Cardinal Optimizer (COPT) User Guide

Contents

Introduction to Cardinal Optimizer
- Overview
- Licenses
- How to Cite
- Contact Information
Installation Guide
- Registration
- Software Installation
  - Windows
  - MacOS
  - Linux
- Setting Up License
- Verify the installation and configuration
- Upgrade
COPT Command-Line
- Overview
- Edit mode
- Script mode
- Shell commands
  - General shell commands
  - COPT shell commands
- Example Usage
COPT Floating Licensing service
- Server Setup
- Client Setup
- Floating Token Server Managing Tool
  - Tool Usage
  - Example Usage
- Running as service
  - Linux
  - MacOS
COPT Compute Cluster Service
- Server Setup
- Client Setup
- COPT Cluster Managing Tool
  - Tool Usage
  - Example Usage
- Running as service
  - Linux
  - MacOS
COPT Web Licensing Service
COPT Quick Start
- C Interface
  - Example details
  - Build and run
- C++ Interface
  - Example details
  - Build and Run
- C# Interface
  - Example details
  - Build and Run
- Java Interface
  - Example details
  - Build and Run
- Python Interface
- AMPL Interface
- Pyomo Interface
  - Installation Guide
  - Example Usage
- PuLP Interface
  - Installation guide
  - Introduction of features
- CVXPY Interface
  - Installation guide
  - Introduction of features
General Constants
- Version information
- Optimization directions
- Infinity and undefined value
- Constraint senses
- Variable types
- SOS constraint types
- Indicator constraint types
- SOC constraint types
- Exponential Cone type
- Quadratic objective function
- Quadratic constraint
- Basis status
- Solution status
- Client configuration
- Callback context
- API function return code
- Client configuration
- Methods for accessing constants
Attributes
- Problem related
- Solution related
- Methods for accessing attributes
Information
- Problem information
- Solution information
- Dual Farkas and primal ray
- Feasibility relaxation information
- Callback information
- Methods for accessing information
Parameters
- Limits and tolerances
- Presolving and scaling
- Linear programming related
- Integer Programming related
- Semidefinite Programming related
- Nonlinear Programming related
- Multi-objective Optimization related
- Parallel computing related
- GPU computing related
- IIS computation related
- Feasibility relaxation related
- Parameter Tuning related
- Callback related
- Other parameters
- Methods for accessing and setting parameters
Modeling and Solving Optimization Problems
- Linear Programming (LP)
- Second-Order Cone Programming (SOCP)
  - Mathematical Model
  - Modeling
- ExpCone Programming
  - Mathematical Formulation
  - Exponential Cone Example
- Semidefinite Programming (SDP)
- Quadratic Programming (QP)
  - Mathematical Formulation
- Quadratically Constrained Programming (QCP)
- General Nonlinear Programming (NLP)
- Mixed-Integer Programming (MIP)
- Special Constraints
Handling Infeasible Models
- IIS for Infeasible Models
- Feasibility Relaxation for infeasible models
  - Computing Feasibility Relaxation
  - Feasrelax-related parameters, attributes and information
MIP Starts
- Utilities of MIP Starts
  - Set and Load MIP Starts
  - Read and Write MIP Starts
- Related Parameters
- Log of MIP starts
  - MIP starts are accepted
  - MIP starts are rejected
MIP Solution Pool
COPT Tuner
- Introduction
- Related parameters
- Provided capabilities
- Example
Callbacks
- Obtaining information during the solving process
- Controlling the MIP solving process
- Using the callback utilities in different APIs
Matrix Modeling Method
- Two different matrix modeling modes
- Multi-dimensional Variables
- Multi-dimensional array operations and expressions
- Matrix Constraints
  - Matrix linear Constraints
  - Quadratic Constraints
- Objective function composed of multi-dimensional variables
Multi-objective Optimization
- Modeling Multiple objectives
- Solving Multiple objectives
- Retrieving the results of multiple objectives
Logging
- Parameters and Functions for Logging
- Basic Information
- Simplex Method Logging
- Barrier Method Logging
  - Presolve
  - Model Information
- Branch-and-Cut Method Logging
- Logging for First-order method (PDLP) of GPU solver
  - GPU hardware information of the machine
  - 3. First-order method PDLP iteration process
File formats
- File format list
- File I/O operations
- Model file introduction
FAQs
- Installation and Licensing Configuration Related
  - MacOS System
  - Windows System
- Modeling and Solving Functions Related
- GPU Usage Related
C API Reference
- Constants
- Attributes
  - Problem information
  - Solution information
- Information
- Callback information
- Parameters
- API Functions
Python API Reference
- Constants
- Python Modeling Classes
- Helper Functions and Utilities
C++ API Reference
- Constants
- Attributes
- Information
- Parameters
- C++ Modeling Classes
C# API Reference
- Constants
- C# Modeling Classes
Java API Reference
- Constants
- Java Modeling Classes