def _wait_hostname_resolution():
    """Wait for the hostname resolution of the container. This is known behavior as the cluster
    boots up and has been documented here:
     https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms-training-algo-running-container.html#your-algorithms-training-algo-running-container-dist-training
    """
    for host in _env.TrainingEnv().hosts:
        _dns_lookup(host)
Esempio n. 2
0
def training_env(resource_config=None,
                 input_data_config=None,
                 hyperparameters=None):

    resource_config = resource_config or env.read_resource_config()
    input_data_config = input_data_config or env.read_input_data_config()
    hyperparameters = hyperparameters or env.read_hyperparameters()

    return env.TrainingEnv(resource_config=resource_config,
                           input_data_config=input_data_config,
                           hyperparameters=hyperparameters)
Esempio n. 3
0
def training_env():
    """Create a TrainingEnv.

    Returns:
        TrainingEnv: an instance of TrainingEnv
    """
    from sagemaker_containers import _env

    return _env.TrainingEnv(resource_config=_env.read_resource_config(),
                            input_data_config=_env.read_input_data_config(),
                            hyperparameters=_env.read_hyperparameters())
Esempio n. 4
0
def training_env(resource_config=None, input_data_config=None, hyperparameters=None):
    """Placeholder docstring"""

    resource_config = resource_config or env.read_resource_config()
    input_data_config = input_data_config or env.read_input_data_config()
    hyperparameters = hyperparameters or env.read_hyperparameters()

    return env.TrainingEnv(
        resource_config=resource_config,
        input_data_config=input_data_config,
        hyperparameters=hyperparameters,
    )