Recalculate item and user vectors on-the-fly

recalculate_user.py

def recalculate_user(user_ratings):
    '''adds new user and its liked items to sparse matrix and returns recalculated recommendations'''

    alpha = 40
    m = load_npz('sparse_user_item.npz')
    n_users, n_movies = m.shape

    ratings = [alpha for i in range(len(user_ratings))]

    m.data = np.hstack((m.data, ratings))
    m.indices = np.hstack((m.indices, user_ratings))
    m.indptr = np.hstack((m.indptr, len(m.data)))
    m._shape = (n_users+1, n_movies)

    # recommend N items to new user
    with open('model.sav', 'rb') as pickle_in:
        model = pickle.load(pickle_in)
    recommended, _ =  zip(*model.recommend(n_users, m, recalculate_user=True))
    
    return recommended, map_movies(recommended)

This code, somehow, seems to have made it onto the screen of a Barbie toy:

https://www.reddit.com/r/ProgrammerHumor/comments/1j2kv2y/

amazing

Nice

Obligatory "I was here". Beautifully obscure and random.