Jason Pekos

`mcmc_pairs()` works differently on an HPC cluster?

3

I can create pairs plots fine on my laptop. ```{r} mcmc_pairs(mcmc_out$draws(), np = nuts_params(mcmc_out)) ``` When I run the identical code on the [Sharcnet](https://www.sharcnet.ca/my/systems/show/114) HPC system, if I include `np...

Update ODE models to Stan 2.24's solvers, as current method is deprecated.

6

Diffeqs should perhaps updated to use the new `ode_rk45 `. I can take a look at this personally, if you want. Sorry for the initial blank issue, I hit enter...

Adds a folder for Turing models, and a few models for testing,

3

## Purpose This PR adds a few Turing models: - `eight_schools_noncentered` and `eight_schools_centered`. (chosen because they're classic) - `earn_height` (chosen because the data needed for this model is a subset...

## Rationale: It would be handy to have some sort of infrastructure for ensuring model implementations in different PPLs are equivalent. ## Idea: Evaluate the log-density over the reference draws,...

## Proposal: modify the posterior `.json` files to specify what data from the dataframe is actually used as an input to the model. ## Rationale: Some models only use a...

Just found out today that Turing has a BinomialLogit: https://github.com/TuringLang/Turing.jl/blob/03eba139853d950119821fbcac847b43b0caaf06/src/stdlib/distributions.jl#L45 This is pretty useful, and so should be in the docs, either as part of a tutorial, or in a...

add new section in HMM tutorial

2

This adds a section to the HMMs tutorial that has a much faster and more concise version of the model (four lines!) using [HiddenMarkovModels.jl](https://github.com/gdalle/HiddenMarkovModels.jl). There's a subsection after that which...