GithubHelp home page GithubHelp logo

metab0t / pyoptinterface Goto Github PK

View Code? Open in Web Editor NEW
161.0 4.0 6.0 3.14 MB

Efficient modeling interface for mathematical optimization in Python

Home Page: https://metab0t.github.io/PyOptInterface/

License: Other

CMake 1.00% C++ 66.07% Python 32.28% Julia 0.63% AMPL 0.02%
conic-programs linear-programming mathematical-modelling mathematical-programming mixed-integer-programming modeling-language optimization python quadratic-programming

pyoptinterface's Issues

求和方法

您好！我对于求和的部分还有些疑问，如果您能不吝赐教，我将非常感激。

第一个问题关于求和方式。我从您的文档中学到了两种求和的方法，第一种是用ExprBuilder，比如

expr = poi.ExprBuilder()
for y in range(100):
expr += block.μ[y]

这里block是一个SimpleNamespace

第二种是用poi.quicksum。从您的例子中看主要是用在tupledict中。

我的问题是，如果我想求若干变量之和，用哪种方法比较好？

比如 x = [model.add_variable(lb=0) for _ in range(10)]，然后我想限制这些x的和小于等于10。那我应该写：

expr = poi.ExprBuilder()
for i in range(10):
expr += x[i]
model.add_linear_constraint(expr, poi.Eq, 10)

还是：

model.add_linear_constraint(poi.quick_sum(x), poi.Eq, 10)

第二个问题是quicksum与quicksum_，这两个很相似函数的适用类型分别是什么？他们有什么区别？

谢谢！

Advice re performance

Hey, nice library! I'm having a play with it to see how it performs with a very simple model with varying numbers of variables.

With the script below I'm getting these results

100 vars:	2.49 ms
1000 vars:	9.67 ms
10000 vars:	717.49 ms
20000 vars:	2794.24 ms
50000 vars:	17723.34 ms

It's looking like the runtime is very non-linear. The same problem in other similar solvers tends to increase quite linearly.
Profiling this via kernprof -l -v example.py (pip install line_profiler first and decorate run with @profile) I see that the bulk of the time seems to be spent in the optimisation:

Timer unit: 1e-06 s

Total time: 20.9067 s
File: example.py
Function: run at line 6

Line #      Hits         Time  Per Hit   % Time  Line Contents
==============================================================
     6                                           @profile
     7                                           def run(n_vars: int):
     8         5          2.0      0.4      0.0      t1 = time.perf_counter()
     9         5        604.0    120.8      0.0      model = highs.Model()
    10         5        184.0     36.8      0.0      model.set_raw_parameter("log_to_console", False)
    11         5          6.0      1.2      0.0      model.set_raw_parameter("output_flag", False)
    12                                           
    13         5     129217.0  25843.4      0.6      x = model.add_variables(list(range(n_vars)), lb=0.0, ub=1.0)
    14         5      10588.0   2117.6      0.1      model.add_linear_constraint(poi.quicksum(x), poi.Eq, 1.0)
    15         5      31477.0   6295.4      0.2      objective = poi.quicksum([i * x[i] for i in range(n_vars)])
    16         5       1035.0    207.0      0.0      model.set_objective(objective, poi.ObjectiveSense.Maximize)
    17         5   20733173.0 4146634.6     99.2      model.optimize()
    18                                           
    19         5        276.0     55.2      0.0      assert math.isclose(model.get_value(objective), n_vars - 1)
    20         5          6.0      1.2      0.0      t2 = time.perf_counter()
    21         5        123.0     24.6      0.0      print(f"{n_vars} vars:\t{(t2 - t1) * 1000:.2f} ms")

Do you have any advice on whether there's any change I could make to my code to improve it?

import math
import time
import pyoptinterface as poi
from pyoptinterface import highs


def run(n_vars: int):
    t1 = time.perf_counter()
    model = highs.Model()
    model.set_raw_parameter("log_to_console", False)
    model.set_raw_parameter("output_flag", False)

    x = model.add_variables(list(range(n_vars)), lb=0.0, ub=1.0)
    model.add_linear_constraint(poi.quicksum(x), poi.Eq, 1.0)
    objective = poi.quicksum([i * x[i] for i in range(n_vars)])
    model.set_objective(objective, poi.ObjectiveSense.Maximize)
    model.optimize()

    assert math.isclose(model.get_value(objective), n_vars - 1)
    t2 = time.perf_counter()
    print(f"{n_vars} vars:\t{(t2 - t1) * 1000:.2f} ms")


for n_vars_ in (100, 1_000, 10_000, 20_000, 50_000):
    run(n_vars_)

jupyter kernel crashes with doc example 

import pyoptinterface as poi
from pyoptinterface import highs

model = highs.Model()

x = model.add_variable(lb=0, ub=1, domain=poi.VariableDomain.Continuous, name="x")
y = model.add_variable(lb=0, ub=1, domain=poi.VariableDomain.Integer, name="y")

con = model.add_linear_constraint(x+y, poi.Geq, 1.2, name="con")

obj = 2*x
model.set_objective(obj, poi.ObjectiveSense.Minimize)

model.set_model_attribute(poi.ModelAttribute.Silent, False)
model.optimize()

print(model.get_model_attribute(poi.ModelAttribute.TerminationStatus))
# TerminationStatusCode.OPTIMAL

x_val = model.get_value(x)
# 0.2

y_val = model.get_value(y)
# 1.0

When running this example, Jupyter kernel crashes in model.optimize()

The Kernel crashed while executing code in the current cell or a previous cell.
Please review the code in the cell(s) to identify a possible cause of the failure.
Click here for more info.
View Jupyter log for further details.

17:39:18.153 [error] Disposing session as kernel process died ExitCode: 3221225477, Reason:

Windows 11, VSCode, Python 3.11.6

Callback support for HiGHS

Following our previous discussion on HiGHS repo, can callback be supported for HiGHs as well?
I can help with testing.

Pylance无法识别某些类方法

您好！ 我对你的包很感兴趣，最近在尝试用。我目前发现了一个小问题。可能poi (pyoptinterface)与 pyoptinterface.highs.Model的有些属性是动态生成的？比如：

model = highs.Model()
model.add_linear_constraint(...)

这段代码中，pyplance无法识别add_linear_constraint这个类方法。可能这个方法是用C++代码动态生成的？

但这样会提高在撰写代码时的错误率，因为pylance无法识别该方法，无法自动补全，或者将之用代表类方法的颜色高亮显示。

不知道您有没有计划解决这个小问题。

同样的问题也出现在poi.VariableDomain

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.