Esri/deep-learning-frameworks

Installation support for Deep Learning Frameworks for the ArcGIS System

What it solves

This project provides a comprehensive set of pre-configured deep learning and machine learning libraries for ArcGIS users. It eliminates the manual effort of setting up complex AI environments, ensuring that geospatial professionals can perform tasks like feature extraction, pixel classification, and feature categorization without worrying about dependency conflicts or installation errors.

How it works

It acts as a bundled installer that adds a collection of 254 packages—including PyTorch, transformers, Fast.ai, and scikit-learn—directly into the default arcgispro-py3 Python environment used by ArcGIS Pro and Server. It supports both Windows (via .msi installers) and Linux (via shell scripts), and provides specific versions matched to the ArcGIS software version to ensure compatibility. For those in disconnected environments, it also offers separate backbone packages to allow model training without internet access.

Who it’s for

Geospatial analysts and developers using ArcGIS Pro, ArcGIS Server, ArcGIS Drone2Map, or ArcGIS Reality Studio who want to implement AI and deep learning workflows for spatial data science.

Highlights

  • Extensive Library Bundle: Includes 254 packages, including industry standards like PyTorch and scikit-learn.
  • Seamless Integration: Integrates directly with the arcgis.learn module and ArcGIS Geoprocessing tools.
  • Hardware Optimization: Specifically configured for NVIDIA GPUs with CUDA support for accelerated computation.
  • Multi-Platform Support: Provides dedicated installers for Windows and Linux server environments.

Related

  • Project
    opengeos/geoai

    A Python package that integrates AI frameworks with geospatial data analysis, providing tools for satellite imagery processing, model training, and inference.

  • Project
    neuralhydrology/neuralhydrology

    A PyTorch-based Python library designed for deep learning research in hydrology, allowing users to train neural networks via configuration files.

  • Project
    frgfm/holocron

    A library of high-quality PyTorch implementations of recent deep learning tricks and model architectures for computer vision tasks.

  • Project
    kozistr/pytorch_optimizer

    A PyTorch‑compatible library that aggregates 100+ research optimizers, several LR schedulers and loss functions behind a single, consistent API, with optional integrations for low‑precision training.

  • Project
    facebookresearch/pytorch3d

    A PyTorch-based library for 3D Computer Vision research providing differentiable rendering and efficient tools for manipulating 3D meshes and point clouds.