SKIL Documentation

Skymind Intelligence Layer

The community edition of the Skymind Intelligence Layer (SKIL) is free. It takes data science projects from prototype to production quickly and easily. SKIL bridges the gap between the Python ecosystem and the JVM with a cross-team platform for Data Scientists, Data Engineers, and DevOps/IT. It is an automation tool for machine-learning workflows that enables easy training on Spark-GPU clusters, experiment tracking, one-click deployment of trained models, model performance monitoring and more.

Get Started

Release Notes

New Features and Changes in SKIL v1.0.1

  • Python support in workspace notebooks
    • Includes Python support for model server interoperability via SKILContext
    • TensorFlow support in Python notebooks
    • Keras support in Python notebooks
  • TensorFlow and Keras model import support in SKIL model server
  • Client authentication support on remote model inference via endpoints
  • Zeppelin notebook import system for quickly building new experiments
  • Docker image of SKIL system
  • Batch-job system for workspaces
    • REST API for batch job system integration with external orchestration system

Incompatible Changes and Limitations

  • inference model hosting still limited to single server

Known Issues in SKIL v1.0.1

  • Multi-server inference serving not yet stable in v1.0.1

Issues Fixed in SKIL v1.0.1 from v1.0.0

  • Many. Too many to list here