symeval 0.3.0

Symbolic evaluation for engineering calculations — renders expression → numbers with units → result as LaTeX

symeval

Idea

The idea is to build something like handcalcs and Calcpad, but then for Python notebooks (marimo & jupyter) using sympy symbolic math expressions and pint quantities (value + unit) as a starting point, such that we can create Explorable Explanations in Python notebooks.

symeval adds two methods to a sympy expression that allows you to fill in pint quantities (value + unit). These methods are:

• quant_evalf: evaluate a sympy expression using pint units. (I don't remember why exactly, but I decided to not use sympy quantities for some reason, I think because sympy quantities can't handle things like GPa at all or intuitively and because pint units can be rendered nicely using pint's formatter "~L". Could that be?) This is especially useful when you want to calculate the quantities of a new column in a dataframe in a unit-aware manner.
• sym_evalf: evaluate a sympy expression using pint units and return the LaTeX expression of the filled in formula.

Shortly explain what's great about handcalcs, i.e. it's raison d'etre, here. Then shortly explain that it isn't great that it relys on Jupyter cell magic, i.e. what the disadvantages of Jupyter cell magic are (such as the inability to render variables when doing e.g. a = x1 where x1 was calculated in another cell).

Advantages of using sympy:

• ability to use symbolic math with sympy before plugging in the numbers
• ability to apply .evalf or .quant_evalf to a dataframe of values.

Inspiration

• handcalcs: GitHub
• Calcpad: Website; Docs; GitHub
• sympy: Website; Docs; GitHub
• Bret Victor's Explorable Explanations