pawan300/azure_ml_model

Azure ML model

Successfully created two scenario.

First :

  - Saved input data in datalake.
  - Created notebook in databricks :
         - Fetched data from datalake
         - trained a model on that.
         - Registered that model in Azure ML studio via azureml library.
         - Deployed the model with config file, score.py file. 
  - Created a pipeline in ADF and run this notebook successfully.
  - Tested model in the evironment

Second :

  - Saved input data in datalake.
  - Created a notebook in databricks:
         - Fetched data from datalake.
         - Trained a model and monitor using mlflow.
  - Created another notebook in databricks:
         - Fetched run_id of the job and registered it in AzureML environment.
         - Deployed the model with config file, score.py file.
  - Created a pipeline in ADF with two notebook activity and linked both. 
  - Tested model in the environment.