-
I am trying to work out if you can supply custom channel beta priors with different distributions using the MMM class.... For example, if I had 3 channels, and on 2 of them I wanted to supply a HalfNorm prior and on 1 of them I wanted to supply a Normal prior - Can I do this by changing the default_model_config? I have gone through the code but couldn't work it out. |
Beta Was this translation helpful? Give feedback.
Answered by
wd60622
Sep 17, 2024
Replies: 1 comment
-
Hi, You can use a truncated normal distribution and have the lower bound as -np.inf for "Normal" and lower of 0 for "HalfNormal" For additional support, that currently isnt built out |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
raz1470
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi,
You can use a truncated normal distribution and have the lower bound as -np.inf for "Normal" and lower of 0 for "HalfNormal"
For additional support, that currently isnt built out