Error in init() function during azurel ML deployment due to model path definition

Question

Hello Team,

We are trying to deploy a model to Azure ML workspace containing a saved model & One Hot Encoded joblib file.
We are facing issue in init() function.

Please find the below error message:

"message": "Aci Deployment failed with exception: Your container application crashed. This may be caused by errors in your scoring file's init() function.

PFB screenshot of scoring file:

Answer 1

@Sahana Gurumurthy (Accenture International Limite) Thanks, I think the problem is that you rely on AML background process to automatically upload content under ./outputs to AML workspace.
But when the upload is not complete and we immediately call run.register_model which takes the content from AML workspace then the error will happen.
To avoid that situation, you can do it like this:

• Persist model (joblib.dump) to a custom folder other than outputs
• Manually run upload_file to upload the model AML workspace. Name the destination same name with your model file.
• Then run run.register_model.