-
Blocked by Pyro HMC refactoring https://github.com/pyro-ppl/pyro/issues/1816
Initially we can simply wrap Pyro's HMC and NUTS implementations. After NumPyro's distributions are closer to torch.dist…
-
Hello Juan,
I found your post "Flax and Numpyro Toy Example" very cool and interesting. As a PYMC user I wanted to see if I could achieve similar results using PYMC. You can see my implementation at …
-
May be out of scope for this project (or just for now), but @SamuelBrand1 points us to `epidemia`'s interface for counterfactuals, which I think is cool. This is somewhat straightforward to implement,…
-
Given the variety of backgrounds and experiences present across CFA and the fact that CFA members seem, at present time, to constitute most of MSRs early users, I suggest we explicitly write out the a…
-
The code below runs without exceptions, is it desired?
```python
w1 = normal_scale(1, 1)
x = bernoulli(w1)
y = 1
calc = Calculate("numpyro",niter=10000)
ys = calc.sample(w1, [x, x, x], [1, 0, …
-
@thomassargent30 @smit-create, I really enjoyed reading the lecture https://python.quantecon.org/ar1_bayes.html
Some minor suggestions:
It is not clear from reading the lecture why we are using …
jstac updated
2 years ago
-
Paper: [Hierarchical Models for Causal Effects](https://onlinelibrary.wiley.com/doi/abs/10.1002/9781118900772.etrds0160)
-
Hi!
I am having an issue with initialising a model with the init_by_median() strategy for my MCMC NUTS model.
I am drawing the location parameter of a VonMises distribution from a VonMises prior a…
-
### Description
@aseyboldt gave a nice explanation here that we could probably integrate in the docstrings: https://github.com/pyro-ppl/numpyro/pull/1751#issuecomment-1980569811
-
BayesBlend currently works with CmdStanPy models, and ArviZ. Users can, in theory, use any modelling pipeline compatible with ArviZ (e.g. PyStan, CmdStanPy, PyMC, NumPyro etc.), and use ArviZ as the b…