Math expression evaluation library. It supports most of useful math operations and functions. Expressions can contain variables which can be substituted with ints, floats or numpy.ndarrays.

Example usage

from parser import Parser

parser = Parser("(-b + sqrt(b^2-4a c))/(2a)")

parser.variables_names # {'c', 'a', 'b'}

parser.evaluate({"a": 1, "b": -3, "c": 2}) # 1.0

parser.evaluate({"a": [1, 1, 1], "b": [-5, -6, -9], "c": [6, 9, 20]}) # [2. 3. 4.]

Expression syntax

Expression can contain numbers or variable names with functions applied to them, separated with operators or united with braces. Numbers do not support *-less multiplication with spaces. Variables must be separated by space.

Here are examples with _ as space

Wrong	Right
`2_a`	`2a` or `2*a`
`a_2`	`a*2`
`a_2_a`	`a2a`
`2_cos(a)`	`2cos(a)` or `2*cos(a)`
`cos(a)_2`	`cos(a)*2`
`aa`	`a*a` or `a_a`
-	`cos(a)cos(a)` or `cos(a)_cos(a)` or `cos(a)*cos(a)`

Theese are supported:

Functions:

name	math
`abs`	`\\|x\\|`
`acos`	`\cos^{-1}(x)`
`acosh`	`\cosh^{-1}(x)`
`acot`	`\cot^{-1}(x)`
`asin`	`\sin^{-1}(x)`
`asinh`	`\sinh^{-1}(x)`
`atan`	`\tan^{-1}(x)`
`avg`	`\overline X`
`cos`	`\cos(x)`
`cosh`	`\cosh(x)`
`cot`	`\cot(x)`
`exp`	`\exp(x)`
`inf`	`\inf(X)`
`lg`	`\lg(x)`
`ln`	`\ln(x)`
`log10`	`\log_{10}(x)`
`log2`	`\log_2(x)`
`max`	`\sup(X)`
`min`	`\inf(X)`
`prod`	`\displaystyle \prod_{i=0}^n x_i`
`sgn`	`sgn(x)`
`sign`	`sgn(x)`
`sin`	`\sin(x)`
`sinh`	`\sinh(x)`
`sqrt`	`\sqrt{x}`
`sum`	`\displaystyle\sum_{i=0}^n x_i`
`sup`	`\sup(X)`
`tan`	`\tan(x)`
`tanh`	`\tanh(x)`

Operators: +, -, *, /, ^, %

Braces: (), [], {}

Floating points: ., ,

Functions have only one argument, provided in braces. Operators must have two operands except minus (if it is the first character of equation or braced expression).

avg, sum, max, sup, min, inf and prod applied on numpy.ndarray produce float.

! There is no error handling yet !