This release is a pre-release and may not be stable for production use.
Machine Learning for Gravitational Waves from Binary Neutron Star mergers
This package's purpose is to speed up the generation of template gravitational waveforms for binary neutron star mergers by training a machine learning model on a dataset of waveforms generated with some physically-motivated surrogate.
It is able to reconstruct them with mismatches lower than 1/10000, with as little as 1000 training waveforms; the accuracy then steadily improves as more training waveforms are used.
Currently, the only model used for training is TEOBResumS,
but it is planned to introduce the possibility to use others.
The documentation can be found here.
Installation
To install the package, use
pip install mlgw-bns
For more details see the documentation.
Changelog
Changes across versions are documented since version 0.10.1 in the CHANGELOG.
Inner workings
The main steps taken by mlgw_bns to train on a dataset are as follows:
- generate the dataset, consisting of EOB waveforms
- decompose the Fourier transforms of the waveforms into phase and amplitude
- downsample the dataset to a few thousand points
- compute the residuals of the EOB waveforms from PN ones
- apply a PCA to reduce the dimensionality to a few tens of real numbers
- train a neural network on the relation between the waveform parameters and the PCA components
After this, the model can reconstruct a waveform within its parameter space, resampled at arbitrary points in frequency space.
In several of the training steps data-driven optimizations are performed:
- the points at which the waveforms are downsampled are not uniformly chosen: instead, a greedy downsampling algorithm determines them
- the hyperparameters for the neural network are optimized, according to both the time taken for the training and the estimated reconstruction error, also varying the number of training waveforms available.
1 maintainer
Release files for mlgw_bns 0.11.0a1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| mlgw_bns-0.11.0a1.tar.gz | 1.8 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| mlgw_bns-0.11.0a1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 3.6 MB
Release files / mlgw_bns-0.11.0a1.tar.gz
| Download URL | mlgw_bns-0.11.0a1.tar.gz |
|---|---|
| Size | 1.8 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
cbfa08a3a0fee98041ae8faf7b9ce50fbcfdfde86ca28521507381e48d3294f1
|
|
BLAKE2b-256 checksum How to use checksums |
bbe49e80e77f77dd93d3fdaefa3a62aa067f0f0761018ddc5115f3b1e2495c05
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/1.1.14 CPython/3.9.11 Linux/5.15.0-10033-tuxedo
|
Release files / mlgw_bns-0.11.0a1-py3-none-any.whl
| Download URL | mlgw_bns-0.11.0a1-py3-none-any.whl |
|---|---|
| Size | 1.8 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
23af898c106c7084240b48356aa540c965e75753510d8345a606ea4256365696
|
|
BLAKE2b-256 checksum How to use checksums |
10382478907d7e8046f946f8ebd46a25a80f1f83155dc61322cc4774b65d727e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/1.1.14 CPython/3.9.11 Linux/5.15.0-10033-tuxedo
|
