DocsICP FilingConsole
Documentation
Home Page Simple Log Service Technical Solutions Intelligent Anomaly Analysis Time series forecasting Anomaly types

Anomaly types

Updated at:
Copy as MD
Product
My Favorites

This topic describes the anomaly types that are available in the results of time series forecasting.

The time series forecasting results are stored in a Logstore named internal-ml-log. You can use the __tag__:__data_type__ field to analyze anomalies. For more information, see Result fields.

Condition

Description

  • The __tag__:__data_type__ field is set to job_error_message.

  • The result.entity and result.metric fields are not empty.

This indicates an anomaly in the forecast for a specific time series.

You can use the result.error_type and result.error_msg fields to view details about the exception.

  • The __tag__:__data_type__ field is set to job_error_message.

  • The result.entity and result.metric fields are empty.

An exception occurs in the forecasting operation indicated by an forecasting ID.

You can use the result.error_type and result.error_msg fields to view details about the exception.

Previous：AlgorithmNext：Drill-down analysis