OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Terratorch

TerraTorch is a Python PyTorch domain library for fine-tuning and using pretrained geospatial foundation models.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Terratorch — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use TerraTorch when a project needs a PyTorch framework for adapting pretrained geospatial models to an Earth-observation task, with trainer, model-composition and data-loading tools in one library.

Who it is for

  • Developers adapting geospatial foundation models to Earth-observation datasets and tasks.
  • Machine-learning practitioners building segmentation, classification or pixel-wise regression workflows for geospatial data.

Documented uses

  • Fine-tune a geospatial foundation model for image segmentation, classification or pixel-wise regression.
  • Combine a model backbone and decoder using model factories.
  • Use provided geospatial datasets and datamodules with project data.
  • Launch fine-tuning through the CLI and configuration files or through Jupyter notebooks.
Before you choose

Requirements & limits

The repository lists an Apache-2.0 license. Installation options documented in the README include pip, conda-forge and pipx.

  • The library supports segmentation, classification and pixel-wise regression fine-tuning.
  • Fine-tuning can be started through the CLI with configuration files or through Jupyter notebooks.
  • GDAL is a stated installation requirement, and the README recommends conda-forge when it is not already set up.
  • TerraTorch provides a training and inference framework but does not host models; check the license of any model used for the intended purpose.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install TerraTorch with pip

Prepare a Python environment with TerraTorch installed.

Before you start

  • Python package installation through pip
  • pip 21.8 or later for use of the project’s pyproject.toml
  • GDAL installed

Steps

  1. If needed, upgrade pip with the command below.

    Documented command
    python -m pip install --upgrade pip

  2. Install GDAL; the README recommends conda install -c conda-forge gdal when GDAL is not already set up.

  3. Install a stable TerraTorch release with the command below.

    Documented command
    pip install terratorch==

Expected outcome

TerraTorch is installed from a stable point release, with the documented GDAL prerequisite addressed.

Guide limitations

  • The installation instructions identify GDAL setup as potentially complex.
  • This recipe covers installation only, not a fine-tuning run.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
torchgeo/terratorch · Official homepage title · retained in its source language
How we review sourcesSuggest a correction