] = df['full_text'].fillna('').astype(str)
embeddings = model.encode(df['full_text'].tolist(), show_progress_bar=False)
df['embeddings'] = embeddings.tolist()
embedding_tensor = torch.tensor(embeddings)

def recommend_movies(user_input, top_k=5):
    if not user_input.strip():
        return "❗Please enter a movie description or genre."
    user_embedding = model.encode(user_input, convert_to_tensor=True)
    similarities = util.cos_sim(user_embedding, embedding_tensor)[0]
    top_indices = similarities.argsort(descending=True)[:top_k]
    results = []
    for idx in top_indices:
        row = df.iloc[idx.item()]
        results.append(f"🎬 **{row['title']}**\n{row['overview'][:300]}...")
    return "\n\n".join(results)

demo = gr.Interface(
    fn=recommend_movies,
    inputs=gr.Textbox(lines=2, placeholder="Describe a movie, mood, or genre..."),
    outputs=gr.Markdown(),
    title="🎬 Movie Recommender",
    description="Type in a description, vibe, or genre and get movie suggestions!"
)

demo.launch()