Machine Learning Dataset Extension Specification

• Title: Machine Learning Dataset
• Identifier: https://raw.githubusercontent.com/earthpulse/ml-dataset/main/json-schema/schema.json
• Field Name Prefix: ml-dataset
• Scope: Catalog
• Extension Maturity Classification: Proposal
• Owner: @earthpulse

This document explains the Machine Learning Dataset Extension to the SpatioTemporal Asset Catalog (STAC) specification.

• Examples:
  • Catalog example: Shows the basic usage of the extension in a STAC Catalog
• JSON Schema
• Changelog

Fields

The fields in the table below can be used in these parts of STAC documents:

• Catalogs
• Collections
• Item Properties
• Assets (for both Collections and Items, incl. Item Asset Definitions in Collections)
• Links

Field Name	Type	Description
ml-dataset:name	string	The name of the dataset
ml-dataset:tasks	array	List of (suggested) tasks that can be solved with the dataset
ml-dataset:inputs-type	string	Type of the inputs (text, image, satellite image, video, ... or combination)
ml-dataset:annotations-type	string	Type of annotations (raster, vector, ...) (not present for unsupervised learning)
ml-dataset:quality-metrics	array	List of quality metrics that define the quality of the dataset
ml-dataset:version	string	Dataset version
ml-dataset:splits	array	List of the splits names. Suggested are Training, Validation, Test, Legacy, Benchmark
ml-dataset:split-items	array	List of the Items that conform the split
ml-dataset:split	string	Name of the split the Item is included

Additional Field Information

template:new_field

This is a much more detailed description of the field template:new_field...

XYZ Object

This is the introduction for the purpose and the content of the XYZ Object...

Field Name	Type	Description
x	number	REQUIRED. Describe the required field...
y	number	REQUIRED. Describe the required field...
z	number	REQUIRED. Describe the required field...

Relation types

The following types should be used as applicable rel types in the Link Object.

Type	Description
ml-dataset:splits	Links to train, test, validation splits if defined

Contributing

All contributions are subject to the STAC Specification Code of Conduct. For contributions, please follow the STAC specification contributing guide Instructions for running tests are copied here for convenience.

Running tests

The same checks that run as checks on PR's are part of the repository and can be run locally to verify that changes are valid. To run tests locally, you'll need npm, which is a standard part of any node.js installation.

First you'll need to install everything with npm once. Just navigate to the root of this repository and on your command line run:

npm install

Then to check markdown formatting and test the examples against the JSON schema, you can run:

npm test

This will spit out the same texts that you see online, and you can then go and fix your markdown or examples.

If the tests reveal formatting problems with the examples, you can fix them with:

npm run format-examples