'Error in init() function during azure ML deployment due to model path definition [closed]
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.
Code:
Solution 1:[1]
- Have you registered the model in your workspace by using the
register()function on the model object. - Otherwise, there will be no model path, which may result in failure.
- Run service.get_logs() to get log information from the unhealthy service to see what is causing it to fail.
- Please refer to debug-locally for more information.
- You can debug your scoring file locally for more information. View the diagnostic events to check status of container, it may help you to debug the issue.
- View the diagnostic events to examine the state of the container; this may assist you in troubleshooting the problem.
Steps to register and deploy the model
- Register the Model
- Define Azure ML Deployment Configuration
- Create Environment Configuration file
- Create Azure ML Scoring file
- Deploy the model to Azure Container Instance
- Expose Web service
- Refer Register the model and SO thread for more information
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | HarshithaVeeramalla-MT |