Emulator of spectral models in physics using neural networks.
Project description
specsimile
Neural network emulator for spectral models in physics.
About
This package approximates the spectrum of a parametric physics simulator. The approximation can be a neural network, or, through a intermediate parametric approximate function.
Example: The true physics model is a black body with an area A and temperature T (two parameters). Between 400nm and 500nm, we approximate this model with a power law (normalisation N at 450nm, index p) with two parameters. specsimile learns the best mapping (A,T)->(N,p) to accurately reproduce the spectrum.
The trained emulator is reusable and has infinite resolution.
You can help by testing specsimile and reporting issues. Code contributions are welcome. See the Contributing page.
Usage
Read the full documentation at:
Licence
GPLv3 (see LICENCE file). If you require another license, please contact me.
Release Notes
0.0.1 (2026-03-02)
First version
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
File details
Details for the file specsimile-0.0.1.tar.gz.
File metadata
- Download URL: specsimile-0.0.1.tar.gz
- Upload date:
- Size: 171.8 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.1.0 CPython/3.12.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
77d5254ec07ccc44e2dd472286d612e79059acf51bdc183fcce107e13e5849b0
|
|
| MD5 |
f97b530556bb045ad67929c4462deed6
|
|
| BLAKE2b-256 |
1f0ab89ce7575ec198a02ef6899eb43ac580e6ffd0311a7317384e93c68e1201
|
