Search for functions with custom filters

You can search for benchmark functions in BOCoDe using custom filters for dimensionality, number of objectives, number of constraints, and other criteria.

Example Usage

import bocode
from bocode import DataType

# Available DataTypes:
#    DataType.CONTINUOUS
#    DataType.DISCRETE
#    DataType.CATEGORICAL

# Retrieve all available categories
available_categories = bocode.filter_functions().keys()

# Example of filtering functions using all available filters
filtered_functions_list = bocode.filter_functions(
    dimension_filter = lambda dim: dim>=5, # Include only functions with dimensionality of 5 or greater
    input_type_filter = lambda input_type: input_type==DataType.CONTINUOUS, # Include only continuous functions
    objectives_filter = lambda n: n==1, # Include only single-objective functions
    constraints_filter = lambda c: c==0, # Include functions with 0 constraints (Functions with only simple bound constraints)
    category_filter = lambda x: x!="CEC.CEC2020_RW_Constrained" # Exclude all CEC2020 functions
    )

print(filtered_functions_list)

Output:

{'Synthetics': ['bocode.Synthetics.DixonPrice', 'bocode.Synthetics.Griewank', 'bocode.Synthetics.Levy', 'bocode.Synthetics.Michalewicz', 'bocode.Synthetics.Powell', 'bocode.Synthetics.Rastrigin', 'bocode.Synthetics.Rosenbrock', 'bocode.Synthetics.StyblinskiTang', 'bocode.Synthetics.Cosine8', 'bocode.Synthetics.Hartmann6D', 'bocode.Synthetics.SVM'],
 'LassoBench': ['bocode.LassoBench.LassoBreastCancer', 'bocode.LassoBench.LassoDiabetes', 'bocode.LassoBench.LassoDNA', 'bocode.LassoBench.LassoLeukemia', 'bocode.LassoBench.LassoRCV1', 'bocode.LassoBench.LassoSyntHard', 'bocode.LassoBench.LassoSyntHigh', 'bocode.LassoBench.LassoSyntMedium', 'bocode.LassoBench.LassoSyntSimple'],
 'Engineering': ['bocode.Engineering.RobotPush', 'bocode.Engineering.Rover'],
 ...
 }