Skip to main content

A Machine Learning and Data-Driven Systems Framework and Toolkit

Project description

Frater

Frater - A Machine Learning and Data-Driven Systems Framework and Toolkit

Frater is a framework and toolkit whose goal is to simplify and streamline building large scale machine learning and data driven systems in research and production, as well as providing insight into a system at each step of the pipeline. Currently, transitioning a machine learning project from a research model to a production system requires a lot of time and effort. Even more, building systems that use more than one model can be an even bigger headache. Along with this transition process, there is a need to build out ways to retrieve and understand the data passing through the system.

Frater accomplishes its goal by providing a set of tools that abstract away the engineering side of machine learning for researchers, while still letting software engineers build powerful systems with the work done by their research counterparts. The plan is to provide a hub for any task that would come up in the process of building machine learning systems:

  • Running experiments
  • Designing systems
  • Developing new models and system components
  • Analyzing and visualizing results
  • Sharing and using pre-built components
  • Resource management and configuration (GPU, CPU, memory)

Frater will also provide an API for developers and researchers to build components to use in a Frater system. Under the hood, Frater will run each component as a Docker container, which allows for portability and flexibility. All of this will be available through a web interface as well as a CLI. Frater will be able to be installed on to a local system, or deployed in a cloud environment.

Install

Requirements:

  • python 3.7+ To install Frater, execute the following:
pip install frater

Getting Started

Frater API

To start using the Frater API

import frater

Frater System

We’re currently looking for people interested in helping to make Frater’s vision into a reality. If you’re interested, contact John Henning at

Links:

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for frater, version 0.3.1.15
Filename, size File type Python version Upload date Hashes
Filename, size frater-0.3.1.15-py3-none-any.whl (39.6 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size frater-0.3.1.15.tar.gz (21.0 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page