OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Edgeml

Edgeml is a machine-learning library for resource-constrained edge devices, with algorithms, training examples and inference tools intended to reduce model storage, prediction latency and energy demands. Its repository covers TensorFlow and PyTorch architectures, selected C++ training and inference code, C inference implementations, and a floating-point-to-fixed-point quantization tool.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Edgeml — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Edgeml when the target workload has tight storage, latency, energy or RAM constraints and local, offline predictions are a goal.

Who it is for

  • Developers exploring offline machine-learning inference on IoT devices or sensors.
  • Practitioners comparing compact classifiers and recurrent models for embedded workloads.
  • Teams investigating fixed-point inference or memory-constrained model deployment.

Documented uses

  • Train compact supervised-learning models for resource-limited devices.
  • Explore classifiers for local prediction, recurrent models for time-series or streaming data, and methods for anomaly detection.
  • Review inference code and quantization tooling relevant to edge-device implementations.
Before you choose

Requirements & limits

  • The repository offers separate TensorFlow and PyTorch architecture packages and examples; choose the relevant framework and directory.
  • C++ training and inference code is listed for Bonsai and ProtoNN; C inference code covers various algorithms.
  • SeeDot is a separate quantization tool for adapting models to fixed-point arithmetic.
  • The README directs users to installation and run instructions within the relevant directories.

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

Choose an EdgeML implementation area

Locate the documented code and training examples for an EdgeML algorithm.

Before you start

  • Access to the EdgeML repository.

Steps

  1. Choose the TensorFlow implementation area in tf and its examples in examples/tf, or the PyTorch area in pytorch and its examples in examples/pytorch.

  2. For C++ training or inference code, inspect cpp, which contains Bonsai and ProtoNN code.

  3. For inference code in C, inspect c_reference; for fixed-point quantization tooling, inspect tools/SeeDot.

  4. Follow the install and run instructions in the README pages within the selected directories.

Expected outcome

You can identify the repository area and its directory-specific instructions for the selected implementation path.

Guide limitations

  • The repository overview does not give the installation or execution steps for individual directories.
  • The listed C++ training and inference code covers Bonsai and ProtoNN.
Documentation 1 ↗
Evidence and transparency

Sources & review

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