You are looking at documentation for an older release. Not what you want? See the current release documentation.

    Getting Started  »

Loud ML Reference

---

• Loud ML Reference: master (current)develop1.11.21.31.41.5
• Getting Started
  • Basic Concepts
  • Setup Your config.yml File
  • Loud ML Python API Client
  • Historical Data
  • First Model
  • Evaluate
  • Forecast Future Data
  • Detecting Outliers
• Set up Loud ML
  • Installing Loud ML
    • Install Loud ML with Debian Package
    • Install Loud ML with RPM
    • Installing and running Loud ML with Docker
  • Configuring Loud ML
    • Configuring system settings
  • Stopping Loud ML
• Upgrade Loud ML
  • Rolling upgrades
  • Training upgrades
• CLI
  • Create Model Command
  • Delete Model Command
  • List Models Command
  • Training Model Command
  • Eval Command
  • Forecast Command
• Feature DSL
  • Timeseries Features DSL
  • A First Example: Abnormal Dips in User Traffic
  • Advanced Example: Using Filters
• API Conventions
  • Common options
• APIs
  • Bucket API
  • Jobs API
    • Getting job information
    • Cancelling one or more jobs
  • Scheduled Jobs API
    • Getting Scheduled Job Information
    • Get Scheduled Job API
    • Create Scheduled Job API
    • Delete Scheduled Job API
  • Models API
    • Create Model API
    • Delete Model API
    • List Models API
    • Train Model API
    • Activate Model API
    • Forecast API
  • Hooks API
    • Add Hook API
    • Delete Hook API
    • Putting it all together: An email notification example
• Glossary of terms
• Release Notes
  • 1.5.0 Release Notes
  • 1.4.2 Release Notes
  • 1.4.0 Release Notes
  • 1.3.0 Release Notes
  • 1.2.0 Release Notes
  • 1.0.0 Release Notes
• Frequently Asked Questions
  • FAQs
    • How often to train a given model?
    • Are there limits to the size of the dataset I can use for training?
    • What buckets can I easily use with Loud ML
    • What algorithms does Loud ML use to generate models?
    • On which infrastructure Loud ML runs on?
    • Can the time series model handle unobserved, missing values, or nan values?
    • Do I need to split data into train/test set for evaluation?
    • Do I need to train one model per time series?
    • Can I pass time-dependent features or scalar features?
    • How much data is required to train a model?
    • Can the time series in the dataset have different frequencies?
    • What is span and how should I set it?

    Getting Started  »