blog
blog copied to clipboard
huggingface
Hugging face endpoints on Azure
Hi, I followed the deployment procedure of Hugging face endpoint on Azure (in exactly the same way as described in this video https://www.youtube.com/watch?v=MlWyhdrgBwg&ab_channel=HuggingFace), however I am receiving an error message: "Failed to create HuggingFace.Endpoint resource in 'testhuggingfaceapp'. Error: The Resource 'Microsoft.CustomProviders/resourceProviders/public' under resource group 'mrg-huggingface-20221221111533' was not found. For more details please go to https://aka.ms/ARMResourceNotFoundFix"
When looking into activity log for more information I get the following information: "The resource provider 'public' received a non-success response 'BadRequest' from the downstream endpoint for the request 'PUT' on 'HuggingFace.Endpoint/a6d8a3b5-2489-4c69-a98c-8c224ec2e79e'. Please refer to additional info for details."
And in JSON of the activity log error it says: "statusMessage": "{"error":{"code":"DownstreamEndpointError","message":"The resource provider 'public' received a non-success response 'BadRequest' from the downstream endpoint for the request 'PUT' on 'HuggingFace.Endpoint/a6d8a3b5-2489-4c69-a98c-8c224ec2e79e'. Please refer to additional info for details.","details":[{"code":"Downstream","message":"{\"error\":\"Subscription 60792e14-b5c7-4b72-8ee0-8136b660385c does not have enough quota for Standard_DS2_v2.\\nCurrently available: 0\\n Requested: 4.\\n Please contact Azure support to increase quota.\"}"}]}}"
I increased the quota on Standard DS Family vCPUs and Standard DSv2 Family vCPUs in my subscription to 100 and tried to deploy the endpoint again on Standard_DS2_v2 instance type but I am not able to solve the problem.
How can I fix the error any successfully deploy the model endpoint?
I increased the quota on Standard DS Family vCPUs and Standard DSv2 Family vCPUs in my subscription to 100 and tried to deploy the endpoint again on Standard_DS2_v2 instance type but I am not able to solve the problem.
Have you increase your quota for regular Compute or Machine Learning?
