environments documentation

Environments

Categories

• AI Studio

• AOAI

• AssetManagement

• AutoML

• AutoML Image training

• Common

• Data Import

• Demand Forecasting

• Designer

• Inference

• Labeling

• MaaS

• Models

• PyTorch

• Responsible AI

• Retrieval Augmented Generation

• Scikit-learn

• System

• Tensorflow

• Training

All environments

---

• acft-hf-nlp-data-import

  Environment used by Hugging Face NLP Finetune components

• acft-hf-nlp-gpu

  Environment used by Hugging Face NLP Finetune components

• acft-mmdetection-image-gpu

  Environment used by MMDetection Image Finetune components

• acft-mmtracking-video-gpu

  Environment used by MMTracking Video Finetune components

• acft-multimodal-gpu

  Environment used by Multimodal classification Finetune components

• acft-transformers-image-gpu

  Environment used by HuggingFace Transformers Image Finetune components

• acpt-automl-image-framework-selector-gpu

  Environment used by framework selector component for automl image workloads

• acpt-pytorch-1.13-cuda11.7

  Recommended environment for Deep Learning with PyTorch on Azure containing the Azure ML SDK with the latest compatible versions of Ubuntu, Python, PyTorch, CUDA\RocM, combined with optimizers like ORT Training,+DeepSpeed+MSCCL+ORT MoE and more. The image introduces preview of new fastcheckpointin...

• acpt-pytorch-2.2-cuda12.1

  Recommended environment for Deep Learning in public preview with PyTorch on Azure containing the Azure ML SDK with the latest compatible versions of Ubuntu, Python, PyTorch, CUDA\RocM, combined with optimizers like ORT Training,+DeepSpeed+MSCCL+ORT MoE and more. The image introduces newly release...

• acpt-pytorch-2.2-cuda12.1-profiler

  Recommended environment for Deep Learning in public preview with PyTorch on Azure containing the Azure ML SDK with the latest compatible versions of Ubuntu, Python, PyTorch, CUDA\RocM, combined with optimizers like ORT Training,+DeepSpeed+MSCCL+ORT MoE and more. The image introduces newly release...

• ai-ml-automl

  An environment used by Azure ML AutoML for training models.

• ai-ml-automl-dnn

  An environment used by Azure ML AutoML for training models.

• ai-ml-automl-dnn-forecasting-gpu

  An environment used by Azure ML AutoML for training models.

• ai-ml-automl-dnn-gpu

  An environment used by Azure ML AutoML for training models.

• ai-ml-automl-dnn-text-gpu

  An environment used by Azure ML AutoML for training models.

• ai-ml-automl-dnn-text-gpu-ptca

  An environment used by Azure ML AutoML for training models.

• ai-ml-automl-dnn-vision-gpu

  An environment used by Azure ML AutoML for training models.

• ai-ml-automl-gpu

  An environment used by Azure ML AutoML for training models.

• ai-studio-dev

  (Public Preview) Environment for Generative AI on Azure containing the Prompt flow SDK with the latest compatible versions of Debian Linux and Python. This environment is part of a preview feature, and subject to the supplemental terms of use for [Microsoft Azure Previews](https://azure.microsoft...

• aoai-data-upload-finetune

  Environment used by proxy AOAI components

• automl-dnn-vision-gpu

  GPU based environment for finetuning AutoML legacy models for image tasks.

• automl-gpu

  An environment for automl inferencing (part of demand forecasting).

• data-labeling

  CPU based environment for AML Data Labeling.

• data-labeling-sam

  GPU based environment for AML Data Labeling SAM Embedding Generation.

• docker-tools

  System environment with docker tools including Oras, Trivy.

• foundation-model-inference

  Environment used for deploying model to use DS-MII or vLLM for inference

• general-langchain-app-deployment

  AzureML general environment to deploy and serve a Langchain app.

• lightgbm-3.3

  An environment for tasks such as regression, clustering, and classification with LightGBM. Contains the Azure ML SDK and additional python packages.

• llm-dbcopilot-embeddings

  An environment for Large Language Model Retrieval Augmented Generation standard grounding database components.

• llm-dbcopilot-mir

  An environment for Large Language Model MIR endpoint components.

• llm-rag

  An environment for standard Large Language Model Retrieval Augmented Generation components.

• llm-rag-embeddings

  An environment for standard Large Language Model Retrieval Augmented Generation embedding components.

• minimal-app-quickstart

  AzureML minimal app quickstart environment.

• minimal-py311-inference

  AzureML minimal/Ubuntu 22.04/Python 3.11 cpu environment.

• minimal-py39-inference

  AzureML minimal/Ubuntu 20.04/Python 3.9 cpu environment.

• mldesigner

  CPU based environment for pipelines (ML Designer).

• mldesigner-minimal

  CPU based environment for pipelines (ML Designer) minimal version.

• model-evaluation

  Environment for evaluating mlflow models.

• model-management

  Environment used by Model Management components

• python-sdk-v2

  Environment containing Azure SDK for Python v2

• responsibleai-tabular

  AzureML Responsible AI environment.

• responsibleai-text

  AzureML Responsible AI Text environment.

• responsibleai-vision

  AzureML Responsible AI Vision environment.

• sklearn-1.0

  An environment for tasks such as regression, clustering, and classification with Scikit-learn. Contains the Azure ML SDK and additional python packages.

• sklearn-1.1

  An environment for tasks such as regression, clustering, and classification with Scikit-learn. Contains the Azure ML SDK and additional python packages.

• sklearn-1.5

  An environment for tasks such as regression, clustering, and classification with Scikit-learn. Contains the Azure ML SDK and additional python packages.

• tensorflow-2.16-cuda11

  An environment for deep learning with Tensorflow containing the Azure ML SDK and additional python packages.