-
There are a few enhancements to Replay Poutine that could simplify the implementation of MCMC algorithms:
1. Return param site from the replayed trace instead of from the param store, by implementin…
-
Original ticket https://networkx.lanl.gov/trac/ticket/470
Reported 2010-11-17 by maciej.kurant, assigned to @hagberg.
Many real-life graphs are too big to be downloaded and analyzed in their full siz…
-
Make sure that these boxes are checked before submitting your issue -- thank you!
- [ X ] This issue is for the C core of igraph.
- [ X ] This issue is a bug report or a feature request, not a su…
-
Hamiltonian Monte Carlo (HMC), also known as Hybrid Monte Carlo, is an efficient Markov Chain Monte Carlo (MCMC) method that exploits gradient information to improve on the simpler MCMC methods. See t…
-
For various reasons I'm using the Metropolis-Hastings sampler in emcee. I'm passing it a dict of (many) parameters, and also have to pass it a covariance matrix for the proposal distribution, which pr…
-
Hi all,
Thanks to pymc's developers for putting together such an accessible library for non-statisticians like me. That statement serves as a disclaimer as well: I am a scientist, not a statisticia…
-
The next proposed pull request is for an example implementing particle marginal Metropolis-Hastings for the non-linear state space model described in Section 3.1 of Andrieu et al. (2010). This example…
-
## Description
* [ ] Set jump covariances to zero (i.e. use diagonal jump variance-covariance matrix) until after `n` iterations, as set by the user.
* [ ] Randomly shuffle jump covariance matrix am…
-
- [ ] 2 Mathematical MRF Models
- [x] Ising Model
- [x] Auto-Logistic Model
- [ ] Auto-Binomial Model
- [ ] Auto-Normal Model
- [x] Multi-level Logistic Model (MLL)
- [ ] 3.2 Image Restora…
-
The documentation has a very short section on this (too short to be useful as is). It should be augmented.