doubts about BG/NBD data shape, and parameters from conditional_expected_number function

First of all, I dont have words to compliment and congratulate for this amazing work.

I have a couple doubts about the BG/NBD model

1. Reading the docs, customers with one purchase should be given a recency of 0 which I did but when I tried to train my model I faced a lot of issues as:

   RuntimeWarning: invalid value encountered in multiply
     defvjp(anp.exp,    lambda ans, x : lambda g: ans * g)
   
   RuntimeWarning: divide by zero encountered in log
   

   The model did not converge. Try adding a larger penalizer to see if that helps convergence.
   

   In the end, I could only train my model after exclude customers with one purchase , why that so?

2. In the conditional_expected_number_of_purchases_up_to_time function the t parameter refers to which unit of time?

Again, thank you so much, sorry for my English, congratulations to your work from Brazil.

[wallacez1]

“frequency represents the number of repeat purchases the customer has made. This means that it’s one less than the total number of purchases. This is actually slightly wrong. It’s the count of time periods the customer had a purchase in:” i got a little confuse with that definition

this is sample of my data, i took frequency as the number of repeat purchases that a customer has made and manually assign value 0 to recency.

thank you in advance.

[wallacez1]

customers with one purchase should be given a recency of 0

I think you meant frequency of 0?

Hey @clausherther, took it from the docs:

recency represents the age of the customer when they made their most recent purchases. This is equal to the duration between a customer’s first purchase and their latest purchase. (Thus if they have made only 1 purchase, the recency is 0.)