Eero Linna

I guess one option would be to just omit the math notation for those models, in other words make the math notation recommended but not mandatory.

Is the idea that we would use stan2tfp to convert Stan models to tfp models and then add them to posteriordb? (I mean the collection of posteriors and models and...

Here's a pymc3 model ```python import pymc3 as pm3 def model(data): J = data["J"] # number of schools y_obs = data["y"] # estimated treatment sigma = data["sigma"] # std of...

My PR #118 (sorry forgot to reference this issue in it, did that now) is probably the best way to include the pymc model with the current design. I'm not...

Oh you mean the actual data? I'll first show how it would look like using python as R will have to go through python anyway. ```python import importlib import pymc3...

I agree that `run_pymc3` doesn't have to be in posteriordb.

I think the benefit is clearer semantics. I don't know if waiting will be helpful here as these kind of changes really never become strictly necessary. Or perhaps a better...

Is the idea here that we can get more accurate expectations by using more samples? So we might have a gold standard with 10 000 draws but expectations computed with...

So `posteriors/8_schools.json` might be something like this? ```json { "name": "eight_schools-eight_schools_noncentered", "keywords": ["stan_benchmark"], "model_name": "eight_schools_noncentered", "reference_draws_name": "eight_schools-eight_schools_noncentered", "reference_expectations_name": "something", "data_name": "eight_schools", "dimensions": {"theta":8, "mu":1, "tau":1}, "added_by": "Mans Magnusson", "added_date": "2019-08-12"...

Here's some things that popped to my mind. ## 1 Lets say we have a posterior where the expectations computed with 100 000 draws are more accurate than the ones...