cli

page_type	languages	products	description
sample	azurecli	azure-machine-learning	Top-level directory for official Azure Machine Learning CLI sample code.

Azure Machine Learning 2.0 CLI (preview) examples

Welcome to the Azure Machine Learning examples repository!

Prerequisites

1. An Azure subscription. If you don't have an Azure subscription, create a free account before you begin.
2. A terminal. Install and set up the 2.0 machine learning extension before you begin.

Set up

Clone this repository:

git clone https://github.com/Azure/azureml-examples --depth 1
cd azureml-examples/cli

Run the set up script to create an Azure resource group, machine learning workspace, and set defaults for --resource-group/g and --workspace/w:

bash setup.sh

You also need remote compute targets for most examples:

bash create-compute.sh

Hello world

Run the "hello world" job:

az ml job create -f jobs/hello-world.yml --web --stream

Examples

Scripts

path	status
batch-score.sh
create-compute.sh
deploy-declarative-safe-rollout-online-endpoints.sh
deploy-imperative-safe-rollout-online-endpoints.sh
deploy-managed-online-endpoint-access-resource-sai.sh
deploy-managed-online-endpoint-access-resource-uai.sh
deploy-managed-online-endpoint.sh
deploy-r.sh
deploy-rest.sh
deploy-tfserving.sh
deploy-torchserve.sh
deploy-triton-ensemble-managed-online-endpoint.sh
deploy-triton-managed-online-endpoint.sh
deploy-triton-multiple-models-online-endpoint.sh
hello-world.sh
how-to-deploy-amlarc-endpoint.sh
how-to-deploy-declarative-safe-rollout-amlarc-endpoints.sh
manage-resources.sh
misc.sh
train-rest.sh
train.sh

Jobs (jobs)

path	status	description
jobs/logging/sklearn/iris/job.yml		Train a scikit-learn knn model on the iris dataset. Showcases examples of mlflow logging API's used in training.
jobs/train/fastai/mnist/job.yml		Train a RESNET-18 convolutional neural network (CNN) with fast.ai on the MNIST dataset.
jobs/train/fastai/pets/job.yml		Fine tune a convolutional neural network (CNN) with fast.ai on a pets dataset.
jobs/train/lightgbm/iris-bash/job.yml		Train a LightGBM model on the Iris dataset via Python via Bash script.
jobs/train/lightgbm/iris/job-sweep.yml		Run a hyperparameter sweep job for LightGBM on Iris dataset.
jobs/train/lightgbm/iris/job.yml		Train a LightGBM model on the Iris dataset.
jobs/train/pytorch/word-language-model/job.yml		Train a multi-layer RNN (Elman, GRU, or LSTM) on a language modeling task with PyTorch.
jobs/train/r/accidents/job.yml		Train a GLM using R on the accidents dataset.
jobs/train/r/iris/job.yml		Train an R model on the Iris dataset.
jobs/train/tensorflow/iris/job.yml		Train a Tensorflow Decision Forest on the Iris dataset.
jobs/train/tensorflow/mnist-distributed-horovod/job.yml		Train a basic neural network with TensorFlow on the MNIST dataset, distributed via Horovod.
jobs/train/tensorflow/mnist-distributed/job.yml		Train a basic neural network with TensorFlow on the MNIST dataset, distributed via TensorFlow.
jobs/train/tensorflow/mnist/job.yml		Train a basic neural network with TensorFlow on the MNIST dataset.
jobs/hello-world-env-var.yml		no description
jobs/hello-world.yml		no description

Endpoints (endpoints)

path	status	description

Assets (assets)

path	status	description
assets/data/iris-url.yml		Data asset pointing to Iris CSV on public blob storage.
assets/environment/python-ml-basic-cpu.yml		Environment asset created from a base Docker image plus a Conda environment file.
assets/model/lightgbm-iris.yml		Model asset from local directory.

Contents

directory	description
assets	assets
endpoints	endpoints
jobs	jobs

Contributing

We welcome contributions and suggestions! Please see the contributing guidelines for details.

Code of Conduct

This project has adopted the Microsoft Open Source Code of Conduct. Please see the code of conduct for details.

Reference

• Documentation
• Private previews