Last active
August 28, 2019 12:03
-
-
Save LouisdeBruijn/2a79695ca3633b80154427226c2bf9df to your computer and use it in GitHub Desktop.
Similar users and movies
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
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 | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment