Merge pull request #297 from opensearch-project/anomaly-result-minor
Anomaly result clarification
This commit is contained in:
commit
e27900265f
|
@ -194,7 +194,7 @@ If a detector detects an anomaly late, the result has the following additional f
|
|||
Field | Description
|
||||
:--- | :---
|
||||
`past_values` | The actual input that triggered an anomaly. If `past_values` is null, the attributions or expected values are from the current input. If `past_values` is not null, the attributions or expected values are from a past input (for example, the previous two steps of the data [1,2,3]).
|
||||
`approx_anomaly_start_time` | The approximate time of the actual input that triggers an anomaly. This field helps you understand when a detector flags an anomaly. If the data is not continuous, the actual time that the detector detects the anomaly can be earlier. Both single-stream and high-cardinality detectors don't query previous anomaly results because these queries are expensive operations, especially when a detector has a lot of entities. This is why the accuracy of this field is low.
|
||||
`approx_anomaly_start_time` | The approximate time of the actual input that triggers an anomaly. This field helps you understand when a detector flags an anomaly. Both single-stream and high-cardinality detectors don't query previous anomaly results because these queries are expensive operations. The cost is especially high for high-cardinality detectors that might have a lot of entities. If the data is not continuous, the accuracy of this field is low and the actual time that the detector detects an anomaly can be earlier.
|
||||
|
||||
```json
|
||||
{
|
||||
|
|
Loading…
Reference in New Issue