A user-friendly machine learning Bayesian inference library
Project description
![Downloads tags](https://img.shields.io/github/downloads-pre/hagabbar/vitamin_b/0.1.0/total) ![Commits](https://img.shields.io/github/commit-activity/m/hagabbar/vitamin_b) ![License](https://img.shields.io/github/license/hagabbar/vitamin_b) ![GitHub Stars](https://img.shields.io/github/stars/hagabbar/vitamin_b?style=social) ![Twitter](https://img.shields.io/twitter/follow/gw_hunter?style=social) [![Build Status](https://travis-ci.org/hagabbar/vitamin_b.svg?branch=master)](https://travis-ci.org/hagabbar/vitamin_b)
# [VItamin_B: A Machine Learning Library for Fast Gravitational Wave Posterior Generation](https://arxiv.org/abs/1909.06296) :star: Star us on GitHub — it helps!
Welcome to VItamin_B, a python toolkit for producing fast gravitational wave posterior samples.
This [repository](https://github.com/hagabbar/vitamin_b) is the official implementation of [Bayesian Parameter Estimation using Conditional Variational Autoencoders for Gravitational Wave Astronomy](https://arxiv.org/abs/1909.06296).
Hunter Gabbard, Chris Messenger, Ik Siong Heng, Francesco Tonlini, Roderick Murray-Smith
Official Documentation can be found at [https://hagabbar.github.io/vitamin_b](https://hagabbar.github.io/vitamin_b).
Check out our Blog (to be made), [Paper](https://arxiv.org/abs/1909.06296) and [Interactive Demo](https://colab.research.google.com/github/hagabbar/OzGrav_demo/blob/master/OzGrav_VItamin_demo.ipynb).
Note: This repository is a work in progress. No official release of code just yet.
## Requirements
VItamin requires python3.6. You may use python3.6 by initializing a virtual environment.
` virtualenv -p python3.6 myenv source myenv/bin/activate pip install --upgrade pip `
Make sure to install basemap prior to installing all other packages.
For installing basemap: - Install geos-3.3.3 from source - Once geos is installed, install basemap using pip install git+https://github.com/matplotlib/basemap.git
For other required packages: ` pip install -r requirements.txt `
Install VItamin using pip: ` pip install vitamin-b `
## Training
To train an example model from the paper, try out the [demo](https://colab.research.google.com/github/hagabbar/OzGrav_demo/blob/master/OzGrav_VItamin_demo.ipynb).
Full model definitions are given in models directory. Data is generated from gen_benchmark_pe.py.
## Results
We train using a network derived from first principals: ![](images/network_setup.png)
We track the performance of the model during training via loss curves: ![](images/inv_losses_log.png)
Finally, we produce posteriors after training and other diagnostic tests comparing our approach with 4 other independent methods:
Posterior example: ![](images/corner_testcase0.png)
KL-Divergence between posteriors: ![](images/hist-kl.png)
PP Tests: ![](images/latest_pp_plot.png)
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for vitamin_b-0.2.1-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 61af4542f066efc0234d5b8bea53c79518e134e216d9a32962d88bfd3f569948 |
|
MD5 | e3a15d719d3cd102a91b8a5faadb2596 |
|
BLAKE2b-256 | 0698fa63ef764bbfda1a6b087b48aa59a6cd2553206e7f7f6adea6cd2232d62e |