Custom types (custom/*)
data type |
custom/* |
content types |
/ |
Custom types can be used for data that has very few specific uses or that is only used by a small number of plugins.
Warning
Before creating a new custom format, first consider any of the existing formats for entities, relations, graphs, or executables. Creating a new custom data type should be considered last.
Data Types
custom/kernel-matrix
custom/clusters
custom/attribute-distances
Represents distances between two entities based on a single specific attribute.
The output is delivered as an application/zip archive containing one .json file per attribute.
JSON Schema (per attribute file): A list of objects representing pairwise distances:
ID(string): Unique identifier for the comparison.entity_1_ID(string): ID of the first entity.entity_2_ID(string): ID of the second entity.href(string): Optional reference link.distance(float | null): The calculated distance score. Can be null if missing or invalid.
Example: Attribute dominanteCharaktereigenschaft of the MUSE dataset.
...
{
"ID": "44_21_1__45_7_2__dominanteCharaktereigenschaft",
"entity_1_ID": "44_21_1", (Id of entity with dominanteCharaktereigenschaft=aktiv)
"entity_2_ID": "45_7_2",
"href": "",
"distance": 0.5
},
...
Example Usage
Output generated by Similarities to distances transformers (@v0.2.1).
Input needed for Aggregators (@v0.2.1).
custom/attribute-similarities
Represents pairwise similarity scores between two entities based on a single specific attribute.
The output is delivered as an application/zip archive containing one .json file per attribute.
JSON Schema (per attribute file): A list of objects representing pairwise similarities:
ID(string): Unique identifier for the comparison (format:{ent1_ID}__{ent2_ID}__{attribute}).entity_1_ID(string): ID of the first entity.entity_2_ID(string): ID of the second entity.href(string): Optional reference link.similarity(float | null): The calculated similarity score.
Example: Attribute dominanteCharaktereigenschaft of the MUSE dataset.
...
{
"ID": "44_21_1__45_7_2__dominanteCharaktereigenschaft",
"entity_1_ID": "44_21_1", (Id of entity with dominanteCharaktereigenschaft=aktiv)
"entity_2_ID": "45_7_2",
"href": "",
"similarity": 0.5
},
...
Example Usage
Output generated by Sym Max Mean attribute comparer (@v0.1.2).
Input needed for Similarities to distances transformers (@v0.2.1).
custom/element-similarities
Represents pairwise similarity scores between two element values of entities based on their taxanomy.
The output is delivered as an application/zip archive containing one .json file per compared attribute.
JSON Schema (per attribute file): A list of objects representing the element-to-element similarity:
source(string/any): The first attribute value/element being compared.target(string/any): The second attribute value/element being compared.similarity(float): The calculated similarity score (e.g., Wu-Palmer score).
Example: Attribute dominanteCharaktereigenschaft of the MUSE dataset.
...
{
"source": "aktiv",
"target": "gut",
"similarity": 0.5
},
...
Example Usage
Output generated by Wu Palmer similarities (@v0.2.1).
Input needed for Sym Max Mean attribute comparer (@v0.1.2).
custom/entity-distances
Represents the total, aggregated distance between two entities across multiple attributes.
The output is delivered as a single application/json file.
JSON Schema: A list of objects representing the aggregated entity-to-entity distance:
ID(string): Unique identifier for the comparison (format:{ent1_ID}_{ent2_ID}).entity_1_ID(string): ID of the first entity.entity_2_ID(string): ID of the second entity.href(string): Optional reference link.distance(float): The aggregated distance score (e.g., Mean, Median, Max, Min).
Example: Attribute dominanteCharaktereigenschaft, dominanteFarbe, dominanterAlterseindruck, dominanterZustand, genre of the MUSE dataset.
...
{
"ID": "44_21_1_45_7_2",
"entity_1_ID": "44_21_1",
"entity_2_ID": "45_7_2",
"href": "",
"distance": 0.5133333333333334
},
...
Example Usage
Output generated by Aggregators (@v0.2.1).
Input needed for Multidimensional Scaling (MDS) (@v0.2.1).
custom/nisq-analyzer-result
custom/pca-metadata
custom/plot
Custom data type for presentational plots (e.g., html output from pyplot).
custom/hello-world-output
Custom data type for demo text output of the hello world plugins.
Current Non-Standard Custom Types
This is a list of custom types that are not using the custom/ prefix.
Danger
Do not use these types in new plugins!
representative-circuit/*should be replaced withexecutable/circuitand a data name starting withrepresentative-circuit.plot/*should be replaced withcustom/plottxt/*should be replaced withcustom/textqnn-weights/*vqc-metadata/*