Azure Machine Learning is a cloud service that you use to train, deploy, automate,
and manage machine learning models. This index should assist in navigating the Azure
Machine Learning notebook samples and encourage efficient retrieval of topics and content.
| Title | Task | Dataset | Training Compute | Deployment Target | ML Framework | Tags |
|---|---|---|---|---|---|---|
| Using Azure ML environments | Creating and registering environments | None | Local | None | None | None |
| Estimators in AML with hyperparameter tuning | Use the Estimator pattern in Azure Machine Learning SDK | None | AML Compute | None | None | None |
| Title | Task | Dataset | Training Compute | Deployment Target | ML Framework | Tags |
|---|---|---|---|---|---|---|
| Forecasting BikeShare Demand | Forecasting | BikeShare | Remote | None | Azure ML AutoML | Forecasting |
| Forecasting orange juice sales with deployment | Forecasting | Orange Juice Sales | Remote | Azure Container Instance | Azure ML AutoML | None |
| Register a model and deploy locally | Deployment | None | Local | Local | None | None |
| ⭐Data drift on aks | Filtering | NOAA | Remote | AKS | Azure ML | Dataset, Timeseries, Drift |
| Train and deploy a model using Python SDK | Training and deploying a model from a notebook | Diabetes | Local | Azure Container Instance | None | None |
| ⭐Data drift quickdemo | Filtering | NOAA | Remote | None | Azure ML | Dataset, Timeseries, Drift |
| ⭐Introduction to labeled datasets | Train | Remote | None | Azure ML | Dataset, label, Estimator | |
| ⭐Datasets with ML Pipeline | Train | Fashion MNIST | Remote | None | Azure ML | Dataset, Pipeline, Estimator, ScriptRun |
| ⭐Filtering data using Tabular Timeseiries Dataset related API | Filtering | NOAA | Local | None | Azure ML | Dataset, Tabular Timeseries |
| ⭐Train with Datasets (Tabular and File) | Train | Iris, Diabetes | Remote | None | Azure ML | Dataset, Estimator, ScriptRun |
| Forecasting away from training data | Forecasting | None | Remote | None | Azure ML AutoML | Forecasting, Confidence Intervals |
| Automated ML run with basic edition features. | Classification | Bankmarketing | AML | ACI | None | featurization, explainability, remote_run, AutomatedML |
| Classification of credit card fraudulent transactions using Automated ML | Classification | Creditcard | AML Compute | None | None | remote_run, AutomatedML |
| Automated ML run with featurization and model explainability. | Regression | MachineData | AML | ACI | None | featurization, explainability, remote_run, AutomatedML |
| ⭐Azure Machine Learning Pipeline with DataTranferStep | Demonstrates the use of DataTranferStep | Custom | ADF | None | Azure ML | None |
| Getting Started with Azure Machine Learning Pipelines | Getting Started notebook for ANML Pipelines | Custom | AML Compute | None | Azure ML | None |
| Azure Machine Learning Pipeline with AzureBatchStep | Demonstrates the use of AzureBatchStep | Custom | Azure Batch | None | Azure ML | None |
| Azure Machine Learning Pipeline with EstimatorStep | Demonstrates the use of EstimatorStep | Custom | AML Compute | None | Azure ML | None |
| ⭐How to use ModuleStep with AML Pipelines | Demonstrates the use of ModuleStep | Custom | AML Compute | None | Azure ML | None |
| ⭐How to use Pipeline Drafts to create a Published Pipeline | Demonstrates the use of Pipeline Drafts | Custom | AML Compute | None | Azure ML | None |
| ⭐Azure Machine Learning Pipeline with HyperDriveStep | Demonstrates the use of HyperDriveStep | Custom | AML Compute | None | Azure ML | None |
| ⭐How to Publish a Pipeline and Invoke the REST endpoint | Demonstrates the use of Published Pipelines | Custom | AML Compute | None | Azure ML | None |
| ⭐How to Setup a Schedule for a Published Pipeline | Demonstrates the use of Schedules for Published Pipelines | Custom | AML Compute | None | Azure ML | None |
| How to setup a versioned Pipeline Endpoint | Demonstrates the use of PipelineEndpoint to run a specific version of the Published Pipeline | Custom | AML Compute | None | Azure ML | None |
| ⭐How to use DataPath as a PipelineParameter | Demonstrates the use of DataPath as a PipelineParameter | Custom | AML Compute | None | Azure ML | None |
| ⭐How to use Dataset as a PipelineParameter | Demonstrates the use of Dataset as a PipelineParameter | Custom | AML Compute | None | Azure ML | None |
| How to use AdlaStep with AML Pipelines | Demonstrates the use of AdlaStep | Custom | Azure Data Lake Analytics | None | Azure ML | None |
| ⭐How to use DatabricksStep with AML Pipelines | Demonstrates the use of DatabricksStep | Custom | Azure Databricks | None | Azure ML, Azure Databricks | None |
| ⭐How to use AutoMLStep with AML Pipelines | Demonstrates the use of AutoMLStep | Custom | AML Compute | None | Automated Machine Learning | None |
| ⭐Azure Machine Learning Pipelines with Data Dependency | Demonstrates how to construct a Pipeline with data dependency between steps | Custom | AML Compute | None | Azure ML | None |
| How to use run a notebook as a step in AML Pipelines | Demonstrates the use of NotebookRunnerStep | Custom | AML Compute | None | Azure ML | None |
| Title | Task | Dataset | Training Compute | Deployment Target | ML Framework | Tags |
|---|---|---|---|---|---|---|
| Train a model with hyperparameter tuning | Train a Convolutional Neural Network (CNN) | MNIST | AML Compute | Azure Container Instance | Chainer | None |
| Distributed Training with Chainer | Use the Chainer estimator to perform distributed training | MNIST | AML Compute | None | Chainer | None |
| Training with hyperparameter tuning using PyTorch | Train an image classification model using transfer learning with the PyTorch estimator | ImageNet | AML Compute | Azure Container Instance | PyTorch | None |
| Distributed PyTorch | Train a model using the distributed training via Horovod | MNIST | AML Compute | None | PyTorch | None |
| Distributed training with PyTorch | Train a model using distributed training via Nccl/Gloo | MNIST | AML Compute | None | PyTorch | None |
| Training and hyperparameter tuning with Scikit-learn | Train a support vector machine (SVM) to perform classification | Iris | AML Compute | None | Scikit-learn | None |
| Training and hyperparameter tuning using the TensorFlow estimator | Train a deep neural network | MNIST | AML Compute | Azure Container Instance | TensorFlow | None |
| Distributed training using TensorFlow with Horovod | Use the TensorFlow estimator to train a word2vec model | None | AML Compute | None | TensorFlow | None |
| Distributed TensorFlow with parameter server | Use the TensorFlow estimator to train a model using distributed training | MNIST | AML Compute | None | TensorFlow | None |
| Hyperparameter tuning and warm start using the TensorFlow estimator | Train a deep neural network | MNIST | AML Compute | Azure Container Instance | TensorFlow | None |
| Resuming a model | Resume a model in TensorFlow from a previously submitted run | MNIST | AML Compute | None | TensorFlow | None |
| Training in Spark | Submiting a run on a spark cluster | None | HDI cluster | None | PySpark | None |
| Train on Azure Machine Learning Compute | Submit a run on Azure Machine Learning Compute. | Diabetes | AML Compute | None | None | None |
| Train on local compute | Train a model locally | Diabetes | Local | None | None | None |
| Train in a remote Linux virtual machine | Configure and execute a run | Diabetes | Data Science Virtual Machine | None | None | None |
| Using Tensorboard | Export the run history as Tensorboard logs | None | None | None | TensorFlow | None |
| Train a DNN using hyperparameter tuning and deploying with Keras | Create a multi-class classifier | MNIST | AML Compute | Azure Container Instance | TensorFlow | None |
| Managing your training runs | Monitor and complete runs | None | Local | None | None | None |
| Tensorboard integration with run history | Run a TensorFlow job and view its Tensorboard output live | None | Local, DSVM, AML Compute | None | TensorFlow | None |
| Use MLflow with AML for a local training run | Use MLflow tracking APIs together with Azure Machine Learning for storing your metrics and artifacts | Diabetes | Local | None | None | None |
| Use MLflow with AML for a remote training run | Use MLflow tracking APIs together with AML for storing your metrics and artifacts | Diabetes | AML Compute | None | None | None |
| Title | Task | Dataset | Training Compute | Deployment Target | ML Framework | Tags |
|---|---|---|---|---|---|---|
| Deploy MNIST digit recognition with ONNX Runtime | Image Classification | MNIST | Local | Azure Container Instance | ONNX | ONNX Model Zoo |
| Deploy Facial Expression Recognition (FER+) with ONNX Runtime | Facial Expression Recognition | Emotion FER | Local | Azure Container Instance | ONNX | ONNX Model Zoo |
| ⭐Register model and deploy as webservice | Deploy a model with Azure Machine Learning | Diabetes | None | Azure Container Instance | Scikit-learn | None |
| ⭐Deploy models to AKS using controlled roll out | Deploy a model with Azure Machine Learning | Diabetes | None | Azure Kubernetes Service | Scikit-learn | None |
| Train MNIST in PyTorch, convert, and deploy with ONNX Runtime | Image Classification | MNIST | AML Compute | Azure Container Instance | ONNX | ONNX Converter |
| Deploy ResNet50 with ONNX Runtime | Image Classification | ImageNet | Local | Azure Container Instance | ONNX | ONNX Model Zoo |
| ⭐Convert and deploy TinyYolo with ONNX Runtime | Object Detection | PASCAL VOC | local | Azure Container Instance | ONNX | ONNX Converter |
| Register Spark model and deploy as webservice | Iris | None | Azure Container Instance | PySpark |