app.py

import numpy as np
import pandas as pd
from dotenv import load_dotenv

from langchain_community.document_loaders import TextLoader
from langchain_huggingface import HuggingFaceEmbeddings
# from langchain_text_splitters import CharacterTextSplitter
from langchain_chroma import Chroma
from langchain.schema import Document
import gradio as gr

import os
import requests

load_dotenv()

movies = pd.read_csv("movies_with_emotions.csv")

loader = TextLoader("../tagged_plot.txt", encoding="utf-8")
raw_documents = loader.load()
with open("../tagged_plot.txt", encoding="utf-8") as f:
    lines = f.read().splitlines()

documents = [Document(page_content=line) for line in lines if line.strip()]

embeddings = HuggingFaceEmbeddings(
    model_name="sentence-transformers/all-mpnet-base-v2"
)

db_movies = Chroma(
    persist_directory="chroma_db",
    embedding_function=embeddings
)

def retrieve_semantic_recommendations(
        query: str,
        genre: None,
        tone: None,
        initial_top_k: int=50,
        final_top_k: int=16,
) -> pd.DataFrame:
    recs = db_movies.similarity_search(query, k= initial_top_k)
    movie_list = [(rec.page_content.split()[0].lstrip('"')) for rec in recs]
    movie_recs = movies[movies["Wiki Page"].isin(movie_list)].head(final_top_k)

    if genre != "All":
        movie_recs = movie_recs[movie_recs["simple_genre"] == genre][:final_top_k]
    else:
        movie_recs = movie_recs.head(final_top_k)


    if tone == "Happy":
        movie_recs.sort_values(by="joy", ascending=False, inplace=True)
    elif tone == "Surprising":
        movie_recs.sort_values(by="surprise", ascending=False, inplace=True)
    elif tone == "Angry":
        movie_recs.sort_values(by="anger", ascending=False, inplace=True)
    elif tone == "Suspenseful":
        movie_recs.sort_values(by="fear", ascending=False, inplace=True)
    elif tone == "Sad":
        movie_recs.sort_values(by="sadness", ascending=False, inplace=True)

    return movie_recs


def recommend_movies(
        query: str,
        genre: None,
        tone: None,
):
    recommendations = retrieve_semantic_recommendations(query,genre,tone)
    results = []
    for _,row in recommendations.iterrows():
        plot = row["Plot"]
        truncated_desc_split = plot.split()
        truncated_plot = " ".join(truncated_desc_split[:20]) + "..."

        authors_split = row["Director"].split(",")
        if len(authors_split) == 2:
            authors_str = f"{authors_split[0]} and {authors_split[1]}"
        elif len(authors_split) > 2:
            authors_str = f"{', '.join(authors_split[:-1])}, and {authors_split[-1]}"
        else:
            authors_str = row["Director"]

        cast_str = "N/A"
        if isinstance(row["Cast"],str):
            cast_split = row["Cast"].split(",")
            if len(cast_split) == 2:
                cast_str = f"{cast_split[0]} and {cast_split[1]}"
            elif len(authors_split) > 2:
                cast_str = f"{', '.join(cast_split[:-1])}, and {cast_split[-1]}"
            else:
                cast_str = row["Cast"]

        caption = (f"{row['Title']} (Directed by {authors_str})\n"
                   f"Cast: {cast_str}\nPlot: {truncated_plot}")

        title = row["Title"]
        api_key = os.getenv("OMDB_API_KEY")
        year = row["Release Year"]

        url = f"http://www.omdbapi.com/?apikey={api_key}&t={title}&y={year}"
        data = requests.get(url).json()

        if data["Response"] == "True":
            poster = data["Poster"] + "&fife=w800"
            if data["Poster"] == "N/A":
                poster = "cover-not-found.jpg"
        else:
            poster = "cover-not-found.jpg"

        results.append((poster,caption))
    return results



categories = ["All"] + sorted(movies["simple_genre"].unique())
tones = ["All"] + ["Happy", "Surprising", "Angry", "Suspenseful", "Sad"]



with gr.Blocks(theme = gr.themes.Glass()) as dashboard:
    gr.Markdown("# Movie Recommendation System")

    with gr.Row():
        user_query = gr.Textbox(label = "Please enter description of a movie:",
                                placeholder = "e.g. A movie about animals")
        category_dropdown = gr.Dropdown(choices = categories, label = "Select a genre: ", value = "All")
        tone_dropdown = gr.Dropdown(choices = tones, label = "Select an emotional tone: ", value="All")
        submit_button = gr.Button("Find Recommendations")

    gr.Markdown("## Recommendations")
    output =  gr.Gallery(label = "Recommended movies", columns=8, rows=2)

    submit_button.click(fn = recommend_movies,
                        inputs=[user_query, category_dropdown,tone_dropdown],
                        outputs=output)

    if __name__ == "__main__":
        dashboard.launch()