SuperAnnotate Schemas

1.0.49 verified Sat May 09 auth: no python

A library providing JSON schemas for SuperAnnotate's data formats, used for validating annotations, exports, and imports. Current version: 1.0.49. Release cadence is irregular, with frequent minor updates.

pip install superannotate-schemas

Common errors

error ModuleNotFoundError: No module named 'superannotate' ↓

cause Incorrect import path using dot instead of underscore.

fix

pip install superannotate-schemas and then from superannotate_schemas import ...

error jsonschema.exceptions.ValidationError: ... ↓

cause Annotation does not match the JSON schema. Usually due to missing required fields or wrong data types.

fix

Check the annotation structure against the schema definition. Use validate_or_raise to see detailed error.

Warnings

gotcha Import path uses underscores: superannotate_schemas (not superannotate.schemas). Many users mistakenly use the dot notation due to the package name. ↓

fix Use from superannotate_schemas import ...

gotcha Validation methods return boolean, not raise exceptions by default. To raise on invalid data, use .validate_or_raise(). ↓

fix Call AnnotationSchema.validate_or_raise(annotation) instead of AnnotationSchema.validate(annotation) to get detailed error.

deprecated Some older schema classes like 'OldAnnotationSchema' are deprecated and will be removed in a future version. ↓

fix Migrate to AnnotationSchema or ExportSchema.

Imports

AnnotationSchema

wrong

from superannotate.schemas import AnnotationSchema

correct

from superannotate_schemas import AnnotationSchema

Old path uses dot instead of underscore, common mistake.

ExportSchema

wrong

import superannotate_schemas.ExportSchema

correct

from superannotate_schemas import ExportSchema

Direct import without path traversal is incorrect.

Quickstart

Basic usage to validate an annotation against the schema.

from superannotate_schemas import AnnotationSchema
import os

api_key = os.environ.get('SUPERANNOTATE_API_KEY', '')
# Validate a sample annotation
annotation = {"type": "polygon", "points": [[0,0],[10,0],[10,10],[0,10]]}
is_valid = AnnotationSchema.validate(annotation)
print(is_valid)