How to use the continuum.disjoint.Disjoint function in continuum

To help you get started, we’ve selected a few continuum examples, based on popular ways it is used in public projects.

Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.

github TLESORT / Continual_Learning_Data_Former / tests / test_disjoint.py View on Github external
def test_disjoint_vanilla_test(dataset, n_tasks):
    # no need to download the dataset again for this test (if it already exists)
    input_folder = os.path.join(dir_data, 'Data')
    Disjoint(path=input_folder, dataset=dataset, tasks_number=n_tasks, download=False, train=False)
    check_task_sequences_files(scenario="Disjoint", folder=dir_data, n_tasks=n_tasks, dataset=dataset, train=False)
github TLESORT / Continual_Learning_Data_Former / tests / test_disjoint.py View on Github external
def test_download(tmpdir, dataset):
    continuum = Disjoint(path=tmpdir, dataset=dataset, tasks_number=1, download=False, train=True)

    if continuum is None:
        raise AssertionError("Object construction has failed")
github TLESORT / Continual_Learning_Data_Former / tests / test_disjoint.py View on Github external
def test_disjoint_vanilla_train(dataset, n_tasks):
    # no need to download the dataset again for this test (if it already exists)
    input_folder = os.path.join(dir_data, 'Data')
    Disjoint(path=input_folder, dataset=dataset, tasks_number=n_tasks, download=False, train=True)
    check_task_sequences_files(scenario="Disjoint", folder=dir_data, n_tasks=n_tasks, dataset=dataset, train=True)
github TLESORT / Continual_Learning_Data_Former / continuum / disjoint.py View on Github external
def __init__(self, path="./Data", dataset="MNIST", tasks_number=1, download=False, train=True):
        super(Disjoint, self).__init__(path=path,
                                       dataset=dataset,
                                       tasks_number=tasks_number,
                                       scenario="Disjoint",
                                       download=download,
                                       train=train,
                                       num_classes=10)