cdcapano / epsie Goto Github PK
View Code? Open in Web Editor NEWAn Embarrasingly Parallel Sampler for Inference Estimation
License: GNU General Public License v3.0
An Embarrasingly Parallel Sampler for Inference Estimation
License: GNU General Public License v3.0
The test/
directory is included in the source distribution (tarball) for 1.0.0, but the single file test/_utils.py
is missing, can this be added to MANIFEST.in
so that the full test suite can be executed from the tarball?
The *Eigenvector
proposals sometimes get 0 eigenvalues when run on the test model. This shouldn't happen since the test model is a 2D Gaussian. Example setup that can reproduce:
Model
(with the default seed in this PR)STABILITY_DURATION = 64
ADAPTATION_DURATION = 16
SWAP_INTERVAL = 1
NCHAINS = 4
NTEMPS = 3
ITERINT = 32
nprocs = 1
seed = 266898112101595934258451730772524202746
sampler.run(STABILITY_DURATION)
. Chain 1, coldest temperature, will then have a covariance matrix of:array([[0.17048876, 0.52049453],
[0.52049453, 1.58904647]])
That yields an eigenvalue that is 0.
The issue appears to be that the initial covariance used for the eigenvalues is based on just 5 points. If only two unique points exist, then the estimated covariance will yield only one unique eigenvalue. See the Eigenvector._stability_update
.
The src tarball for epsie-0.4.1 on PyPI doesn't include the LICENSE file, can this be added for future releases (this is a condition of GPL)?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.