**I am trying to build a recommendation engine using the following three columns - customer_id, Content_title, and Clicks. I am able to train for initial few subscribers (say 1000) and after that, it throws an error **
Code executed
if __name__ == '__main__':
import time
start = time.time()
content_1 = []
scores_1 = []
subs_1 = []
rec1 = []
for i in range(0, 10000): #len(subids)-1):
recommended = model.recommend(subids[i], sparse_user_item[i])
rec1.append(recommended)
for j in range (0,10):
subs_1.append(subids[i])
#scores_1 = zip(scores_1,recommended[1])
scores_1.append(list(recommended[1]))
content_1.append(list(recommended[0]))
Error Message
---------------------------------------------------------------------------
IndexError Traceback (most recent call last)
Input In [256], in <cell line: 2>()
9 rec1 = []
11 for i in range(0, 10000): #len(subids)-1):
---> 12 recommended = model.recommend(subids[i], sparse_user_item[i])
13 rec1.append(recommended)
15 for j in range (0,10):
File ~/.local/lib/python3.8/site-packages/implicit/cpu/matrix_factorization_base.py:51, in MatrixFactorizationBase.recommend(self, userid, user_items, N, filter_already_liked_items, filter_items, recalculate_user, items)
48 if user_items.shape[0] != user_count:
49 raise ValueError("user_items must contain 1 row for every user in userids")
---> 51 user = self._user_factor(userid, user_items, recalculate_user)
53 item_factors = self.item_factors
55 # if we have an item list to restrict down to, we need to filter the item_factors
56 # and filter_query_items
File ~/.local/lib/python3.8/site-packages/implicit/cpu/matrix_factorization_base.py:135, in MatrixFactorizationBase._user_factor(self, userid, user_items, recalculate_user)
133 if recalculate_user:
134 return self.recalculate_user(userid, user_items)
--> 135 return self.user_factors[userid]
IndexError: index 86110 is out of bounds for axis 0 with size 40916