Category

Represents a single category in a dataset (e.g. COCO-style). Each category has required id, name, and supercategory, and optional isthing and color. This datatype is used as the schema for each element in Categories and in annotation workflows within the Telekinesis ecosystem.

Field

Required

Field	Type	Description
id	int	Unique category identifier (int).
name	str	Category name (str).
supercategory	str	Supercategory name (str).

Optional

Field	Type	Description
isthing	int	Whether the category is a "thing" (1) or "stuff" (0). Default 0. Must be 0 or 1.
color	list[int] or np.ndarray	RGB color as list or array of 3 integers. Default [0, 0, 0].

Methods

Method	Description
to_dict()	Returns a dictionary with keys: id, name, supercategory, isthing, color.

Example

From the datatypes examples:

from datatypes import datatypes
from loguru import logger

# ------------------------------------------------
# 1. Create Category instance
# ------------------------------------------------
category = datatypes.Category(
    id=1,
    name="object",
    supercategory="thing",
    isthing=1,
    color=[255, 0, 0],
)

# ------------------------------------------------
# 2. Access data via to_dict
# ------------------------------------------------
d = category.to_dict()
logger.info("Category id={}", d["id"])
logger.info("Category name={}", d["name"])
logger.info("Category supercategory={}", d["supercategory"])
logger.info("Category isthing={}", d["isthing"])
logger.info("Category color={}", d["color"])