共用方式為


ModelServingDataset 類別

表示建立模型型 DataDriftDetector 物件時,在內部使用的資料集。

以模型為基礎的 DataDriftDetector 可讓您計算模型訓練資料集與其評分資料集之間的資料漂移。 若要建立以模型為基礎的 DataDriftDetector,請使用 <xref:azureml.datadrift.DataDriftDetector.create_from_model> 方法。

建構函式。

繼承
builtins.object
ModelServingDataset

建構函式

ModelServingDataset(workspace)

參數

名稱 Description
workspace
必要

模型服務資料集的工作區。

workspace
必要
<xref:Workspace>

提供資料集之模型的工作區。

方法

export_to_csv

將模型服務資料集匯出至本機 CSV 檔案。

export_to_csv

將模型服務資料集匯出至本機 CSV 檔案。

export_to_csv(start_time, end_time)

參數

名稱 Description
start_time
必要

以 UTC 開始匯出的開始時間。

end_time
必要

結束匯出的 UTC 結束時間。

傳回

類型 Description
str

匯出之 CSV 檔案的相對路徑。