Skip to content

Commit 5f05539

Browse files
msakandedem108
andauthored
Update articles/machine-learning/how-to-safely-rollout-online-endpoints.md
Co-authored-by: SeokJin Han <[email protected]>
1 parent c1d1818 commit 5f05539

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

articles/machine-learning/how-to-safely-rollout-online-endpoints.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -384,7 +384,7 @@ If you invoke the endpoint with the name of the deployment that will receive tra
384384

385385
### Invoke endpoint without specifying deployment
386386

387-
If you invoke the endpoint without specifying the deployment that will receive traffic, Azure Machine Learning will route the endpoint's incoming traffic to one of the deployments in the endpoint based on traffic control settings.
387+
If you invoke the endpoint without specifying the deployment that will receive traffic, Azure Machine Learning will route the endpoint's incoming traffic to the deployment(s) in the endpoint based on traffic control settings.
388388

389389
Traffic control settings allocate specified percentages of incoming traffic to each deployment in the endpoint. For example, if your traffic rules specify that a particular deployment in your endpoint will receive incoming traffic 40% of the time, Azure Machine Learning will route 40% of the endpoint's traffic to that deployment. If your endpoint contains only one deployment, then the single deployment will receive 100% of incoming requests.
390390

0 commit comments

Comments
 (0)