Skip to content

1.1 chore: Verify and fix link urls #929

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
B13rg merged 4 commits into from
Aug 21, 2025
+30 −30
Merged

1.1 chore: Verify and fix link urls #929

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1323533
Fix url typo
B13rg Aug 1, 2025
e1589e9
Point directly to doi
B13rg Aug 1, 2025
865bafb
fix quoting
B13rg Aug 1, 2025
7732654
Correct namespace URLs to use http://
B13rg Aug 1, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
60 changes: 30 additions & 30 deletions docs/croissant-spec-draft.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
Version 1.1 (draft)

**This is a draft of the Croissant 1.1 specification. This document is a work in progress.
For the latest official specification, please see the [Croissant 1.0 specification](https://http://mlcommons.org/croissant/1.0).**
For the latest official specification, please see the [Croissant 1.0 specification](http://mlcommons.org/croissant/1.0).**
Comment thread
B13rg marked this conversation as resolved.

<!-- Published: -->

Expand Down Expand Up @@ -54,7 +54,7 @@ Creating or changing the metadata is straightforward. A dataset repository can i

### Responsible AI

As AI advances at rapid speed there is increased recognition among researchers, practitioners and policy makers that we need to explore, understand, manage, and assess [its economic, social, and environmental impacts](https://link.springer.com/book/10.1007/978-3-030-30371-6). One of the main instruments to operationalise responsible AI (RAI) is dataset documentation.
As AI advances at rapid speed there is increased recognition among researchers, practitioners and policy makers that we need to explore, understand, manage, and assess [its economic, social, and environmental impacts](https://doi.org/10.1007/978-3-030-30371-6). One of the main instruments to operationalise responsible AI (RAI) is dataset documentation.

This is how Croissant helps address RAI:

Expand Down Expand Up @@ -84,7 +84,7 @@ Croissant metadata is encoded in JSON-LD.
{
"@context": {
"@language": "en",
"@vocab": "https://schema.org/"
"@vocab": "http://schema.org/"
},
"@type": "sc:Dataset",
"name": "simple-pass",
Expand Down Expand Up @@ -405,7 +405,7 @@ These [schema.org](http://schema.org) properties are recommended for every Crois
<tr>
<td><a href="http://schema.org/keywords">keywords</a></td>
<td>
<a href="https://schema.org/DefinedTerm">DefinedTerm</a><br>
<a href="http://schema.org/DefinedTerm">DefinedTerm</a><br>
<a href="http://schema.org/Text">Text</a><br>
<a href="http://schema.org/URL">URL</a>
</td>
Expand Down Expand Up @@ -466,8 +466,8 @@ These [schema.org](http://schema.org) properties are recommended for every Crois
<tr>
<td><a href="http://schema.org/inLanguage">inLanguage</a></td>
<td>
<a href="https://schema.org/Language">Language</a><br>
<a href="https://schema.org/Text">Text</a>
<a href="http://schema.org/Language">Language</a><br>
<a href="http://schema.org/Text">Text</a>
</td>
<td>MANY</td>
<td>The language(s) of the content of the dataset.</td>
Expand Down Expand Up @@ -608,37 +608,37 @@ Most of the important properties needed to describe a `FileObject` are defined i
<th>Description</th>
</thead>
<tr>
<td><a href="https://schema.org/name">sc:name</a></td>
<td><a href="http://schema.org/name">sc:name</a></td>
<td><a href="http://schema.org/Text">Text</a></td>
<td>ONE</td>
<td>The name of the file. As much as possible, the name should reflect the name of the file as downloaded, including the file extension. e.g. "images.zip".</td>
</tr>
<tr>
<td><a href="https://schema.org/contentUrl">sc:contentUrl</a></td>
<td><a href="http://schema.org/contentUrl">sc:contentUrl</a></td>
<td><a href="http://schema.org/URL">URL</a></td>
<td>ONE</td>
<td>Actual bytes of the media object, for example the image file or video file.</td>
</tr>
<tr>
<td><a href="https://schema.org/contentSize">sc:contentSize</a></td>
<td><a href="http://schema.org/contentSize">sc:contentSize</a></td>
<td><a href="http://schema.org/Text">Text</a></td>
<td>ONE</td>
<td>File size in (mega/kilo/…)bytes. Defaults to bytes if a unit is not specified.</td>
</tr>
<tr>
<td><a href="https://schema.org/encodingFormat">sc:encodingFormat</a></td>
<td><a href="http://schema.org/encodingFormat">sc:encodingFormat</a></td>
<td><a href="http://schema.org/Text">Text</a></td>
<td>MANY</td>
<td>The formats of the file, given as a mime type. Unregistered or niche encoding and file formats can be indicated instead via the most appropriate URL, e.g. defining Web page or a Wikipedia/Wikidata entry.</td>
</tr>
<tr>
<td><a href="https://schema.org/sameAs">sc:sameAs</a></td>
<td><a href="http://schema.org/sameAs">sc:sameAs</a></td>
<td><a href="http://schema.org/URL">URL</a></td>
<td>MANY</td>
<td>URL (or local name) of a FileObject with the same content, but in a different format.</td>
</tr>
<tr>
<td><a href="https://schema.org/sha256">sc:sha256</a></td>
<td><a href="http://schema.org/sha256">sc:sha256</a></td>
<td><a href="http://schema.org/Text">Text</a></td>
<td>ONE</td>
<td>Checksum for the file contents.</td>
Expand Down Expand Up @@ -1015,7 +1015,7 @@ The ratings `RecordSet` above corresponds to a CSV table, declared elsewhere as

`RecordSet`s specify where to get their data via the `dataSource` property of Field. `DataSource` is the class describing the data that can be extracted from files to populate a `RecordSet`. This class should be used when the data coming from the source needs to be transformed or formatted to be included in the ML dataset; otherwise a simple `Reference` can be used instead to point to the source.

`DataSource` is a subclassOf: [sc:Intangible](https://schema.org/Intangible) and defines the following properties:
`DataSource` is a subclassOf: [sc:Intangible](http://schema.org/Intangible) and defines the following properties:

<table>
<thead>
Expand Down Expand Up @@ -1190,31 +1190,31 @@ Commonly used atomic data types:
<th>Usage</th>
</thead>
<tr>
<td><a href="https://schema.org/Boolean">sc:Boolean</a></td>
<td><a href="http://schema.org/Boolean">sc:Boolean</a></td>
<td>Describes a boolean.</td>
</tr>
<tr>
<td><a href="https://schema.org/Date">sc:Date</a></td>
<td><a href="http://schema.org/Date">sc:Date</a></td>
<td>Describes a date.</td>
</tr>
<tr>
<td><a href="https://schema.org/Time">sc:Time</a></td>
<td><a href="http://schema.org/Time">sc:Time</a></td>
<td>Describes a time.</td>
</tr>
<tr>
<td><a href="https://schema.org/DateTime">sc:DateTime</a></td>
<td><a href="http://schema.org/DateTime">sc:DateTime</a></td>
<td>Describes a combination of date and time of day.</td>
</tr>
<tr>
<td><a href="https://schema.org/Float">sc:Float</a></td>
<td><a href="http://schema.org/Float">sc:Float</a></td>
<td>Describes a float.</td>
</tr>
<tr>
<td><a href="https://schema.org/Integer">sc:Integer</a></td>
<td><a href="http://schema.org/Integer">sc:Integer</a></td>
<td>Describes an integer.</td>
</tr>
<tr>
<td><a href="https://schema.org/Text">sc:Text</a></td>
<td><a href="http://schema.org/Text">sc:Text</a></td>
<td>Describes a string.</td>
</tr>
</table>
Expand All @@ -1227,7 +1227,7 @@ Other data types commonly used in ML datasets:
<th>Usage</th>
</thead>
<tr>
<td><a href="https://schema.org/ImageObject">sc:ImageObject</a></td>
<td><a href="http://schema.org/ImageObject">sc:ImageObject</a></td>
<td>Describes a field containing the content of an image (pixels).</td>
</tr>
<tr>
Expand All @@ -1251,10 +1251,10 @@ Croissant datasets can use data types from other vocabularies, such as Wikidata.
</thead>
<tr>
<td>
<a href="https://www.wikidata.org/wiki/Q48277">wd:Q48277</a><br>
<a href="http://www.wikidata.org/wiki/Q48277">wd:Q48277</a><br>
(gender)
</td>
<td>Describes a Field or a RecordSet whose values are indicative of someone’s gender. This could be used for instance by RAI frameworks and tools to flag possible biases in the data. Values for this RecordSet can be associated with specific gender URLs (eg: <a href="https://www.wikidata.org/wiki/Q6581097">wd:Q6581097</a>, <a href="https://www.wikidata.org/wiki/Q6581072">wd:Q6581072</a>, etc.). Refer to the "Typed RecordSets > Enumerations" section for an example.</td>
<td>Describes a Field or a RecordSet whose values are indicative of someone’s gender. This could be used for instance by RAI frameworks and tools to flag possible biases in the data. Values for this RecordSet can be associated with specific gender URLs (eg: <a href="http://www.wikidata.org/wiki/Q6581097">wd:Q6581097</a>, <a href="http://www.wikidata.org/wiki/Q6581072">wd:Q6581072</a>, etc.). Refer to the "Typed RecordSets > Enumerations" section for an example.</td>
</tr>
</table>

Expand All @@ -1268,13 +1268,13 @@ In the following example, `color_sample` is a field containing an image, but wit
}
```

In the following example, the `url` field is expected to be a URL, whose semantic type is [City](https://www.wikidata.org/wiki/Q515), so one will expect values of this field to be URLs referring to cities (e.g.: "<https://www.wikidata.org/wiki/Q90>").
In the following example, the `url` field is expected to be a URL, whose semantic type is [City](http://www.wikidata.org/wiki/Q515), so one will expect values of this field to be URLs referring to cities (e.g.: "<http://www.wikidata.org/wiki/Q90>").

```json
{
"@id": "cities/url",
"@type": "cr:Field",
"dataType": ["https://schema.org/URL", "https://www.wikidata.org/wiki/Q515"]
"dataType": ["http://schema.org/URL", "http://www.wikidata.org/wiki/Q515"]
}
```

Expand Down Expand Up @@ -1522,7 +1522,7 @@ Annotations can also appear at the level of a RecordSet. A RecordSet-level annot
],
"annotation" : {
"@type": "cr:Field", "@id": "movies/ratings",
subField: [
"subField": [
{ "@type": "cr:Field", "@id": "movies/ratings/user_id", ...},
{ "@type": "cr:Field", "@id": "movies/ratings/rating", ...},
]
Expand Down Expand Up @@ -1577,9 +1577,9 @@ We now introduce a number of features that are useful in the context of ML data.

### Categorical Data

In machine learning applications, it's often useful to know that some of the data is categorical in nature, and has a finite set of values that can be used, say, for classification. Croissant represents that information by using the [sc:Enumeration](https://schema.org/Enumeration) class from [schema.org](https://schema.org), as a `dataType` on `RecordSet`s that hold categorical data.
In machine learning applications, it's often useful to know that some of the data is categorical in nature, and has a finite set of values that can be used, say, for classification. Croissant represents that information by using the [sc:Enumeration](http://schema.org/Enumeration) class from [schema.org](http://schema.org), as a `dataType` on `RecordSet`s that hold categorical data.

These RecordSets must define a `name` field conforming with the [sc:name](https://schema.org/name) definition, i.e. a human-readable text naming the item. They must also specify a key to identify each possible instance. Enumerations should have a `url` field, which can also be used to uniquely refer to each instance.
These RecordSets must define a `name` field conforming with the [sc:name](http://schema.org/name) definition, i.e. a human-readable text naming the item. They must also specify a key to identify each possible instance. Enumerations should have a `url` field, which can also be used to uniquely refer to each instance.

For example, the [COCO](https://cocodataset.org/#format-data) dataset defines categories and super-categories ([Croissant definition](https://github.com/mlcommons/croissant/blob/main/datasets/1.0/coco2014/metadata.json)), to which are associated other parts of the dataset. Using Croissant, one can describe the COCO super-categories the following way:

Expand Down Expand Up @@ -1840,8 +1840,8 @@ Segmentation mask as an image:
```json
"@context": {
"@language": "en",
"@vocab": "https://schema.org/",
"sc": "https://schema.org/",
"@vocab": "http://schema.org/",
"sc": "http://schema.org/",
"cr": "http://mlcommons.org/croissant/",
"rai": "http://mlcommons.org/croissant/RAI/",
"dct": "http://purl.org/dc/terms/",
Expand Down
⚡