search

aimalz / scippr

Supernova Cosmology Inference with Probabilistic Photometric Redshifts
4 stars 2 forks source link

Self-consistent forward model for mock probabilities #35

Open aimalz opened 7 years ago

aimalz commented 7 years ago

The mock data right now isn't entirely self-consistent. I'm going to restructure it to make realistic probabilities the way chippr does. This issue can be closed when 1) the framework exists, 2) there is documentation of it in the simulation notebook, 3) toy versions of the mock data from it successfully run through the inference process, and 4) the paper draft includes a complete description of the new procedure. Other issues (#4, #12, #13, #30) will address the upgrade from toy functions to realistic data.

aimalz commented 6 years ago

This issue is being addressed via a migration to pystan, which should help with some other issues as well.