Skip to content

Commit 8a75537

Browse files
committed
remove section
1 parent 4f356d9 commit 8a75537

File tree

1 file changed

+0
-5
lines changed

1 file changed

+0
-5
lines changed

articles/ai-studio/how-to/online-evaluation.md

Lines changed: 0 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -17,11 +17,6 @@ author: lgayhardt
1717

1818
[!INCLUDE [feature-preview](../includes/feature-preview.md)]
1919

20-
Continuous advancements in Generative AI have led organizations to build increasingly complex applications to solve various problems (chat-bots, RAG systems, agentic systems, etc.). These applications are being used to drive innovation, improve customer experiences, and enhance decision-making. Although the models (for example, GPT-4) powering these Generative AI applications are extremely capable, continuous monitoring has never been more important to ensure high-quality, safe, and reliable results. Continuous monitoring is effective when multiple perspectives are considered when observing an application. These perspectives include token usage and cost, operational metrics – latency, request count, etc. - and, importantly, continuous evaluation. To learn more about evaluation, see [Evaluation of generative AI applications](../concepts/evaluation-approach-gen-ai.md).
21-
22-
Azure AI and Azure Monitor provide tools for you to continuously monitor the performance of your Generative AI applications from multiple perspectives. With Azure AI Online Evaluation, you can continuously evaluate your application agnostic of where it's deployed or what orchestration framework it's using (for example, LangChain). You can use various [built-in evaluators](../concepts/evaluation-metrics-built-in.md) which maintain parity with the [Azure AI Evaluation SDK](./develop/evaluate-sdk.md) or define your own custom evaluators. By continuously running the right evaluators over your collected trace data, your team can more effectively identify and mitigate security, quality, and safety concerns as they arise, either in pre-production or post-production. Azure AI Online Evaluation provides full integration with the comprehensive suite of observability tooling available in [Azure Monitor Application Insights](/azure/azure-monitor/app/app-insights-overview), enabling you to build custom dashboards, visualize your evaluation results over time, and configure alerting for advanced application monitoring.
23-
24-
In summary, monitoring your generative AI applications has never been more important, due to the complexity and rapid evolvement of the AI industry. Azure AI Online Evaluation, integrated with Azure Monitor Application Insights, enables you to continuously evaluate your deployed applications to ensure that they're performant, safe, and produce high-quality results in production.
2520

2621
## How online evaluation works
2722

0 commit comments

Comments
 (0)