-
|
hey team, first of all, thank you so much for providing this product! I ran the code successfully but the contribution returned all nan. I checked all the input data and there wasn't any null value. How can we understand the potential issues here? I ran mmm.trace["coef_media"] and got valid numbers. Thanks! |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments
-
|
Hi @miaox018 ! I'm not sure offhand, but my first guess would be that you might have some negative numbers in your trace; if so, can you try dropping those channels and see if the plot works? |
Beta Was this translation helpful? Give feedback.
-
|
Thanks! I actually found it later that one media variable has 0 spend during the modeling time period. So it's already resolved now. |
Beta Was this translation helpful? Give feedback.
Thanks! I actually found it later that one media variable has 0 spend during the modeling time period. So it's already resolved now.