noether 0.1.2.post2

Work with physical measurements and constants

Noether (in development: Alpha 0.1.2)

The problem: million-dollar errors are easy to make

We're all human. Whether you're a layman, a student or a NASA senior, it's immensely easy to make simple mistakes, especially when working with physical measurements. A good programming language will shout at you if you try to add a list to a number, so why shouldn't it stop you adding a metre to a minute? Enter Noether.

pip install noether&& python3.7 -im noether

You can use Noether as a calculator or in simulations (though at the moment it's not exactly performant). Noether is intended to compliment numpy, matplotlib and scipy with

• A disturbingly extensive and constantly-updated catalogue of scientific units, measurements and constants
• Automatic unit conversion, and propagation of dimension and statistical error

In the pipeline:

• Advanced support for more sophisticated statistical errors
• Treat functions like code with ASTley, and automagically graph them in a REPL
• Use Astley's code analysis with Numba to ensure both code safety and performance

Units

BMI, bushel and byte; eotvos, erg and electronvolt; parsec, plethron and potrzebie. Noether has (almost) every unit known, with a thorough unit system allowing for automatic dimensional analysis.

>>> 0.5 * gibibyte / second
4.295×10⁹bps <data rate>
>>> a = 12 * pixel / mm
>>> a
1.2×10⁴pix·m⁻¹ <image quality>
>>> display(ppi)
>>> a
304.8ppi <image quality>
>>> (joule / kilogram / kelvin)(12, 2)
12 ± 2K⁻¹·m²·s⁻² <specific heat capacity>
>>> e / electron.mass
(2.187×10¹¹ ± 1649)A·kg⁻¹·s

You may, of course, define your own units and dimensions on the fly:

>>> FF = Unit(furlong / fortnight, 'ff')
>>> display(FF)
>>> c
1.803×10¹²ff <velocity, speed>

>>> health = Dimension.new('health', dimsym='H', unitsym='a')
>>> apple = Unit(health)
>>> apple / day
1.157×10⁻⁵a·s⁻¹

Installation and use

Noether currently requires Python 3.7 or above. It's easy to jump into a quick calculator-like instance quick:

pip install noether
python -im noether

This simply runs import noether as noe and from noether import *. We'll assume the latter has been done; it makes for a neater experience when working with units.

Some yet-undocumented submodules may happen to require astley, numpy, matplotlib and scipy to use:

pip install numpy matplotlib
git clone https://github.com/yunruse/astley
# modify your PATH as needed

Update log

Updates are released to the master branch every couple of weeks. They are not guaranteed to have a sturdy, documented and tested API just yet, but each release should be pretty dang stable.

Every single change I plan to make can be found on my kanban board. Thanks to internet magic, this appears live, exactly as I see and edit. If you have a feature suggestion, drop me an issue on GitHub. Bug fixes will stay in GitHub: they're much more important than features.

• Alpha 0.1.2 (2020-05-07):

  • Added basic config support.
  • This is the first PyPI release.

• Alpha 0.1.1 (2020-02-16):

  • Many bugfixes and units added.
  • Added noether.fraction with ContinuedFraction (and Fraction with shortcut).
  • Introduced basic backend for UnitSystem for quality of life (eg display(imperial)).
  • Made special tweaks so python -m noether lazy-loads at fast pace.

• Alpha 0.1 (2020-01-26):

  • Made many improvements to display of units.
  • The global display was added, alongside a wide range of particles.
  • A test suite begins development.

Thank you!

This project has a decently large scope, and I want it to appeal to anybody who has ever needed to measure anything. Your feedback, no matter how small, means a lot to me!

Legal (and academical)

Copyright (c) Mia yun Ruse (yunru.se) 2018–2020. All scientific data shown remains in the public domain, and is cited where relevant.

This work is licensed under a Creative Commons Attribution 4.0 International license. In non-legal terms: do whatever you like, including science! But if you want to redistribute this project (derived or not), please credit me, or shoot me a message and I'll see what I can do.