New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[ML] Skipping model updates should also skip the anomaly model #217
Comments
dimitris-athanasiou
added a commit
to dimitris-athanasiou/ml-cpp
that referenced
this issue
Oct 16, 2018
Skip model update was not being communicated to the anomaly model resulting to the score/probability of results triggering the skip-model-update rule to reduce. This commit skips updates to the anomaly model as well. Closes elastic#217
dimitris-athanasiou
added a commit
that referenced
this issue
Oct 17, 2018
Skip model update was not being communicated to the anomaly model resulting to the score/probability of results triggering the skip-model-update rule to reduce. This commit skips updates to the anomaly model as well. Closes #217
dimitris-athanasiou
added a commit
that referenced
this issue
Oct 17, 2018
Skip model update was not being communicated to the anomaly model resulting to the score/probability of results triggering the skip-model-update rule to reduce. This commit skips updates to the anomaly model as well. Closes #217
dimitris-athanasiou
added a commit
to dimitris-athanasiou/ml-cpp
that referenced
this issue
Oct 17, 2018
Skip model update was not being communicated to the anomaly model resulting to the score/probability of results triggering the skip-model-update rule to reduce. This commit skips updates to the anomaly model as well. Closes elastic#217
dimitris-athanasiou
added a commit
that referenced
this issue
Oct 19, 2018
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Let us imagine a job which has a simple
high_count
detector. The detector also has a rule toskip_model_update
when the count is over100
. Then our data has a normal event rate which is below100
. But at some point there are several buckets over100
. We shall expect the model to remain unaffected and that all those buckets are detected as anomalies.However, that is not the case. The screenshot below shows what currently happens.
This is most probably because the anomaly model is not skipped. Thus, we start learning that anomalies as normal and their score drops. We should ensure the anomaly model is also skipped when such a rule exists.
The text was updated successfully, but these errors were encountered: