search

stan-dev / posteriordb

Database with posteriors of interest for Bayesian inference
176 stars 36 forks source link

Duplicate eight schools model #139

Closed bbbales2 closed 4 years ago

bbbales2 commented 4 years ago

eight_schools and 8schools appear to be the same model

bbbales2 commented 4 years ago

Other duplicates are:

arma and arma-arma11

pkpd and one_comp_mm_elim_abs-one_comp_mm_elim_abs

low_dim_gauss_mix_collapse and low_dim_gauss_mix_collapse-low_dim_gauss_mix_collapse

MansMeg commented 4 years ago

Yes. This is the case. There are posterior with aliases. (arma, pkpd) to simplify quick access to specific compute comparison models from Michaels repo.

How would you like the API to work? Maybe have an include aliases in the names() function?

A quick fix is to only use the posteriors that have a "-". No alias has those dashes.

MansMeg commented 4 years ago

@bbbales2, do you have a take on this? What do you think would be the best way to handle alias so they dont end up as duplicates for you?

bbbales2 commented 4 years ago

How would you like the API to work? Maybe have an include aliases in the names() function?

I think the answer is the default way to iterate through all models shouldn't give duplicates.

And by default way I mean whatever I did, which maybe was bad lol. I vaguely remember I just tried to copy the tutorial.

So whatever the tutorial does just shouldn't give duplicates. If it already is that way and I just messed things up, I wouldn't be surprised.

MansMeg commented 4 years ago

Now this has been solved with alias.