Similar users and movies

similar.py

def most_similar_items(item_id, n_similar=10):
    '''computes the most similar items'''
    with open('model.sav', 'rb') as pickle_in:
        model = pickle.load(pickle_in)

    similar, _ = zip(*model.similar_items(item_id, n_similar)[1:])

    return map_movies(similar)


def most_similar_users(user_id, n_similar=10):
    '''computes the most similar users'''
    sparse_user_item = load_npz("sparse_user_item.npz")
    
    with open('model.sav', 'rb') as pickle_in:
        model = pickle.load(pickle_in)

    # similar users gives back [(users, scores)]
    # we want just the users and not the first one, because that is the same as the original user
    similar, _ = zip(*model.similar_users(user_id, n_similar)[1:])

    # orginal users items
    original_user_items = list(sparse_user_item[user_id].indices)

    # # this maps back user_ids to their information, which is useful for visualisation
    similar_users_info = map_users(similar)
    # # now we want to add the items that a similar used has rated
    for user_info in mapped:
        # we create a list of items that correspond to the simillar user ids
        # then compare that in a set operation to the original user items
        # as a last step we add it as a key to the user information dictionary
        user_info['items'] = set(list(sparse_user_item[user_info['user_id']].indices)) & set(original_user_items)

    return similar_users_info