Getting Started

Installation

To use BOCoDe, first install it using pip:

(.venv) $ pip install bocode

Quick Start

Here’s a simple example of how to use BOCoDe:

import bocode
import torch

# Retrieve all available benchmark problems in BOCoDe by searching with no filters
all_problems = bocode.filter_functions()
print(all_problems)

# Instantiate a synthetic benchmark problem
problem = bocode.Synthetics.Ackley()

# Evaluate at a point
x = torch.Tensor([[0.0] * problem.dim])
values, constraints = problem.evaluate(x)

print(f"Ackley function value at origin: {values[0]}")

Basic Concepts

BOCoDe provides a standardized interface for various benchmark problems. Each problem has:

• An objective function to be optimized

• Input dimension (number of decision variables)

• Bounds on the decision variables

• Optional constraints that must be satisfied

• Known optimum value (for some problems)