pints
pints copied to clipboard
pints-team
Probabilistic Inference on Noisy Time Series
Useful to diagnose whether slow convergence is being caused by step sizes being too large or small etc.
As requested by @mirams Sounds sensible
When models are close to unidentified, in adaptive covariance MCMC, the covariance matrix can become rank-deficient. If that occurs, the linear algebra used to form it may fail and halt...
These are essential for the paper as they are ways to check a model's fit to data. We already do a bit of this in notebooks but we should do...
This is a sampling diagnosis plot.
- [ ] AR1LogLikelihood - [ ] ARMA11LogLikelihood - [ ] GaussianIntegratedUniformLogLikelihood #763 - [ ] CauchyLogLikelihood - [x] GaussianKnownSigmaLogLikelihood - [x] GaussianLogLikelihood - [ ] MultiplicativeGaussianLogLikelihood - [x] ScaledLogLikelihood...
I constantly need to check the shape of priors that I am using and so need to do: - values = np.linspace(min, max, 1000) - y = map(lambda x: log_prior(x),...
In Neal [paper](https://projecteuclid.org/download/pdf_1/euclid.aos/1056562461): - [x] univariate slice sampling - [x] slice-sampling-within-Gibbs multivariate slice - [x] multivariate slice sampling - [x] over-relaxed slice sampling for univariate and multivariate cases (some of...