CloudML Experiments

Manages Champion Model by Recording and Comparing Experiment Information of AI model learning

CloudML Experiments executes the AI model developed with CloudML Notebook and CloudML Studio to make sure that model learning is carried out seamlessly and check results. A dashboard provides an at-a-glance view of the execution history and status of each model.

Overview

  • Easy View of Experiment Status

    Data analyst selects a champion model after recording the experiment information generated during the model learning process. A dashboard provides an at-a-glance view of the execution history and status of each model.

  • Easy Model Comparison

    Managing the entire lifecycle of an AI model, from managing versions and results of ML experiments for each model to comparing and verifying results, is easy.

  • Central Feature Repository

    A fully-managed repository records and shares the traits of the ML model for training and inferences. The recorded features can later be reused for training and inference, cutting the development process that could otherwise take months to a fraction of the time.

01

04

Service Architecture

    SCP CloudML
  • CloudML Notebook: Model development and learning, Parameter tuning
  • CloudML Studio: Workflow development, Reporting and analytics App development
  • CloudML Pipeline: Allocate resource by component, Manage model lifecycle, Execute pipeline
  • CloudML Experiments: Save and manage experiment data, Model Registry, Model Verification
  • Kubernetes Engine
  • Data Scientist → ML model development/tuning, learning, inference analysis → SCP CloudML(CloudML Notebook) ← Utilize Kubernetes Service → Kubernetes Service (DNS, VPC, Virtual Server, NAS, Security Group, Cloud Monitoring)
  • Data Scientist → ML model development/tuning, learning, inference analysis → SCP CloudML(CloudML Studio) ← Integrate Container Image Registry → Container Image Registry
  • Data Scientist → ML model development/tuning, learning, inference analysis → SCP CloudML(CloudML Pipeline) ← Integrate user authentication → IDP(Identity Provider)*
  • MLOps Engineer → ML model development/tuning, learning, inference analysis → SCP CloudML(CloudML Experiments) ← Integrate DevOps Tool → DevOps Tool* (Nexus, GitHub)
  • MLOps Engineer → ML model development/tuning, learning, inference analysis → SCP CloudML(Kubernetes Engine) ← Save/utilize data set and model → Object Storage*
* To be provided

Key Features

  • Provide execution dashboard

    - View experiment execution list and current status : Running, Finished, Failed
    - View the meta information of project : project ID, user ID, image
    - Entire experiment list and detailed model history

  • Manage and compare MLflow based experiments

    - Manage entire experiments and view detailed information by model
    - Compare and analyze past experiment history of models for experiment management
    - Model Verification function for model verification
    - Provide Model Registry for managing and distributing model image

Pricing

    • Billing
    • Container pod usage time of Kubernetes Engine occupied by CloudML Experiments (vCore/hr)
Let’s talk

Whether you’re looking for a specific business solution or just need some questions answered, we’re here to help

Get in Touch
Share