Skip to content

Conversation

@SamDuffield
Copy link
Contributor

Adds a compulsory init_likelihood argument to all init_prepare functions. It has default True so doesn't change any existing tests or code, although if we agree on this API we probably want to add tests for init_likelihood=False.

As discussed in #178 this is required for factorial models.

@SamDuffield SamDuffield requested a review from Sahel13 February 7, 2026 14:11
@SamDuffield
Copy link
Contributor Author

Also I'm open to different options for the name of the arg, e.g. apply_update, apply_initial_observation ...?

@Sahel13
Copy link
Collaborator

Sahel13 commented Feb 7, 2026

I think apply_initial_observation, or a shorter apply_init_obs, is better.

@SamDuffield
Copy link
Contributor Author

As mentioned in #184 we made the soft decision not to have this arg as it's cleaner to have a unified API controlled by model_inputs rather than adding args unnecessarily.

Soft decision is to refactor to remove initial observation everywhere.

@SamDuffield SamDuffield marked this pull request as draft February 9, 2026 14:25
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants