Skip to content
This repository has been archived by the owner on Mar 21, 2024. It is now read-only.

Latest commit

 

History

History
28 lines (18 loc) · 2.15 KB

move_model.md

File metadata and controls

28 lines (18 loc) · 2.15 KB
Raw

Move a model to other workspace

The InnerEye models on AzureML are composed of two parts in order to reproduce the same settings at inference and training time:

  • Model: The model is registered in the AzureML registry and contains the code used at training time and pytorch checkpoint
  • Environment: The Azure ML environment used to train the model. This contains the docker image with all the dependencies that were used for training

If you want to export a model from one Workspace to another you can use the following command to download and upload a model from an AzureML workspace. This script does not use settings.yml, it uses interactive authentication, and the workspace specified in the parameters.

  • Download to path:

python InnerEye/Scripts/move_model.py -a download --path ./ --workspace_name "<name>" --resource_group "<name>" --subscription_id "<sub_id>" --model_id "name:version"

The model will be downloaded to a subfolder called name_version (based on the --model_id parameter) of the path in --path parameter. For example, with --model_id "my_model:100 this will be my_model_100. The subfolder will contain two folders: one for the MODEL and one for the ENVIRONMENT files.

  • Upload from path:

python InnerEye/Scripts/move_model.py -a upload --path ./ --workspace_name "<name>" --resource_group "<name>" --subscription_id "<sub_id>" --model_id "name:version"

The model is expected to be in the format above, i.e. contained in a subfolder called name_version of the path in --path parameter. The model will be automatically given the next available id once uploaded to the workspace. For example, if my_model:100 already exists then it will be called my_model:101.

Once in place you may want to run inference on the model with partial test data, i.e. test data from patients for whom some of the labels are missing. Normally inference on partial test data would raise an exception. To allow inference to continue over partial test data add the flag --allow_incomplete_labels to your inference call, for example python InnerEye/ML/runner.py --allow_incomplete_labels --no-train --azureml --model=<model_name> --run_recovery_id=<recovery_id>