反馈

AutoMLBaseSettings 类

  • 参考

保留并验证 AutoML 试验的设置。

继承
builtins.object
AutoMLBaseSettings

构造函数

AutoMLBaseSettings(path: Optional[str] = None, iterations: int = 1000, data_script: Optional[str] = None, primary_metric: Optional[str] = None, task_type: Optional[str] = None, positive_label: Optional[Any] = None, test_size: Optional[float] = None, validation_size: Optional[float] = None, n_cross_validations: Optional[Union[str, int]] = None, y_min: Optional[float] = None, y_max: Optional[float] = None, num_classes: Optional[int] = None, featurization: Union[str, azureml.automl.core.featurization.featurizationconfig.FeaturizationConfig] = 'auto', max_cores_per_iteration: int = 1, max_concurrent_iterations: int = 1, iteration_timeout_minutes: Optional[int] = None, mem_in_mb: Optional[int] = None, enforce_time_on_windows: bool = True, experiment_timeout_minutes: Optional[int] = None, experiment_exit_score: Optional[float] = None, blocked_models: Optional[List[str]] = None, blacklist_models: Optional[List[str]] = None, allowed_models: Optional[List[str]] = None, whitelist_models: Optional[List[str]] = None, exclude_nan_labels: bool = True, verbosity: int = 20, debug_log: Optional[str] = 'automl.log', debug_flag: Optional[Dict[str, Any]] = None, enable_voting_ensemble: bool = True, enable_stack_ensemble: Optional[bool] = None, ensemble_iterations: Optional[int] = None, model_explainability: bool = True, enable_tf: bool = True, enable_subsampling: Optional[bool] = None, subsample_seed: Optional[int] = None, cost_mode: int = 0, is_timeseries: bool = False, enable_early_stopping: bool = True, early_stopping_n_iters: int = 10, enable_onnx_compatible_models: bool = False, enable_feature_sweeping: bool = False, enable_nimbusml: Optional[bool] = None, enable_streaming: Optional[bool] = None, force_streaming: Optional[bool] = None, label_column_name: Optional[str] = None, weight_column_name: Optional[str] = None, cv_split_column_names: Optional[List[str]] = None, enable_local_managed: bool = False, vm_type: Optional[str] = None, track_child_runs: bool = True, show_deprecate_warnings: Optional[bool] = True, forecasting_parameters: Optional[azureml.automl.core.forecasting_parameters.ForecastingParameters] = None, allowed_private_models: Optional[List[str]] = None, scenario: Optional[str] = None, environment_label: Optional[str] = None, save_mlflow: bool = False, test_include_predictions_only: bool = False, run_source: Optional[str] = None, partition_column_names: Optional[List[str]] = None, **kwargs: Any)

参数

path
默认值: None
iterations
默认值: 1000
data_script
默认值: None
primary_metric
默认值: None
task_type
默认值: None
positive_label
默认值: None
test_size
默认值: None
validation_size
默认值: None
n_cross_validations
默认值: None
y_min
默认值: None
y_max
默认值: None
num_classes
默认值: None
featurization
默认值: auto
max_cores_per_iteration
默认值: 1
max_concurrent_iterations
默认值: 1
iteration_timeout_minutes
默认值: None
mem_in_mb
默认值: None
enforce_time_on_windows
默认值: True
experiment_timeout_minutes
默认值: None
experiment_exit_score
默认值: None
blocked_models
默认值: None
blacklist_models
默认值: None
allowed_models
默认值: None
whitelist_models
默认值: None
exclude_nan_labels
默认值: True
verbosity
默认值: 20
debug_log
默认值: automl.log
debug_flag
默认值: None
enable_voting_ensemble
默认值: True
enable_stack_ensemble
默认值: None
ensemble_iterations
默认值: None
model_explainability
默认值: True
enable_tf
默认值: True
enable_subsampling
默认值: None
subsample_seed
默认值: None
cost_mode
默认值: 0
is_timeseries
默认值: False
enable_early_stopping
默认值: True
early_stopping_n_iters
默认值: 10
enable_onnx_compatible_models
默认值: False
enable_feature_sweeping
默认值: False
enable_nimbusml
默认值: None
enable_streaming
默认值: None
force_streaming
默认值: None
label_column_name
默认值: None
weight_column_name
默认值: None
cv_split_column_names
默认值: None
enable_local_managed
默认值: False
vm_type
默认值: None
track_child_runs
默认值: True
show_deprecate_warnings
默认值: True
forecasting_parameters
默认值: None
allowed_private_models
默认值: None
scenario
默认值: None
environment_label
默认值: None
save_mlflow
默认值: False
test_include_predictions_only
默认值: False
run_source
默认值: None
partition_column_names
默认值: None

方法

as_serializable_dict
from_string_or_dict

将包含设置的字符串或字典转换为 AutoMLBaseSettings 对象。

如果提供的值已经是 AutoMLBaseSettings 对象，则只需传递它。

as_serializable_dict 

as_serializable_dict() -> Dict[str, Any]

from_string_or_dict

将包含设置的字符串或字典转换为 AutoMLBaseSettings 对象。

如果提供的值已经是 AutoMLBaseSettings 对象，则只需传递它。

static from_string_or_dict(val: Union[Dict[str, Any], str, azureml.automl.core.automl_base_settings.AutoMLBaseSettings]) -> azureml.automl.core.automl_base_settings.AutoMLBaseSettings

参数

val
必需

要转换的输入数据

返回

AutoMLBaseSettings 对象

属性

debug_log

MAXIMUM_DEFAULT_ENSEMBLE_SELECTION_ITERATIONS 

MAXIMUM_DEFAULT_ENSEMBLE_SELECTION_ITERATIONS = 15

MAXIMUM_EXPERIMENT_TIMEOUT_MINUTES 

MAXIMUM_EXPERIMENT_TIMEOUT_MINUTES = 525600

MAXIMUM_ITERATION_TIMEOUT_MINUTES 

MAXIMUM_ITERATION_TIMEOUT_MINUTES = 43200

MAXIMUM_MEM_IN_MB 

MAXIMUM_MEM_IN_MB = 1073741824

MAX_CORES_PER_ITERATION 

MAX_CORES_PER_ITERATION = 16384

MAX_LAG_LENGTH 

MAX_LAG_LENGTH = 2000

MAX_N_CROSS_VALIDATIONS 

MAX_N_CROSS_VALIDATIONS = 1000

MINIMUM_REQUIRED_ITERATIONS_ENSEMBLE 

MINIMUM_REQUIRED_ITERATIONS_ENSEMBLE = 2

MIN_EXPTIMEOUT_MINUTES 

MIN_EXPTIMEOUT_MINUTES = 15