SERVIR-ACES

Spec sheet

Purpose

ACES is a Python module for generating training data and training machine learning models for remote sensing applications.Read 5 Oct 2026 · servir.github.io

Support

The project directs users to file GitHub issues to report bugs or submit feedback.Read 5 Oct 2026 · servir.github.io

Apache Beam

ACES integrates with Apache Beam for data processing.Read 5 Oct 2026 · servir.github.io

Colab note

The tutorial offers a Jupyter Notebook Colab example and notes that GPU and other resources may not be fully available in the unpaid Colab version.Read 5 Oct 2026 · servir.github.io

Configuration

ACES uses environment variables in a configuration file to set data and output directories, model type, features, labels, and training parameters.Read 5 Oct 2026 · servir.github.io

Contribution model

The contribution page describes the project as volunteer-driven and welcomes contributions.Read 5 Oct 2026 · servir.github.io

Data processing

ACES provides data processing and data loading from Google Earth Engine.Read 5 Oct 2026 · servir.github.io

Earth Engine requirement

The installation documentation says Google Earth Engine is used to process data into a usable TF format and directs users to sign up for an Earth Engine account.Read 5 Oct 2026 · servir.github.io

Installation

The documentation provides installation from PyPI with pip and installation of a development version from GitHub.Read 5 Oct 2026 · servir.github.io

Model support

ACES supports training and evaluating DNN, CNN, and U-Net models, and running inference with trained models.Read 5 Oct 2026 · servir.github.io

Optional dependencies

The installation documentation offers an extra option to install all optional dependencies.Read 5 Oct 2026 · servir.github.io

Remote sensing

ACES supports feature extraction for remote sensing applications.Read 5 Oct 2026 · servir.github.io

Where Specifiction places it

• AI Image Recognition Software#32 of 376.1