Batch Scoring on Azure for Deep Learning Models

This tutorial demonstrates how to deploy a deep learning model to Azure for batch scoring.

File System

.
├── bait/
│   ├── cluster_setup.py
│   ├── config.py
│   ├── config_template.py
│   ├── fileshare_setup.py
│   ├── job_setup.py
├── scoring_script/
│   ├── pytorch_classification/
│   │   └── score0.py
│   └── tf_mnist/
│       └── score0.py
├── training_script/
|   ├── pytorch_classification/
|   │   ├── train0.ipynb
|   │   └── train0.py
|   └── tf_mnist/
|       ├── train0.ipynb
|       └── train0.py

--- Files below this point will be generated in the tutorial

├── data/
│   └── pytorch_classification/
├── model/
│   ├── pytorch_classification/
│   └── tf_mnist/
└── func/
    ├── blobtrig/
    │   ├── function.json
    │   ├── host.json
    │   ├── __init__.py
    │   ├── readme.md
    │   └── sample.dat
    ├── host.json
    ├── local.settings.json
    └── requirements.txt

There are a few main folders to take note of in this repository:

/bait

This folder contains all the Batch AI scripts, including:

• cluster_setup.py - executed locally
• fileshare_setup.py - executed locally
• job_setup.py - executed by Functions V2

It also contains a config_template.py file, which needs to be renamed as config.py and filled out.

/func

This folder contains everything needed to run your functions v2. (TODO - maybe this should be made by the user?)

/models

This folder is where we store the model files that we will use for scoring.

/scoring_script

This folder contains the scoring script that will use a model in the /models directory. This scoring script will be executed on nodes in the Batch AI cluster.

/training_script

This folder contains the training scripts used to generate the models in the /models directory. This training script will be executed locally on a GPU enabled VM.

Contributing

This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.microsoft.com.

When you submit a pull request, a CLA-bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., label, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.