Rename latent_individual and direct_model for consistency

[athowes]

Adam to fill...

[seabbs]

At the same time as addressing this we might want to address the naming structure for custom distributions:

epidist/R/latent_lognormal.R

Line 59 in 03609f9

log_lik_latent_lognormal <- function(i, prep) {

[athowes]

At the same time as addressing this we might want to address the naming structure for custom distributions:

Yes more broadly this is the naming for custom families. Which would include the name in epidist_family and then the name in log_lik, epred and pred methods.

[athowes]

What about just:

• direct_model
• individual_model
• cohort_model

i.e. do we need the latent_ part in latent_individual?

[athowes]

• latent_model
• marginal_model

[athowes]

@kgostic notes this needs to be documented clearly.

[athowes]

• naive_model
• latent_model
• marginal_model

@athowes to implement name change.

[seabbs]

Thinking about this a bit more I think it needs to be marginalized_model to make sense? I am not sure we would talk about something being marginal?

[athowes]

https://www.pymc.io/projects/docs/en/stable/api/gp/generated/pymc.gp.Marginal.html

[seabbs]

ha nice okay good point. Also pymc is looking cool these days

[athowes]

Closed by #441.