FastScore

FastScore is an analytic deployment environment, enabling organizations to iterate and deploy more models into production faster. With an agnostic deployment environment, FastScore helps organization with heterogeneous analytic creation tools and a variety of data sources to quickly and easily turn "models" into "assets" and build competency around monetizing those assets.

NEW FEATURES

Version 1.3 of FastScore elevates schemas to first-class citizens in the FastScore ecosystem, as well as adds sensors and support for models in both Python 2.7.x and 3.6.x.

  • Schemas
  • Sensors
  • Updated FastScore command line interface (CLI)
  • An additional streaming transport: "pigging" (buffer/batch flushing), together with a REST api batch processing function
  • Configurable stream buffering

WANT TO KEEP UP? SUBSCRIBE TO RECEIVE UPDATES ON FASTSCORE


Universal Deployment, Start to Finish

Multi-Language Support

THE FASTSCORE ENGINE DEPLOYS MODELS FROM MOST COMMON DATA SCIENCE LANGUAGES

  • Simulatneously deploy models from a variety of languages
  • Leverage the best data science tools for each problem or model
  • Enables rapid iteration and model tuning in each language with production support

Compare Model Metrics

Compare Model Metrics

With FastScore's sensors, it's now possible to monitor key aspects of your model's performance and pipeline.

  • Monitor how much memory, CPU or time it takes for your algorithm to actually evaluate data
  • Easily compare model performance from DIFFERENT languages
  • Instrumentation helps debug model performance in production

Fit to your Business

Supports Enterprise Systems Migration

Fastscore leverages modern, microservices based deployment technolgoy to allow simple adoption

  • FastScore is deployed as a docker container
  • IT teams can leverage full support from docker ecosystem:  
    monitoring, networking and orchestration
  • On Premise, in-cloud or hybrid - your choice

Best of all, it's easy to start. All it takes is three files (model, sample input data, sample output data...just to make sure we got it right) and we bet we can show you how to use our software to deploy your model anywhere.  Give us a test!

Request a Demo