helpers.py

def create_user_movie_df():
    import pandas as pd
    movie = pd.read_csv('datasets/movie.csv')
    rating = pd.read_csv('datasets/rating.csv')
    df = movie.merge(rating, how="left", on="movieId")
    df['title'] = df.title.str.replace('(\(\d\d\d\d\))', '')
    df['title'] = df['title'].apply(lambda x: x.strip())
    a = pd.DataFrame(df["title"].value_counts())
    rare_movies = a[a["title"] <= 1000].index
    common_movies = df[~df["title"].isin(rare_movies)]
    user_movie_df = common_movies.pivot_table(index=["userId"], columns=["title"], values="rating")
    return user_movie_df