Share via

AnomalyDetectorClientModelInfo interface

  • Reference
Package:
@azure/ai-anomaly-detector

Train result of a model including status, errors and diagnose info for model and variables.

Properties

alignPolicy
diagnosticsInfo

NOTE: This property will not be serialized. It can only be populated by the server.
displayName

An optional field. The name of the model whose maximum length is 24.
endTime

A required field, indicating the end time of training data. Should be date-time.
errors

Error messages when failed to create a model. NOTE: This property will not be serialized. It can only be populated by the server.
slidingWindow

An optional field, indicating how many previous points will be used to compute the anomaly score of the subsequent point.
source

Source link to the input variables. Each variable should be a csv file with two columns, timestamp and value. By default, the file name of the variable will be used as its variable name.
startTime

A required field, indicating the start time of training data. Should be date-time.
status

Model training status. NOTE: This property will not be serialized. It can only be populated by the server.

Property Details

alignPolicy 

alignPolicy?: AlignPolicy

Property Value

AlignPolicy

diagnosticsInfo

NOTE: This property will not be serialized. It can only be populated by the server.

diagnosticsInfo?: DiagnosticsInfo

Property Value

DiagnosticsInfo

displayName

An optional field. The name of the model whose maximum length is 24.

displayName?: string

Property Value

string

endTime

A required field, indicating the end time of training data. Should be date-time.

endTime: Date

Property Value

Date

errors

Error messages when failed to create a model. NOTE: This property will not be serialized. It can only be populated by the server.

errors?: AnomalyDetectorClientErrorResponse[]

Property Value

AnomalyDetectorClientErrorResponse[]

slidingWindow

An optional field, indicating how many previous points will be used to compute the anomaly score of the subsequent point.

slidingWindow?: number

Property Value

number

source

Source link to the input variables. Each variable should be a csv file with two columns, timestamp and value. By default, the file name of the variable will be used as its variable name.

source: string

Property Value

string

startTime

A required field, indicating the start time of training data. Should be date-time.

startTime: Date

Property Value

Date

status

Model training status. NOTE: This property will not be serialized. It can only be populated by the server.

status?: AnomalyDetectorClientModelStatus

Property Value

AnomalyDetectorClientModelStatus