Open ayanray089 opened 4 years ago
@ayanray089 yes it's possible. I supposed that you have ssen this https://github.com/hastic/hastic-server/wiki/Custom-model But be prepared that it's not going to be easy.
Thanks!
Do you have any working example?
Thanks, Ayan
On Sun, 12 Apr 2020 at 23:28, Alexey Velikiy notifications@github.com wrote:
@ayanray089 https://github.com/ayanray089 yes it's possible. I supposed that you have ssen this https://github.com/hastic/hastic-server/wiki/Custom-model But be prepared that it's not going to be easy.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/hastic/hastic-server/issues/887#issuecomment-612771522, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIDQ34YMQOQPAUV7J7ES4MDRMKWIRANCNFSM4MGWI5EA .
@ayanray089 please write email if you really want to go through this:) ping@hastic.io
Hi,
I have a custom ML model for anomaly detection of time series data which has been already trained. Is it possible to use that custom ML model for predicting the Anomalous time windows after the user labels a time window, instead of default algorithms which hastic uses. I want the results to be displayed in grafana through hastic-grafana panel. Reference to a working example will help.
Thanks, Ayan