-
First of all, thanks for this awesome library ! I have been going through the Rethinking translation (along with the book), it's been great.
I have developed a bayesian linear regression model, th…
-
Here's an example memory breakdown showing a minimal DEzs MCMC configuration where the chains are of length 100k showing the mcmcOut object is using 1.2GB of memory:
```R
mcmcOut = runMCMC(bayesianS…
-
I am wondering why geoopt requires retraction to be 'reversible' (not sure what this refers to) for samplers to work. Why wouldn't, for example, SGRLD work on a sphere? The sphere manifold (whose retr…
-
Hi,
I'm attempting to get dynesty to converge "nicely". My model has 19 parameters and I run it with:
```python
ndim = 19
dsampler = dynesty.DynamicNestedSampler(ln_like, prior_transform, ndim =…
-
Dear Paul:
I can usually get to the bottom of problems with R packages but this one has me stumped. brm loads okay and accepts commands but quits when it is trying to compile. Below I'm attaching t…
-
Hi, on a local cluster I installed cobaya with pip and auto-installed CAMB code and cosmo data with cobaya-install (except for clik which needed a bit more work). My planck-likelihood (coming from the…
-
I think the current dragging algorithm is invalid (returns incorrect results): I think there should be drag_interp_steps+1 in both factors at https://github.com/CobayaSampler/cobaya/blob/master/cobaya…
-
Hi! Recently the Turing and ArviZ teams joined forces to implement a unified, as-lightweight-as-possible implementation of diagnostics for MCMC-generated samples. We started with [MCMCChains's collect…
-
Currently, random.random() is called for uniform random numbers, and normal variables are generated via a polar method implementation. Switching these to numpy's equivalent functions should improve pe…
-
Hi, this is a great software that I am trying to incorporate into my own project to run multiple MCMC chains in parallel. However I am facing a problem with the program just freezing after processing,…