src.environment.problem.MTO.CEC2017MTO.cec2017mto_dataset

Problem Difficulty Classification

Difficulty Mode

Training Set

Testing Set

easy

0, 1, 2, 3, 4, 5

6, 7, 8

difficult

6, 7, 8

0, 1, 2, 3, 4, 5

Note: When difficulty is ‘all’, both training and testing sets contain all problems (0-8).

Module Contents

Classes

CEC2017MTO_Tasks

CEC2017MTO_Dataset

Introduction

CEC2017MTO proposes 9 multi-task benchmark problems to represent a wider range of multi-task optimization problems.

Functions

mat2np

API

src.environment.problem.MTO.CEC2017MTO.cec2017mto_dataset.mat2np(file_obj)[source]
class src.environment.problem.MTO.CEC2017MTO.cec2017mto_dataset.CEC2017MTO_Tasks(tasks)[source]

Initialization

reset()[source]
__str__()[source]
update_T1()[source]
class src.environment.problem.MTO.CEC2017MTO.cec2017mto_dataset.CEC2017MTO_Dataset(data, batch_size=1)[source]

Bases: torch.utils.data.Dataset

Introduction

CEC2017MTO proposes 9 multi-task benchmark problems to represent a wider range of multi-task optimization problems.

Original Paper

“Evolutionary Multitasking for Single-objective Continuous Optimization: Benchmark Problems, Performance Metric, and Bseline Results.”

Official Implementation

CEC2017MTO

License

None

Problem Suite Composition

The CEC2017MTO problem suite contains a total of 9 benchmark problems, each consisting of two basic functions. These nine benchmark problems are classified according to the degree of intersection and the inter-task similarity between the two constitutive functions: P1. Complete intersection and high similarity(CI+HS) P2. Complete intersection and medium similarity(CI+MS)
P3. Complete intersection and low similarity(CI+LS) P4. Partial intersection and high similarity(PI+HS) P5. Partial intersection and medium similarity(PI+MS)
P6. Partial intersection and low similarity(PI+LS) P7. No intersection and high similarity(NI+HS) P8. No intersection and medium similarity(NI+MS) P9. No intersection and low similarity(NI+LS)

Methods:

  • __getitem__(item): Retrieves a batch of data based on the specified index.

  • __len__(): Returns the total number of datasets in the collection.

  • __add__(other): Combines the current dataset with another CEC2017MTO_Dataset instance.

  • shuffle(): Randomly shuffles the dataset indices.

  • get_datasets(version, train_batch_size, test_batch_size, difficulty, user_train_list, user_test_list): Static method to generate training and testing datasets based on the specified difficulty level or user-defined task lists.

Raises:

  • ValueError: Raised in the following cases:

    • If difficulty, user_train_list, and user_test_list are all None.

    • If an invalid difficulty value is provided.

Initialization

Introduction

Initializes the CEC2017MTO Dataset with datas.

Args:

  • data (list): A list of task datasets, where each dataset contains multiple tasks.

  • batch_size (int, optional): The size of each batch for data retrieval. Defaults to 1.

Attributes:

  • data (list): The dataset containing tasks for the CEC2017MTO problem suite.

  • batch_size (int): The size of each batch for data retrieval. Defaults to 1.

  • maxdim (int): The maximum dimensionality across all tasks in the dataset.

  • N (int): The total number of datasets in the collection.

  • ptr (list): A list of indices for batching the dataset.

  • index (numpy.ndarray): An array of shuffled indices for dataset access.

__getitem__(item)[source]

Introduction

Retrieves a batch of tasks of the the CEC2017MTO benchmark suite based on the given index.

Args:

  • item (int, optional): Specifies which batch of tasks of the CEC2017MTO benchmark is selected.

Returns:

  • list: A list containing a batch of tasks of the CEC2017MTO benchmark.

__len__()[source]

Introduction

Returns the total number of tasks in the CEC2017MTO benchmark suite.

Returns:

  • int: The size of the CEC2017MTO benchmark suite.

__add__(other: src.environment.problem.MTO.CEC2017MTO.cec2017mto_dataset.CEC2017MTO_Dataset)[source]

Introduction

Combines two datasets into a single dataset.

Returns:

  • Object: The combined new dataset of the CEC2017MTO benchmark suite.

shuffle()[source]

Introduction

Randomly shuffles the order of tasks in the dataset.

static get_datasets(version='numpy', train_batch_size=1, test_batch_size=1, difficulty=None, user_train_list=None, user_test_list=None)[source]

Introduction

Generates training and testing datasets for the CEC2017MTO Dataset benchmark suite based on specified difficulty or user-defined function lists.

Args:

  • version (str, optional): Specifies the implementation version to use for function instances. Accepts ‘numpy’ or any other string for alternative (e.g., ‘torch’). Defaults to ‘numpy’.

  • train_batch_size (int, optional): Batch size for the training dataset. Defaults to 1.

  • test_batch_size (int, optional): Batch size for the testing dataset. Defaults to 1.

  • difficulty (str, optional): Difficulty level for dataset split. Accepts ‘easy’, ‘difficult’, ‘all’, or None. If None, user_train_list and user_test_list must be provided.

  • user_train_list (list of int, optional): List of function IDs to include in the training set. Used if difficulty is None.

  • user_test_list (list of int, optional): List of function IDs to include in the testing set. Used if difficulty is None.

Returns:

  • tuple: A tuple containing two CEC2017MTO_Dataset objects:

    • The first is the training dataset.

    • The second is the testing dataset.

Raises:

  • ValueError: If neither difficulty nor both user_train_list and user_test_list are provided.

  • ValueError: If an invalid difficulty value is specified.