Skip to content

Commit 800482b

Browse files
Merge pull request #221720 from mrbullwinkle/mrb_12_16_2022_metrics_advisor_ui
[Applied AI Services] [Metrics Advisor] Update what's new
2 parents 1dd7ce9 + a3a929c commit 800482b

File tree

1 file changed

+7
-1
lines changed

1 file changed

+7
-1
lines changed

articles/applied-ai-services/metrics-advisor/whats-new.md

Lines changed: 7 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -8,14 +8,20 @@ manager: nitinme
88
ms.service: applied-ai-services
99
ms.subservice: metrics-advisor
1010
ms.topic: overview
11-
ms.date: 05/25/2022
11+
ms.date: 12/16/2022
1212
ms.author: mbullwin
1313
---
1414

1515
# Metrics Advisor: what's new in the docs
1616

1717
Welcome! This page covers what's new in the Metrics Advisor docs. Check back every month for information on service changes, doc additions and updates this month.
1818

19+
## December 2022
20+
21+
**Metrics Advisor new portal experience** has been released in preview! [Metrics Advisor Studio](https://metricsadvisor.appliedai.azure.com/) is designed to align the look and feel with the Cognitive Services studio experiences with better usability and significant accessibility improvements. Both the classic portal and the new portal will be available for a period of time, until we officially retire the classic portal.
22+
23+
We **strongly encourage** you to try out the new experience and share your feedback through the smiley face on the top banner.
24+
1925
## May 2022
2026

2127
**Detection configuration auto-tuning** has been released. This feature enables you to customize the service to better surface and personalize anomalies. Instead of the traditional way of setting configurations for each time series or a group of time series. A guided experience is provided to capture your detection preferences, such as the level of sensitivity, and the types of anomaly patterns, which allows you to tailor the model to your own needs on the back end. Those preferences can then be applied to all the time series you're monitoring. This allows you to reduce configuration costs while achieving better detection results.

0 commit comments

Comments
 (0)