pymc
pymc copied to clipboard
pymc-devs
Bayesian Modeling and Probabilistic Programming in Python
When does the use of Potential have an impact on **posterior** predictive sampling? https://github.com/pymc-devs/pymc/blob/6ef135b6cbeaee3c3be1efb7a5af1ddfdbc3c861/pymc/model/core.py?plain=1#L2274-L2278 It seems like sampling distributions only given InferenceData.posterior samples would not be affected by Potentials. Is...
### Before ```python ``` ### After ```python gp = pm.gp.HSGP(m=[25, 25], c=4.0, cov_func=cov_func) f = gp.prior("f", X=X, n_outputs=10) ``` ### Context for the issue: Hi, First of all big fan...
Hi @jessegrabowski , @ricardoV94 ! I've implemented the `rng_fn` for the designated classes `CAR/ICAR` in the issue [#7713](https://github.com/pymc-devs/pymc/issues/7713) , so could you please review and let me know if there...
The models that are now allowed following https://github.com/pymc-devs/pymc/pull/7656 have a disconnected node in the model graph. The sampling is as expected. It is just the graphviz representation that is incorrect....
### Describe the issue: Progress bar throws an error when a nested CompoundStep is found in the sampling flow of a model. Once the progress bar is deactivated, i..e, `progressbar=False`,...
### Description I was talking to @theorashid who linked me to [this case study](https://mc-stan.org/learn-stan/case-studies/icar_stan.html) of CAR priors. It seems like they're just MvNormals, but with degenerate covariance matrices. We can...
`find_contrained_prior` was deprecated in v5.17.0 (https://github.com/pymc-devs/pymc/pull/7458), released on Oct 3, 2024. This PR removes it completely. ---- 📚 Documentation preview 📚: https://pymc--7680.org.readthedocs.build/en/7680/
### Description Suppose I want to make a log-normal "by hand": ```py def d(mu, sigma, size=None): return pt.exp(pm.Normal.dist(mu=mu, sigma=sigma, size=size)) with pm.Model() as m: y = pm.CustomDist('y', 0, 1, dist=d,...
## Description For drawing from an LKJCholeskyCorr distribution. Seems nice, but needs a forward transform. Marked as maintenance because our current transformation breaks for n > about 5. ## Related...
Only model objects will have a default Ipython/Jupyter representation. This simplifies the codebase and saves half a second of import time. Before: ```python import pymc as pm with pm.Model() as...
