question

SampatVarun-0080 avatar image
0 Votes"
SampatVarun-0080 asked YutongTie-MSFT commented

Azure ML (Designer): Endpoint deploys initially, but can't consume or test it after that.

Hello,

I recently deployed an Azure Machine Learning Model (Designer). I have followed this tutorial (but for my model). The inference pipeline looks like this:

109810-inference-pipeline.png

This endpoint was then deployed with the following configurations:

  • Compute Type: Azure Container Instances

  • CPU : 2

  • Memory: 2 Gb

From my understanding, when a model-endpoint is deployed initially, there is a test run (which tries to consume the endpoint) that is automatically run by Azure ML, and this has always worked for me (I have tried AKS too, and ACI with lower compute power). Attached, please find the logs.

109955-initial-test-endpoint.txt

After the test run, I can either test it from the interface itself, or I can execute some Python code (which Azure ML provides) that can consume it for me. In both cases, I get a 500-Bad Gateway error (checked through Deployment Logs):
109962-endpoint-test.txt

This is what the Python code outputs:
109917-screenshot-2021-06-28-at-85746-pm.png

I have monitored the status of the endpoint while I try to consume it, and it is always healthy.

I hope I have provided enough details. If not, please let me know what else is needed to understand the issue here. Any help is appreciated.

azure-machine-learningazure-machine-learning-inference
· 2
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

Hello,

Thanks for reaching out to us. Could you please check if what you have done is the same as the guidance: https://docs.microsoft.com/en-us/azure/machine-learning/algorithm-module-reference/assign-data-to-clusters ?

The process should be:
110027-image.png

Please let me know how it goes.


Regards,
Yutong


0 Votes 0 ·
image.png (97.1 KiB)

Hi,

There doesn't seem to be any issue with Assign Data To Clusters. When I run the inference pipeline, I can visualize the outputs without any issues. Nonetheless, I retried the implementation using the instructions above and I am still facing the same issue as before. I have tried setting Check for append or uncheck for result only to True & False and in both cases the behaviour is the same as mentioned above.

Regards,
Varun

0 Votes 0 ·

0 Answers