Metadata-Version: 2.1
Name: solveminmax
Version: 0.1.2
Summary: A package to solve the sum of min/max equations
Home-page: https://github.com/hegelim/solve-sum-minmax
Author: Yewen Zhou
Author-email: yz4175@columbia.edu
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE

[![Github All Releases](https://img.shields.io/github/license/Hegelim/solve-sum-minmax)]()
[![Github All Releases](https://img.shields.io/github/v/release/Hegelim/solve-sum-minmax)]()
[![Github All Releases](https://img.shields.io/github/downloads/Hegelim/solve-sum-minmax/total)]()
[![Github All Releases](https://img.shields.io/github/issues/Hegelim/solve-sum-minmax)]()  
**About**:  
`solveminmax` is a Python module that allows you to solve a sum of min/max equations 
by taking advantage of the powerful `sympy` library. For instance, say you want to 
solve this equation: min(400, 500x) + min(200, 500x) + min(0, 500x) = 700 
with the assumption that x is within range (0, 1).  
In Math, the rigorous way would 
require you to set up all possible conditions, which 
might result in huge computation. 
Currently, there aren't any available packages in Python
that allows you to solve this kind of equation fastly and efficiently. Thus,
this package is developed to fill the void and hopefully be of use to the broad 
population.  
****
**Quick Start**:  
Let's say you want to solve the equation 
min(500, 600a) + max(400, 500a) = 500. However, solving it in Math means you 
would have to set up the conditions, then solve the check for each one of them, 
which sounds like a lot of work, especially for smart people like you 
who know how to take advantage of existing tools. So you ask yourself,
"What if there is a library that lets me solve it like a piece of cake?" Well, 
there is a library for you now! First off, you need to install it via pip 
in your terminal like below:  
```
pip install solveminmax
```
Then to solve your problem, simply type in these
codes in your Python console: 
```
>>> from solveminmax import solver
>>> eq = "min(500, 600*a) + max(400, 500*a) = 500"
>>> solver.auto_solve(eq, "a")
FiniteSet(1/6)
```
Whola! In fact, this is a pretty complex problem, but 
you just solved it with 3 lines of code. But hold on, what does it mean? 
Let's break it down: the core function 
`auto_solve` takes in two required parameters 
`equation` and `var_name`. `equation` takes in a string of the equation you want to solve 
and `var_name` lets you define your variable with flexibility, such as `"a"`
or `"x"`, although currently, it only supports `"a"`. 
You can also pass in `"low"`, `"high"`, which lets you specify the range 
of your variable. Further details are included in the docstring 
if you are interested.  
****
**Perks**:  
* **Fast**: the module solves a set of complex min and max equations usually 
  under 100 ms, depending on your hardware. 
For example, for an equation as complex as below, it takes 7 ms on average to 
  give you a solution:
  
```
>>> from solveminmax import solver 
>>> eq = "max(600*a, 400) + min(200*a, 500) + min(100, 300*a) + 50*a = 600"
>>> %timeit solver.auto_solve(eq, "a")
FiniteSet(4/11)
7.1 ms ± 225 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)
```
* **Accurate**: it handles exact Rational numbers and intervals. 
* **Flexible**：you have a lot of flexibility in defining your equation, 
  see below. 
****
**Format guide**:  
Because the module depends heavily on regular expressions, please follow 
the guide on how to define your equations carefully, or the module might break.   
In a nutshell, wrap the equation you want to solve in a string with the format
similar to the example: 
```
>>> eq = "20 + max(600*a, 400) + min(200*a, 500) + min(100, 300*a) + 50*a = 600"
```
Before we delve into explanations in details, let's define a few terms: 
* **value_term**: the value you want to solve your equation for, here it's `600`. 
* **minmax_term**: it is what it means in English, for example, `max(600*a, 400)`.
* **cons_var_term**: terms with constants times variables, such as `50*a`.

In brief, what you **can** do include:
* put the variable either in the 1st or 2nd place inside the parenthesis, for 
example, either `min(200, 300*a)` or `min(300*a, 200)` is fine.
* use min and max together in one equation.
* use + and/or -. 
* have constants in front of min or max, such as `2*min(400, 400a)`.
* have any space between each component.
* have leading 0s before variable, such as `min(0*a, 200)`.
* have constants inside min or max, such as `min(20, 30)`.

What you **can't** do include: 
* use `==` instead of `=`.
* for the `cons_var_term`, have variables before constants, such as `a*50` 
instead of `50*a`.
* missing any parenthesis. 
* use other operators instead of + or -.
* missing any necessary `*` operator for each variable.
* put any constants on the left-hand side of the equation. Do me a favor, if 
you have any constants, subtract it from the right-hand side and 
  rearrange your terms before using the module. 
****
**Limitations**:  
* Currently, the module only supports `"a"` as the variable. 
* Because the module heavily depends on regular expressions, 
  the user needs to follow the format of the 
equation carefully, or the module might break.
* The equation must be uni-variate, i.e., there can only be one independent 
variable. 
****
**Version history**: 
Version | Core Ideas | Return Rationals | Return Intervals | Error Handling
------------ | ------------- | ------------- | ------------- | ------------- 
v0.0.1 | numerical methods | No | No | No
v0.0.2 | numerical methods | No | No | No
v0.0.3 | combinations | Yes | No | No  
v0.0.4 | intervals | Yes | Yes | No 
v0.1.0 | intervals | Yes | Yes | Yes 
****
**Contact**:  
* **Email**: yz4175@columbia.edu
* **Collaboration**: collaborations are welcomed, please email me if you 
are interested.


