Azure Machine Learning Curated Environments

This article lists the curated environments in Azure Machine Learning, and the packages and channels that are pre-installed in them. Curated environments are provided by Azure Machine Learning and are available in your workspace by default. They are backed by cached Docker images, reducing the run preparation cost and allowing for faster deployment time. Use these environments to quickly get started with various machine learning frameworks.

Note

This list is updated as of September 2020. Use the Python SDK to get the most updated list. For more information, see the environments article.

AzureML-AutoML

Package channels:

  • anaconda
  • conda-forge
  • pytorch

Conda packages:

  • python
  • numpy
  • scikit-learn
  • pandas
  • py-xgboost
  • fbprophet
  • holidays
  • setuptools-git
  • psutil

Pip packages:

  • azureml-core
  • azureml-pipeline-core
  • azureml-telemetry
  • azureml-defaults
  • azureml-interpret
  • azureml-explain-model
  • azureml-automl-core
  • azureml-automl-runtime
  • azureml-train-automl-client
  • azureml-train-automl-runtime
  • inference-schema
  • py-cpuinfo

AzureML-AutoML-DNN

Package channels:

  • anaconda
  • conda-forge
  • pytorch

Conda packages:

  • python
  • numpy
  • scikit-learn
  • pandas
  • py-xgboost
  • fbprophet
  • holidays
  • setuptools-git
  • pytorch
  • cudatoolkit
  • psutil

Pip packages:

  • azureml-core
  • azureml-pipeline-core
  • azureml-telemetry
  • azureml-defaults
  • azureml-interpret
  • azureml-explain-model
  • azureml-automl-core
  • azureml-automl-runtime
  • azureml-train-automl-client
  • azureml-train-automl-runtime
  • inference-schema
  • pytorch-transformers
  • spacy
  • en_core_web_sm
  • py-cpuinfo

AzureML-AutoML-DNN-GPU

Package channels:

  • anaconda
  • conda-forge
  • pytorch

Conda packages:

  • python
  • numpy
  • scikit-learn
  • pandas
  • fbprophet
  • holidays
  • setuptools-git
  • pytorch
  • cudatoolkit
  • psutil

Pip packages:

  • azureml-core
  • azureml-pipeline-core
  • azureml-telemetry
  • azureml-defaults
  • azureml-interpret
  • azureml-explain-model
  • azureml-automl-core
  • azureml-automl-runtime
  • azureml-train-automl-client
  • azureml-train-automl-runtime
  • inference-schema
  • horovod
  • pytorch-transformers
  • spacy
  • en_core_web_sm
  • py-cpuinfo

AzureML-AutoML-DNN-Vision-GPU

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-dataset-runtime
  • azureml-contrib-dataset
  • azureml-telemetry
  • azureml-automl-core
  • azureml-automl-runtime
  • azureml-train-automl-client
  • azureml-defaults
  • azureml-interpret
  • azureml-explain-model
  • azureml-train-automl-runtime
  • azureml-train-automl
  • azureml-contrib-automl-dnn-vision

AzureML-AutoML-GPU

Package channels:

  • anaconda
  • conda-forge
  • pytorch

Conda packages:

  • python
  • numpy
  • scikit-learn
  • pandas
  • fbprophet
  • holidays
  • setuptools-git
  • psutil

Pip packages:

  • azureml-core
  • azureml-pipeline-core
  • azureml-telemetry
  • azureml-defaults
  • azureml-interpret
  • azureml-explain-model
  • azureml-automl-core
  • azureml-automl-runtime
  • azureml-train-automl-client
  • azureml-train-automl-runtime
  • inference-schema
  • py-cpuinfo

AzureML-Chainer-5.1.0-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • chainer
  • mpi4py

AzureML-Chainer-5.1.0-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • chainer
  • cupy-cuda90
  • mpi4py

AzureML-Dask-CPU

Package channels:

  • conda-forge
  • pytorch
  • defaults

Conda packages:

  • python

Pip packages:

  • adlfs
  • azureml-core
  • azureml-dataset-runtime
  • dask[complete]
  • dask-ml[complete]
  • distributed
  • fastparquet
  • fsspec
  • joblib
  • jupyterlab
  • lz4
  • mpi4py
  • notebook
  • pyarrow

AzureML-Dask-GPU

Package channels:

  • conda-forge

Conda packages:

  • python
  • matplotlib

Pip packages:

  • azureml-defaults
  • adlfs
  • azureml-core
  • dask[complete]
  • dask-ml[complete]
  • distributed
  • fastparquet
  • fsspec
  • joblib
  • jupyterlab
  • lz4
  • mpi4py
  • notebook
  • pyarrow

AzureML-Hyperdrive-ForecastDNN

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-pipeline-core
  • azureml-telemetry
  • azureml-defaults
  • azureml-automl-core
  • azureml-automl-runtime
  • azureml-train-automl-client
  • azureml-train-automl-runtime
  • azureml-contrib-automl-dnn-forecasting

AzureML-Minimal

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults

AzureML-PySpark-MmlSpark-0.15

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core

AzureML-PyTorch-1.0-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod

AzureML-PyTorch-1.0-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod

AzureML-PyTorch-1.1-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.1-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.2-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.2-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.3-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.3-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.4-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.4-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.5-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.5-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.6-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-PyTorch-1.6-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • torch
  • torchvision
  • mkl
  • horovod
  • tensorboard
  • future

AzureML-Scikit-learn-0.20.3

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • scikit-learn
  • scipy
  • joblib

AzureML-TensorFlow-1.10-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow
  • horovod

AzureML-TensorFlow-1.10-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow-gpu
  • horovod

AzureML-TensorFlow-1.12-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow
  • horovod

AzureML-TensorFlow-1.12-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow-gpu
  • horovod

AzureML-TensorFlow-1.13-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow
  • horovod

AzureML-TensorFlow-1.13-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow-gpu
  • horovod

AzureML-TensorFlow-2.0-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow
  • horovod

AzureML-TensorFlow-2.0-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow-gpu
  • horovod

AzureML-TensorFlow-2.1-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow
  • horovod

AzureML-TensorFlow-2.1-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow-gpu
  • horovod

AzureML-TensorFlow-2.2-CPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow
  • horovod

AzureML-TensorFlow-2.2-GPU

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • tensorflow-gpu
  • horovod

AzureML-Tutorial

Package channels:

  • anaconda
  • conda-forge

Conda packages:

  • python
  • pandas
  • numpy
  • tqdm
  • scikit-learn
  • matplotlib

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-telemetry
  • azureml-train-restclients-hyperdrive
  • azureml-train-core
  • azureml-widgets
  • azureml-pipeline-core
  • azureml-pipeline-steps
  • azureml-opendatasets
  • azureml-automl-core
  • azureml-automl-runtime
  • azureml-train-automl-client
  • azureml-train-automl-runtime
  • azureml-train-automl
  • azureml-train
  • azureml-sdk
  • azureml-interpret
  • azureml-tensorboard
  • azureml-mlflow
  • mlflow
  • sklearn-pandas

AzureML-VowpalWabbit-8.8.0

Package channels:

  • conda-forge

Conda packages:

  • python

Pip packages:

  • azureml-core
  • azureml-defaults
  • azureml-dataset-runtime[fuse,pandas]