Monte Carlo sampling notebook
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
README.md
Sampling.html
Sampling.ipynb

README.md

This is a notebook with some sampling algorithms implemented in numpy/autograd. You will need Python 3, matplotlib, numpy, scipy, and autograd to run them.

I wanted to get a better understanding of MCMC parameter tuning, especially Hamiltonian Monte Carlo, so I decided to implement these routines myself to observe their dynamics. These implementions are unoptimized and unaccelerated but relatively clear and therefore worth sharing.

Please let me know if you have any notes or suggestions by making a pull request or contacting me via the details on my profile.