telekinesis-datatypes 0.0.1rc3

Official core data types and serialization utilities package for the Telekinesis SDK and APIs.

Telekinesis Examples | Telekinesis Data

Telekinesis Datatypes

Telekinesis Datatypes is a core library providing canonical, strongly typed data structures for robotics and computer vision within the Telekinesis ecosystem. It includes:

• 3D data types: point clouds, meshes, transforms, and geometric primitives
• 2D data types: images, bounding boxes, masks, and pixel-space geometry
• Standardized representations for common perception and geometry formats
• Efficient serialization and deserialization for reliable data exchange

This library is used internally by the Telekinesis SDK as the foundation for data exchange across perception, planning, and learning components.

Installation

You will need to have Python 3.11 or higher set up to use the Telekinesis Datatypes Package.

Run the following command to install the Telekinesis Datatypes Package.

pip install telekinesis-datatypes

Example

import numpy as np
from datatypes import datatypes

# Create Rgba32 colors for R, G, B
red = datatypes.Rgba32([255, 0, 0, 255])
green = datatypes.Rgba32([0, 255, 0, 255])
blue = datatypes.Rgba32([0, 0, 255, 255])

print(f"Red color (packed uint32): {red.rgba}")
print(f"Green color (packed uint32): {green.rgba}")
print(f"Blue color (packed uint32): {blue.rgba}")

# Use __int__() to convert to integer
red_int = int(red)

print(f"Red as int: {red_int}")
print(f"Direct comparison: int(red) == red.rgba: {red_int == red.rgba}")

Expected output:

Red color (packed uint32): 4278190335
Green color (packed uint32): 16711935
Blue color (packed uint32): 65535
Red as int: 4278190335
Direct comparison: int(red) == red.rgba: True

Resources

• Examples
  Runnable usage examples for Telekinesis Datatypes: Telekinesis Examples (see examples/datatypes_examples.py)

• Documentation
  Full API reference and datatype descriptions: Datatypes Documentation

  Complete documentation: Telekinesis Docs

Support

For issues and questions:

• Create an issue in the GitHub repository.
• Contact the Telekinesis development team.

GitHub  •  LinkedIn  •  X  •  Discord