adelvalle62/loudml

Loud ML is the first open-source AI solution for ICT and IoT automation

Loud ML - Reveal the hidden

Loud ML is an open source inference engine for metrics and events, and the fastest way to embed machine learning in your time series application. This includes APIs for storing and querying data, processing it in the background for ML or detecting outliers for alerting purposes, and more.

Help make this document better

This page, as well as the rest of our docs, are open-source and available on GitHub. We welcome your contributions.

• To report a problem in the documentation, or to submit feedback and comments, please open an issue on GitHub.

An Open-Source AI Library for Time Series Data

Loud ML is an open source time series inference engine built on top of TensorFlow. It's useful to forecast data, detect outliers, and automate your process using future knowledge.

Features

• Built-in HTTP API that facilitates the integration in other applications
• Data agnostic. The ML engine consumes data from different buckets to achieve seamless data experience. Supported data buckets include:
  • ElasticSearch
  • InfluxDB
  • MongoDB
  • OpenTSDB. Contributed by Volodymyr Sergeyev
• JSON configuration
• Simple to install and manage
• Donut unsupervised learning model arXiv 1802.03903
• Data processing in near real-time: data buckets are queried at regular intervals and feed the inference engine to return results

Installation

Local install

loudmld can be installed using pip similar to other Python packages. Do not use sudo with pip. It is usually good to work in a virtualenv or venv to avoid conflicts with other package managers and Python projects. For a quick introduction see Python Virtual Environments in Five Minutes

Run inside a virtualenv:

make install

Getting Started

Running loudmld

You can start the Loud ML model server using:

• systemctl start loudmld if you have installed Loud ML using an official Debian or RPM package, and are running a distro with systemd.
• loudmld if you have built Loud ML from source.

loudmld -c <path/to/config.yml file>

Running loudml command-line interface

One extra package is needed to run the command line interface.

If you've installed loudml-python locally, the loudml command should be available via the command line. Executing loudml will start the CLI and automatically connect to the local Loud ML model server instance (assuming you have already started the server with systemctl start loudmld or by running loudmld directly).

pip install loudml-python

The Python client library is open source

Contributors wanted! Official client libraries for Javascript, Java, Ruby, Go can be found at: https://github.com/loudml

Running unit tests

make test

Building Packages

make clean && make rpm

make clean && make repo

Documentation

• Read more about the design goals and motivations of the project.
• Follow the getting started guide to learn the basics in just a few minutes.
• Learn more about Loud ML's key concepts.

Contributing

If you're feeling adventurous and want to contribute to Loud ML, see our contributing doc for info on how to make feature requests, build from source, and run tests.

Licensing

See LICENSE

Looking for Support?

Contact contact@loudml.io to learn how we can best help you succeed.