https://github.com/neurodata/uncertainty-forest
Conditional probability, conditional entropy, and mutual information estimation in Python. https://arxiv.org/abs/1907.00325
https://github.com/neurodata/uncertainty-forest
Last synced: about 1 year ago
JSON representation
Conditional probability, conditional entropy, and mutual information estimation in Python. https://arxiv.org/abs/1907.00325
- Host: GitHub
- URL: https://github.com/neurodata/uncertainty-forest
- Owner: neurodata
- Created: 2019-11-27T18:55:38.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2020-10-15T01:56:40.000Z (over 5 years ago)
- Last Synced: 2025-04-03T15:04:53.638Z (about 1 year ago)
- Language: Jupyter Notebook
- Size: 63.5 MB
- Stars: 12
- Watchers: 6
- Forks: 5
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Uncertainty Forest
This repo contains code and demos of estimation procedures for posterior distributions, conditional entropy, and mutual information between random variables `X` and `Y`.
- [Overview](#overview)
- [System Requirements](#system-requirements)
# Overview
To reproduce any of the figures, navigate to the corresponding directory, and run the Jupyter notebook.
```
cd figs/fig1
jupyter nbconvert --to notebook --inplace --execute figure-1.ipynb --ExecutePreprocessor.timeout=-1
```
Commands are similar for Figures 2 and 3. The application and hypothesis test code can be found in the `figs/application` director. The above commands convert the notebook to a Python file and produces the figures as PDFs. An alternate option is to open to the notebook, and select "Restart and Run All".
# System Requirements
## Hardware requirements
UF requires only a standard computer with enough RAM to support the in-memory operations.
## Software requirements
### OS Requirements
The code has been tested on the following systems:
+ macOS: Mojave (10.14.1)
+ Linux: Ubuntu 16.04
### Python Dependencies
The code mainly depends on the Python scientific stack.
```
numpy
scipy
scikit-learn
joblib
matplotlib
seaborn
tqdm
```