xplor provides a unified framework for building Operation Research models using polars DataFrames. By leveraging polars' performance and ergonomic API, xplor makes mathematical optimization more intuitive and maintainable.
Building large-scale optimization models in Python can become inefficient. Standard approaches often require iterating over rows, manipulating Python dictionaries, or using slow list comprehensions, which makes the logic slow and difficult to maintain as the model grows.
xplor solves this core issue by using a Polars-based DataFrame API to construct your model logic. This provides two significant benefits:
- Vectorized Performance: Model operations (like grouping, joining, filtering, and aggregation) are executed using Polars' high-performance Rust backend. You are no longer manipulating slow Python objects*; the logic is executed at native speed.
- Maintainability: By relying on the familiar and expressive DataFrame API, your model logic becomes significantly more readable and maintainable.
Additionally, a huge benefit of xplor is its solver agnosticism. It abstracts away the specific syntax of underlying optimization engines (like Gurobi or MathOpt), allowing you to define a single model and easily benchmark its performance across multiple solvers.
* Clarification: Variable, variable expression, and constraint objects themselves are still Python objects (e.g., gurobi.Var or mathopt.Variable), but the data manipulation and application of model logic are managed by Rust via Polars' User-Defined Functions (UDFs), enabling vectorized performance.
xplor requires Python 3.11+.
Install the base package, which provides the core Polars integration and abstract modeling interface:
xplor comes with few dependencies. To use a specific solver you need to add optional dependency group.
pip install xplor[gurobi, ortools, cplex]or install all of them
pip install xplor[all]Note: Gurobi requires a valid license (academic or commercial) to run. Please follow the official Gurobi documentation for setup instructions.
Note: CPLEX requires a valid license (academic or commercial) to run. Please follow the official CPLEX documentation for setup instructions.
Have a look to the documentation.
xplor is in active development.
- ✅ Gurobi backend
- ✅ OR-Tools backend
- ✅ Hexaly backend
- ✅ CPLEX backend (via docplex)
- 🚧 Support for non-linear expressions.
- 🚧 Support for additional solvers (HiGHS, ...).
Want to contribute? Read our contributing guide.
This project is licensed under the MIT License - see the LICENSE file for details.
- gurobipy-pandas for inspiration.
- polars for the amazing DataFrame library.
- Gurobi for the optimization solver.
- OR-Tools for the optimization solver.
- IBM CPLEX for the optimization solver.