##### /
###### /
/# / /
/ / /
/ /
## ## /## ### /### /### ### /### /### /###
## ## / ### ##/ ###/ /## / ##/ ###/ /## / / ### /
## ## / ### ## ###/ ###/ ## ###/ ###/ / ###/
## ## ## ### ## ## ## ## ## ## ## ##
## ## ######## ## ## ## ## ## ## ## ##
# ## ####### ## ## ## ## ## ## ## ##
/ ## ## ## ## ## ## ## ## ##
/##/ / #### / ## ## ## ## ## ## ## /#
/ ############/ ######/ ### ### ### ### ### ### ####/ ##
/ ######### ##### ### ### ### ### ### ### ### ##
#
##
An extensible mini-language to generate mathematical notation for executable and testable Python.
TL;DR:
- Write your mathematical formula with Lemma
- Execute them from Python code
- Output formula definitions in LaTeX maths notation
- Ensures your implementation and documentation match
- Great for reproducible research and teaching/learning
Follow the tutorial on Binder:
Then check out:
- Define mathematical formulae that can be formatted as LaTeX and executed as functions.
- Check all of the steps in your equation transformations produce the same results in software.
- Runs on Python - you can use defined formulae in your existing Python projects, and use Python libraries from your formulae.
- Fully extensible - built with Hy (a Lisp running on Python) to provide powerful tools for defining your own mathematical operations.
- Supports reproducible research by ensuring the notation in your paper matches the behaviour of your (testable) code.
- Enables developers to work through math in a more exploratory way. Use an equation to record the steps in your step-by-step algebraic transformations, and write test-cases to check your working. Then, get the LaTeX to document your work.
- Bridges the gap between mathematical notation and code to simplify teaching in domains that depend on both.
- handcalcs
- handcalcs has a similar goal of producing LaTeX notation for your Python code, but it takes the approach of directly interpreting Python syntax. It supports some common mathematical notation, but does not appear to have the same focus on user extensibility (to new math domains) and customisability (for fine-grained control of generated notation) that Lemma does.
- latexify
- Likewise, latexify directly compiles a Python function into LaTeX, but it seems to have many of the same limitations in extensibility and customisability as handcalcs.
- Mathematica/Wolfram Language
- The Wolfram Language has similar motivations to Lemma around bridging the gap between mathematical notation and executable code. You can generate LaTeX from your code, but it's a proprietary language. Also, because Lemma is built on top of Python, it can work with your existing Python code and libraries.
- SymPy
- Like the Wolfram Language, SymPy is designed for symbolic computation. You can generate LaTeX from your expressions, but it seems that control over the notation is limited (expressions are automatically simplified, and LaTeX formatting options are controlled by keyword arguments that apply to the entire expression). Lemma is designed to be extensible so that you can define exactly how you want the LaTeX to be generated for your use case, and so that you can provide notation for any Python code you like (not just symbolic computation). A library for using SymPy's symbolic computation powers from Lemma would be a good extension though...
- pytexit
- Translates a string of Python code to LaTeX. Limited features supported, and requires managing Python code in strings.
- LaTeXCalc
- Interprets and executes LaTeX math notation. Limited library of math functions available.
- Penrose
- While Penrose is a math language for generating diagrams, while Lemma generates LaTeX notation and executable code.
- Write tutorial
- Add more operators to
algebra.hy - Add unit tests