Transform2D

Represents a rigid 2D transform (SE(2)): a rotation plus a translation.

Parameters

Field	Type	Description
data	np.ndarray | list | tuple	Array-like input of shape (3, 3), converted to a contiguous float32 array.

Raises

Exception	Condition
TypeError	data can't be converted to float32 (e.g. non-numeric elements)
ValueError	data is not rank-2, or its shape isn't (3, 3)
ValueError	data contains a non-finite value (NaN/Inf)
ValueError	Last row isn't [0, 0, 1] (within transform_atol)
ValueError	The 2x2 rotation block isn't orthonormal (within transform_atol), or its determinant isn't ~1.0 (a determinant of -1 is a reflection, not a rotation)

Attributes

Attribute	Type	Description
data	np.ndarray	Defensive copy of the underlying (3, 3) float32 matrix. Assigning a new value re-validates it the same way as construction.
shape	tuple[int, ...]	Always (3, 3).
ndim	int	Always 2.
dtype	np.dtype	Always float32.
size	int	Always 9.
transform_atol	float	Class-level absolute tolerance (1e-4) used when validating the last row and the rotation block.

Methods

Method	Description
to_numpy(copy=True)	Returns the matrix as np.ndarray. Pass copy=False for a reference to the internal array instead — faster, but mutating it mutates the Transform2D too.
copy()	Returns a new Transform2D with an independent copy of the data.
Transform2D.coerce(value)	Returns value unchanged if it's already a Transform2D; otherwise wraps an array-like into one (running full validation). Raises TypeError for any other input.

Operators

Operation	Behavior
t == other	True only if other is also a Transform2D with element-equal data. False for anything else.
len(t)	Always 3 (length of the first axis).
np.asarray(t)	Works directly via __array__. Always returns a copy; use to_numpy(copy=False) for a zero-copy view.
hash(t)	Not supported — mutable via the data setter.

Visualization

datatypes.visualize(transform, entity_path=...) logs the transform's posed frame (X/Y axes rotated by atan2(data[1,0], data[0,0]) and translated by data[:2, 2]) alongside a labeled world-origin frame, so the rotation/translation are visible relative to identity. Passing label="..." attaches a floating text label at the posed frame's translation.

Example

"""Demonstrates the Telekinesis Transform2D datatype."""

import time

import numpy as np
from loguru import logger
import rerun as rr

from telekinesis import datatypes


def transform2d_example():
    """Demonstrate creation, access, visualization, update, NumPy interop, and serialization."""

    # ======================= Create ============================================
    theta = np.pi / 4
    matrix = np.array(
        [
            [np.cos(theta), -np.sin(theta), 1.0],
            [np.sin(theta), np.cos(theta), 2.0],
            [0.0, 0.0, 1.0],
        ]
    )
    transform2d = datatypes.Transform2D(matrix)
    logger.info(f"Original Transform2D: {transform2d}")

    # ======================= Inspect ===========================================
    data = transform2d.data
    shape = transform2d.shape
    size = transform2d.size
    dtype = transform2d.dtype
    ndim = transform2d.ndim
    numpy_array = transform2d.to_numpy()
    copy = transform2d.copy()

    logger.info(f"shape={shape}, size={size}, ndim={ndim}, dtype={dtype}")
    logger.info(f"Transform2D data:\n{data}")
    logger.info(f"NumPy array:\n{numpy_array}")
    logger.info(f"Copied Transform2D: {copy}")

    # ======================= Visualize =========================================
    rr.init("transform2d_example", spawn=True)
    datatypes.visualize(
        transform2d,
        entity_path="/Transform2D/main",
        label="My Transform2D",
    )

    # ======================= Update ============================================
    updated_theta = np.pi / 2
    updated_matrix = np.array(
        [
            [np.cos(updated_theta), -np.sin(updated_theta), 1.5],
            [np.sin(updated_theta), np.cos(updated_theta), 2.5],
            [0.0, 0.0, 1.0],
        ]
    )
    transform2d.data = updated_matrix
    logger.info(f"Updated Transform2D: {transform2d}")
    datatypes.visualize(
        transform2d,
        entity_path="/Transform2D/updated",
        label="Updated Transform2D",
    )

    # ======================= Arithmetic ========================================
    total = np.array([1, 1, 0]) + numpy_array
    logger.info(f"Sum of Transform2D with numpy array: {total}")

    # ======================= Serialize / Deserialize ===========================
    start = time.perf_counter()
    serialized = datatypes.serialize(transform2d)
    serialization_ms = (time.perf_counter() - start) * 1000

    start = time.perf_counter()
    deserialized = datatypes.deserialize(serialized)["param_0"]
    deserialization_ms = (time.perf_counter() - start) * 1000

    logger.info(f"Deserialized Transform2D: {deserialized}")
    logger.info(f"Round-trip successful: {transform2d == deserialized}")
    logger.info(f"Serialization time: {serialization_ms:.3f} ms")
    logger.info(f"Deserialization time: {deserialization_ms:.3f} ms")


if __name__ == "__main__":
    transform2d_example()