https://github.com/junpenglao/glmm-in-python

Generalized linear mixed-effect model in Python
https://github.com/junpenglao/glmm-in-python

bayesian-inference glmm linear-mixed-models pymc3 statistics

Last synced: about 1 month ago
JSON representation

Generalized linear mixed-effect model in Python

• Host: GitHub
• URL: https://github.com/junpenglao/glmm-in-python
• Owner: junpenglao
• License: gpl-3.0
• Created: 2016-06-29T11:32:01.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2018-08-17T05:47:47.000Z (over 6 years ago)
• Last Synced: 2024-08-03T15:06:56.864Z (5 months ago)
• Topics: bayesian-inference, glmm, linear-mixed-models, pymc3, statistics
• Language: Jupyter Notebook
• Homepage:
• Size: 11.6 MB
• Stars: 175
• Watchers: 16
• Forks: 47
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

        
# Generalized Linear Mixed‐effects Model in Python

or the many ways to perform GLMM in python playground

  

A comparison among:  

[StatsModels](https://github.com/statsmodels/statsmodels)  

[Theano](https://github.com/Theano/Theano)  

[PyMC3](https://github.com/pymc-devs/pymc3)(Base on Theano)  

[TensorFlow](https://github.com/tensorflow/tensorflow)  

[Stan](https://github.com/stan-dev/stan) and [pyStan](https://github.com/stan-dev/pystan)  

[Keras](https://github.com/fchollet/keras)  

[edward](https://github.com/blei-lab/edward)  

Whenever I try on some new machine learning or statistical package, I will fit a mixed effect model. It is better than linear regression (or MNIST for that matter, as it is just a large logistic regression) since linear regressions are almost too easy to fit. Hence this collection of codes that all doing (more or less) the same thing.

## TODO

Estimate uncertainty related to model parameter using [dropout](http://mlg.eng.cam.ac.uk/yarin/blog_3d801aa532c1ce.html) in Theano and TensorFlow  

[DROPOUT AS A BAYESIAN APPROXIMATION](http://mlg.eng.cam.ac.uk/yarin/publications.html#Gal2015Bayesian)  

K-Fold Cross Validation and Leave-One-Out (LOO)  

[WAIC and cross-validation in Stan](http://www.stat.columbia.edu/~gelman/research/unpublished/waic_stan.pdf)  

[tyarkoni/PPS2016](https://github.com/tyarkoni/PPS2016)

## More information (codes) could be found below (to name a few):  

[paul-buerkner/brms](https://github.com/paul-buerkner/brms)  

[vasishth/BayesLMMTutorial](https://github.com/vasishth/BayesLMMTutorial)  

[jonsedar/pymc3_vs_pystan](https://github.com/jonsedar/pymc3_vs_pystan)  

[Example from PyMC3](http://pymc-devs.github.io/pymc3/notebooks/GLM-hierarchical.html)  

[tyarkoni/nipymc](https://github.com/tyarkoni/nipymc)  

[bambinos/bambi](https://github.com/bambinos/bambi)