data_science_usecase_icon_solutionspace
rule

SigOpt’s hyperparameter optimization technologies enable you to boost the performance of your AI, ML, backtest, and simulation models with optimal hyperparameter, architecture, and feature transformation tuning. The highly scalable, global AWS infrastructure is an ideal solution for the robust performance requirements of machine learning workloads.  

This solution is provided by SigOpt, an AWS Partner Network (APN) Partner with AWS Machine Learning Competency designation.

  • Learn

    SigOpt_logo_horiz
    • Improve machine learning models 100x faster with state-of-the-art Bayesian optimization
    • Reduce costs by minimizing the waste of computational and personnel resources
    • Easily integrate SigOpt with your existing machine learning tools
    • Scale the infrastructure supporting your machine learning projects up and down on-demand
    • Keep your data and models secure with black box optimization  
    • Black box optimization, which enables the SigOpt team to assess the inputs and outputs of your model without needing to look at the underlying data
    • Client Libraries for Python, R, and Java
    • Open Source integrations for Spark and scikit-learn
    • Support SLAs across every pricing plan
    • Single sign-on and team-based permissions

    With SigOpt on AWS, you can amplify your research by optimizing your model’s hyperparameters faster and more accurately than grid search, random search and manual tuning. The solution is also faster, more stable, and easier to use than open source solutions, and leverages AWS for fast deployment, scalability, and cost-effectiveness.

    Learn more about SigOpt.

    Discover value from this solution by gaining access to AWS credits to support a Proof-of-Concept (POC) and consulting offers from AWS Competency Partners that include AWS co-investments. Click below for more information.

    Get started
  • Launch

    aws-marketplace-200

    SigOpt is delivered through a software-as-a-service (SaaS) model that enables you to deploy quickly and pay only for what you use. In addition to a free trial, two pricing plans are available through AWS Marketplace: 15 experiments per month, and 100 experiments per month.

    When deployed from AWS Marketplace, setup takes less than 15 minutes and the cost of SigOpt will be rolled directly into your AWS bill. Custom pricing plans are also available by contacting SigOpt.

    Discover value from this solution by gaining access to AWS credits to support a Proof-of-Concept (POC) and consulting offers from AWS Competency Partners that include AWS co-investments. Click below for more information.

    Get started
  • Work with an APN Partner

    Discover value from this solution by gaining access to AWS credits to support a Proof-of-Concept (POC) and consulting offers from AWS Competency Partners that include AWS co-investments. Click below for more information.

    Get started
  • Purchase

    aws-marketplace-200

    SigOpt is offered in AWS Marketplace based on different subscription duration and number of experiments per month options.

    • 15 experiments / month-With 500 observations, 20 parameters, 4x parallelism
    • 100 experiments / month-With 1000 observations, 100 parameters, 10x parallelism

    Discover value from this solution by gaining access to AWS credits to support a Proof-of-Concept (POC) and consulting offers from AWS Competency Partners that include AWS co-investments. Click below for more information.

    Get started
  • Get help

    icon-aws-credits-solutionspace

    AWS Credits

    icon-aws-talk-solutionspace

    Other AWS-related Inquiries

    Discover value from this solution by gaining access to AWS credits to support a Proof-of-Concept (POC) and consulting offers from AWS Competency Partners that include AWS co-investments. Click below for more information.