You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A modelbase has the ability to watch for changes in its model directory. Currently it only reacts on new models in the directoy, but does not reload existing models if they are changed. @jong42 requested this feature. Please discuss if this is a desired feature or not, and - if desired - implement it.
The text was updated successfully, but these errors were encountered:
Con:
I am unsure this is a good way to support the incremental workflow of bayesian modelling with ppls. The reason is that overwriting a model makes it impossible to compare the new (incrementally changed) model with the old one.
Pro:
It seems the natural thing to do, and the above bayseian modelling is a very specific use case.
Conclusion:
I believe the feature should be implemented, but maybe we need some way of 'auotmatically' name incrementally changed models for the ppls use case?
A modelbase has the ability to watch for changes in its model directory. Currently it only reacts on new models in the directoy, but does not reload existing models if they are changed.
@jong42 requested this feature. Please discuss if this is a desired feature or not, and - if desired - implement it.
The text was updated successfully, but these errors were encountered: