metrics 模块

计算可用指标。

函数

assert_metrics_sane

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

assert_metrics_sane(scores, task)

参数

scores
task

compute_mean_cv_scores

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score.scoring.aggregate_scores。

compute_mean_cv_scores(scores: List[Dict[Any, Any]], metrics: List[str], logger: Optional[logging.Logger] = None) -> Dict[str, Union[float, Dict[str, Any]]]

参数

scores
metrics
logger
默认值: None

compute_metrics

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score.scoring。

compute_metrics(y_pred: numpy.ndarray, y_test: numpy.ndarray, metrics: Optional[List[str]] = None, task: str = 'classification', sample_weight: Optional[numpy.ndarray] = None, num_classes: Optional[int] = None, class_labels: Optional[numpy.ndarray] = None, trained_class_labels: Optional[numpy.ndarray] = None, y_transformer: Optional[sklearn.base.TransformerMixin] = None, y_max: Optional[float] = None, y_min: Optional[float] = None, y_std: Optional[float] = None, bin_info: Optional[Dict[str, float]] = None, horizons: Optional[numpy.ndarray] = None) -> Dict[str, Union[float, Dict[str, Any]]]

参数

y_pred
y_test
metrics
默认值: None
task
默认值: classification
sample_weight
默认值: None
num_classes
默认值: None
class_labels
默认值: None
trained_class_labels
默认值: None
y_transformer
默认值: None
y_max
默认值: None
y_min
默认值: None
y_std
默认值: None
bin_info
默认值: None
horizons
默认值: None

compute_metrics_classification

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score.scoring.score_classification。

compute_metrics_classification(y_pred_probs: Union[numpy.ndarray, pandas.core.frame.DataFrame], y_test: numpy.ndarray, metrics: List[str], num_classes: Optional[int] = None, sample_weight: Optional[numpy.ndarray] = None, class_labels: Optional[numpy.ndarray] = None, trained_class_labels: Optional[numpy.ndarray] = None, y_transformer: Optional[sklearn.base.TransformerMixin] = None, logger: Optional[logging.Logger] = None) -> Dict[str, Union[float, Dict[str, Any]]]

参数

y_pred_probs
y_test
metrics
num_classes
默认值: None
sample_weight
默认值: None
class_labels
默认值: None
trained_class_labels
默认值: None
y_transformer
默认值: None
logger
默认值: None

compute_metrics_forecasting

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score.scoring.score_forecasting。

compute_metrics_forecasting(y_pred: numpy.ndarray, y_test: numpy.ndarray, horizons: numpy.ndarray, metrics: List[str], y_max: Optional[float] = None, y_min: Optional[float] = None, y_std: Optional[float] = None, sample_weight: Optional[numpy.ndarray] = None, bin_info: Optional[Dict[str, float]] = None, logger: Optional[logging.Logger] = None, X_test: Optional[pandas.core.frame.DataFrame] = None, X_train: Optional[pandas.core.frame.DataFrame] = None, y_train: Optional[numpy.ndarray] = None, grain_column_names: Optional[List[str]] = None) -> Dict[str, Union[float, Dict[str, Any]]]

参数

y_pred
y_test
horizons
metrics
y_max
默认值: None
y_min
默认值: None
y_std
默认值: None
sample_weight
默认值: None
bin_info
默认值: None
logger
默认值: None
X_test
默认值: None
X_train
默认值: None
y_train
默认值: None
grain_column_names
默认值: None

compute_metrics_regression

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score.scoring.score_regression。

compute_metrics_regression(y_pred: numpy.ndarray, y_test: numpy.ndarray, metrics: List[str], y_max: Optional[float] = None, y_min: Optional[float] = None, y_std: Optional[float] = None, sample_weight: Optional[numpy.ndarray] = None, bin_info: Optional[Dict[str, float]] = None, logger: Optional[logging.Logger] = None) -> Dict[str, Union[float, Dict[str, Any]]]

参数

y_pred
y_test
metrics
y_max
默认值: None
y_min
默认值: None
y_std
默认值: None
sample_weight
默认值: None
bin_info
默认值: None
logger
默认值: None

get_all_nan

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

get_all_nan(task)

参数

task

get_default_metric_with_objective

不推荐使用此函数。 请使用 azureml.automl.core.shared.utilities。

get_default_metric_with_objective(task)

参数

task

get_default_metrics

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

get_default_metrics(task)

参数

task

get_max_values

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

get_max_values(task, for_assert_sane=False)

参数

task
for_assert_sane
默认值: False

get_metric_ranges

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

get_metric_ranges(task, for_assert_sane=False)

参数

task
for_assert_sane
默认值: False

get_min_values

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

get_min_values(task)

参数

task

get_scalar_metrics

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

get_scalar_metrics(task)

参数

task

get_worst_values

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

get_worst_values(task, for_assert_sane=False)

参数

task
for_assert_sane
默认值: False

is_better

不推荐使用此函数。 请使用 azureml.automl.runtime.shared.score 函数。

is_better(val1, val2, metric=None, task=None, objective=None)

参数

val1
val2
metric
默认值: None
task
默认值: None
objective
默认值: None

minimize_or_maximize

不推荐使用此函数。 请使用 azureml.automl.core.shared.utilities。

minimize_or_maximize(metric, task=None)

参数

metric
task
默认值: None