After creating a model with Azure designer and creating a Real-time inference pipeline I would like to use the trained artifacts in a local script. I'm trying to look for the model.py in the Azure Storage Explorer but cannot find it, or the python where the trained model is called using pytorch.
2 Answers
To use a real-time inference pipeline you must call the REST endpoint with an authentication key.
See: https://thenewstack.io/tutorial-create-training-and-inferencing-pipelines-with-azure-ml-designer/
The end of the post shows how to make the call.
The model artifacts can be downloaded using az ml model download - that will get all of the files.
Related