rowan¶
Welcome to the documentation for rowan, a package for working with quaternions! Quaternions, which form a number system with various interesting properties, were originally developed for classical mechanics. Although they have since been largely displaced from this application by vector mathematics, they have become a standard method of representing rotations in three dimensions. Quaternions are now commonly used for this purpose in various fields, including computer graphics and attitude control.
The package is built entirely on top of NumPy and represents quaternions using NumPy arrays, meaning that all functions support arbitrarily high-dimensional arrays of quaternions.
Quaternions are encoded as arrays of shape (..., 4)
, with the convention that the final dimension of an array (a, b, c, d)
represents the quaternion a + bi + cj + dk
.
This package provides tools for standard algebraic operations on quaternions as well as a number of additional tools for e.g. measuring distances between quaternions, interpolating between them, and performing basic point-cloud mapping.
A particular focus of the rowan package is working with unit quaternions, which are a popular means of representing rotations in 3D.
In order to provide a unified framework for working with the various rotation formalisms in 3D, rowan allows easy interconversion between these formalisms.
Core features of rowan include (but are not limited to):
- Algebra (multiplication, exponentiation, etc).
- Derivatives and integrals of quaternions.
- Rotation and reflection operations, with conversions to and from matrices, axis angles, etc.
- Various distance metrics for quaternions.
- Basic point set registration, including solutions of the Procrustes problem and the Iterative Closest Point algorithm.
- Quaternion interpolation (slerp, squad).
rowan¶
Overview
rowan.conjugate |
Conjugates an array of quaternions. |
rowan.inverse |
Compute the inverse of an array of quaternions. |
rowan.exp |
Compute the natural exponential function \(e^q\). |
rowan.expb |
Compute the exponential function \(b^q\). |
rowan.exp10 |
Compute the exponential function \(10^q\). |
rowan.log |
Compute the quaternion natural logarithm. |
rowan.logb |
Compute the quaternion logarithm to some base b. |
rowan.log10 |
Compute the quaternion logarithm base 10. |
rowan.multiply |
Multiplies two arrays of quaternions. |
rowan.divide |
Divides two arrays of quaternions. |
rowan.norm |
Compute the quaternion norm. |
rowan.normalize |
Normalize quaternions. |
rowan.rotate |
Rotate a list of vectors by a corresponding set of quaternions. |
rowan.vector_vector_rotation |
Find the quaternion to rotate one vector onto another. |
rowan.from_euler |
Convert Euler angles to quaternions. |
rowan.to_euler |
Convert quaternions to Euler angles. |
rowan.from_matrix |
Convert the rotation matrices mat to quaternions. |
rowan.to_matrix |
Convert quaternions into rotation matrices. |
rowan.from_axis_angle |
Find quaternions to rotate a specified angle about a specified axis. |
rowan.to_axis_angle |
Convert the quaternions in q to axis-angle representations. |
rowan.from_mirror_plane |
Generate quaternions from mirror plane equations. |
rowan.reflect |
Reflect a list of vectors by a corresponding set of quaternions. |
rowan.equal |
Check whether two sets of quaternions are equal. |
rowan.not_equal |
Check whether two sets of quaternions are not equal. |
rowan.isfinite |
Test element-wise for finite quaternions. |
rowan.isinf |
Test element-wise for infinite quaternions. |
rowan.isnan |
Test element-wise for NaN quaternions. |
Details
The rowan package for working with quaternions.
The core rowan
package contains functions for operating on
quaternions. The core package is focused on robust implementations of key
functions like multiplication, exponentiation, norms, and others. Simple
functionality such as addition is inherited directly from NumPy due to
the representation of quaternions as NumPy arrays. Many core NumPy functions
implemented for normal arrays are reimplemented to work on quaternions (
such as allclose()
and isfinite()
). Additionally, NumPy
broadcasting
is enabled throughout rowan unless otherwise specified. This means that
any function of 2 (or more) quaternions can take arrays of shapes that do
not match and return results according to NumPy’s broadcasting rules.
-
rowan.
allclose
(p, q, **kwargs)¶ Check whether two sets of quaternions are all close.
This is a direct wrapper of the corresponding NumPy function.
Parameters: - p ((…, 4)
numpy.ndarray
) – First array of quaternions. - q ((…, 4)
numpy.ndarray
) – Second array of quaternions. - **kwargs – Keyword arguments to pass to np.allclose.
Returns: Whether all of
p
andq
are close.Return type: Example:
>>> rowan.allclose([1, 0, 0, 0], [1, 0, 0, 0]) True
- p ((…, 4)
-
rowan.
conjugate
(q)¶ Conjugates an array of quaternions.
Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Conjugates of q
.Return type: (…) numpy.ndarray
Example:
>>> rowan.conjugate([0.5, 0.5, -0.5, 0.5]) array([ 0.5, -0.5, 0.5, -0.5])
-
rowan.
divide
(qi, qj)¶ Divides two arrays of quaternions.
Division is non-commutative; this function returns \(q_i q_j^{-1}\).
Parameters: - qi ((…, 4)
numpy.ndarray
) – Dividend quaternions. - qj ((…, 4)
numpy.ndarray
) – Divisor quaternions.
Returns: Element-wise quotients of
q
(obeying broadcasting rules up to the last dimension ofqi
andqj
).Return type: (…)
numpy.ndarray
Example:
>>> rowan.divide([1, 0, 0, 0], [2, 0, 0, 0]) array([0.5, 0. , 0. , 0. ])
- qi ((…, 4)
-
rowan.
exp
(q)¶ Compute the natural exponential function \(e^q\).
The exponential of a quaternion in terms of its scalar and vector parts \(q = a + \boldsymbol{v}\) is defined by exponential power series: formula \(e^x = \sum_{k=0}^{\infty} \frac{x^k}{k!}\) as follows:
\[\begin{split}\begin{align} e^q &= e^{a+v} \\ &= e^a \left(\sum_{k=0}^{\infty} \frac{v^k}{k!} \right) \\ &= e^a \left(\cos \lvert \lvert \boldsymbol{v} \rvert \rvert + \frac{\boldsymbol{v}}{\lvert \lvert \boldsymbol{v} \rvert \rvert} \sin \lvert \lvert \boldsymbol{v} \rvert \rvert \right) \end{align}\end{split}\]Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Exponentials of q
.Return type: (…) numpy.ndarray
Example:
>>> rowan.exp([1, 0, 0, 0]) array([2.71828183, 0. , 0. , 0. ])
-
rowan.
expb
(q, b)¶ Compute the exponential function \(b^q\).
We define the exponential of a quaternion to an arbitrary base relative to the exponential function \(e^q\) using the change of base formula as follows:
\[\begin{split}\begin{align} b^q &= y \\ q &= \log_b y = \frac{\ln y}{\ln b}\\ y &= e^{q\ln b} \end{align}\end{split}\]Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Exponentials of q
.Return type: (…) numpy.ndarray
Example:
>>> rowan.expb([1, 0, 0, 0], 2) array([2., 0., 0., 0.])
-
rowan.
exp10
(q)¶ Compute the exponential function \(10^q\).
Wrapper around
expb()
.Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Exponentials of q
.Return type: (…) numpy.ndarray
Example:
>>> rowan.exp10([1, 0, 0, 0]) array([10., 0., 0., 0.])
-
rowan.
equal
(p, q)¶ Check whether two sets of quaternions are equal.
This function is a simple wrapper that checks array equality and then aggregates along the quaternion axis.
Parameters: - p ((…, 4)
numpy.ndarray
) – First array of quaternions. - q ((…, 4)
numpy.ndarray
) – Second array of quaternions.
Returns: Whether
p
andq
are equal.Return type: (…)
numpy.ndarray
of boolExample:
>>> rowan.equal([1, 0, 0, 0], [1, 0, 0, 0]) True
- p ((…, 4)
-
rowan.
from_axis_angle
(axes, angles)¶ Find quaternions to rotate a specified angle about a specified axis.
All angles are assumed to be counterclockwise rotations about the axis.
Parameters: - axes ((…, 3)
numpy.ndarray
) – An array of vectors (the axes). - angles (float or (…, 1)
numpy.ndarray
) – An array of angles in radians. Will be broadcasted to match shape of axes as needed.
Returns: The corresponding rotation quaternions.
Return type: (…, 4)
numpy.ndarray
Example:
>>> import numpy as np >>> rowan.from_axis_angle([[1, 0, 0]], np.pi/3) array([[0.8660254, 0.5 , 0. , 0. ]])
- axes ((…, 3)
-
rowan.
from_euler
(alpha, beta, gamma, convention='zyx', axis_type='intrinsic')¶ Convert Euler angles to quaternions.
For generality, the rotations are computed by composing a sequence of quaternions corresponding to axis-angle rotations. While more efficient implementations are possible, this method was chosen to prioritize flexibility since it works for essentially arbitrary Euler angles as long as intrinsic and extrinsic rotations are not intermixed.
Parameters: - alpha ((…)
numpy.ndarray
) – Array of \(\alpha\) values in radians. - beta ((…)
numpy.ndarray
) – Array of \(\beta\) values in radians. - gamma ((…)
numpy.ndarray
) – Array of \(\gamma\) values in radians. - convention (str) – One of the 12 valid conventions xzx, xyx, yxy, yzy, zyz, zxz, xzy, xyz, yxz, yzx, zyx, zxy.
- axes (str) – Whether to use extrinsic or intrinsic rotations.
Returns: Quaternions corresponding to the input angles.
Return type: (…, 4)
numpy.ndarray
Example:
>>> rowan.from_euler(0.3, 0.5, 0.7) array([0.91262714, 0.29377717, 0.27944389, 0.05213241])
- alpha ((…)
-
rowan.
from_matrix
(mat, require_orthogonal=True)¶ Convert the rotation matrices mat to quaternions.
This method uses the algorithm described by Bar-Itzhack in [Itzhack00]. The idea is to construct a matrix K whose largest eigenvalue corresponds to the desired quaternion. One of the strengths of the algorithm is that for nonorthogonal matrices it gives the closest quaternion representation rather than failing outright.
[Itzhack00] Itzhack Y. Bar-Itzhack. “New Method for Extracting the Quaternion from a Rotation Matrix”, Journal of Guidance, Control, and Dynamics, Vol. 23, No. 6 (2000), pp. 1085-1087 https://doi.org/10.2514/2.4654 Parameters: mat ((…, 3, 3) numpy.ndarray
) – An array of rotation matrices.Returns: The corresponding rotation quaternions. Return type: (…, 4) numpy.ndarray
Example:
>>> rowan.from_matrix([[1, 0, 0], [0, 1, 0], [0, 0, 1]]) array([ 1., -0., -0., -0.])
-
rowan.
from_mirror_plane
(x, y, z)¶ Generate quaternions from mirror plane equations.
Reflection quaternions can be constructed from the form \((0, x, y, z)\), i.e. with zero real component. The vector \((x, y, z)\) is the normal to the mirror plane.
Parameters: - x ((…)
numpy.ndarray
) – First planar component. - y ((…)
numpy.ndarray
) – Second planar component. - z ((…)
numpy.ndarray
) – Third planar component.
Returns: Quaternions reflecting about the input plane \((x, y, z)\).
Return type: (…)
numpy.ndarray
Example:
>>> rowan.from_mirror_plane(*(1, 2, 3)) array([0., 1., 2., 3.])
- x ((…)
-
rowan.
inverse
(q)¶ Compute the inverse of an array of quaternions.
Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Inverses of q
.Return type: (…) numpy.ndarray
Example:
>>> rowan.inverse([1, 0, 0, 0]) array([ 1., -0., -0., -0.])
-
rowan.
isclose
(p, q, **kwargs)¶ Element-wise check of whether two sets of quaternions are close.
This function is a simple wrapper that checks using the corresponding NumPy function and then aggregates along the quaternion axis.
Parameters: - p ((…, 4)
numpy.ndarray
) – First array of quaternions. - q ((…, 4)
numpy.ndarray
) – Second array of quaternions. - **kwargs – Keyword arguments to pass to np.isclose.
Returns: Whether
p
andq
are close element-wise.Return type: (…)
numpy.ndarray
of boolExample:
>>> rowan.allclose([[1, 0, 0, 0]], [[1, 0, 0, 0]]) True
- p ((…, 4)
-
rowan.
isinf
(q)¶ Test element-wise for infinite quaternions.
A quaternion is defined as infinite if any elements are infinite.
Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternionsReturns: Whether q
is infinite.Return type: (…) numpy.ndarray
of boolExample:
>>> import numpy as np >>> rowan.isinf([np.nan, 0, 0, 0]) False
-
rowan.
isfinite
(q)¶ Test element-wise for finite quaternions.
A quaternion is defined as finite if all elements are finite.
Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Whether q
is finite.Return type: (…) numpy.ndarray
of boolExample:
>>> rowan.isfinite([1, 0, 0, 0]) True
-
rowan.
isnan
(q)¶ Test element-wise for NaN quaternions.
A quaternion is defined as NaN if any elements are NaN.
Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Whether q
is NaN.Return type: (…) numpy.ndarray
of boolExample:
>>> import numpy as np >>> rowan.isnan([np.nan, 0, 0, 0]) True
-
rowan.
is_unit
(q)¶ Check if all input quaternions have unit norm.
Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Whether or not all inputs are unit quaternions. Return type: (…) numpy.ndarray
of boolExample:
>>> rowan.is_unit([10, 0, 0, 0]) False
-
rowan.
log
(q)¶ Compute the quaternion natural logarithm.
The natural of a quaternion in terms of its scalar and vector parts \(q = a + \boldsymbol{v}\) is defined by inverting the exponential formula (see
exp()
), and is defined by the formula \(\frac{x^k}{k!}\) as follows:\[\begin{equation} \ln(q) = \ln\lvert\lvert q \rvert\rvert + \frac{\boldsymbol{v}}{\lvert\lvert \boldsymbol{v} \rvert\rvert} \arccos\left(\frac{a}{q}\right) \end{equation}\]Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Logarithms of q
.Return type: (…) numpy.ndarray
Example:
>>> rowan.log([1, 0, 0, 0]) array([0., 0., 0., 0.])
-
rowan.
logb
(q, b)¶ Compute the quaternion logarithm to some base b.
The quaternion logarithm for arbitrary bases is defined using the standard change of basis formula relative to the natural logarithm.
\[\begin{split}\begin{align} \log_b q &= y \\ q &= b^y \\ \ln q &= y \ln b \\ y &= \log_b q = \frac{\ln q}{\ln b} \end{align}\end{split}\]Parameters: - q ((…, 4)
numpy.ndarray
) – Array of quaternions. - n ((…)
numpy.ndarray
) – Scalars to use as log bases.
Returns: Logarithms of
q
.Return type: (…)
numpy.ndarray
Example:
>>> rowan.logb([1, 0, 0, 0], 2) array([0., 0., 0., 0.])
- q ((…, 4)
-
rowan.
log10
(q)¶ Compute the quaternion logarithm base 10.
Wrapper around
logb()
.Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Logarithms of q
.Return type: (…) numpy.ndarray
Example:
>>> rowan.log10([1, 0, 0, 0]) array([0., 0., 0., 0.])
-
rowan.
multiply
(qi, qj)¶ Multiplies two arrays of quaternions.
Note that quaternion multiplication is generally non-commutative, so the first and second set of quaternions must be passed in the correct order.
Parameters: - qi ((…, 4)
numpy.ndarray
) – Array of left quaternions. - qj ((…, 4)
numpy.ndarray
) – Array of right quaternions.
Returns: Element-wise products of
q
(obeying broadcasting rules up to the last dimension ofqi
andqj
).Return type: (…)
numpy.ndarray
Example:
>>> rowan.multiply([1, 0, 0, 0], [2, 0, 0, 0]) array([2., 0., 0., 0.])
- qi ((…, 4)
-
rowan.
norm
(q)¶ Compute the quaternion norm.
Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Norms of q
.Return type: (…) numpy.ndarray
Example:
>>> rowan.norm([10, 0, 0, 0]) 10.0
-
rowan.
normalize
(q)¶ Normalize quaternions.
Parameters: q ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Normalized versions of q
.Return type: (…, 4) numpy.ndarray
Example:
>>> rowan.normalize([10, 0, 0, 0]) array([1., 0., 0., 0.])
-
rowan.
not_equal
(p, q)¶ Check whether two sets of quaternions are not equal.
This function is a simple wrapper that checks array equality and then aggregates along the quaternion axis.
Parameters: - p ((…, 4)
numpy.ndarray
) – First array of quaternions. - q ((…, 4)
numpy.ndarray
) – Second array of quaternions.
Returns: Whether
p
andq
are unequal.Return type: (…)
numpy.ndarray
of boolExample:
>>> rowan.not_equal([-1, 0, 0, 0], [1, 0, 0, 0]) True
- p ((…, 4)
-
rowan.
power
(q, n)¶ Compute the power of a quaternion \(q^n\).
Quaternions raised to a scalar power are defined according to the polar decomposition angle \(\theta\) and vector \(\hat{u}\): \(q^n = \lvert\lvert q \rvert\rvert^n \left( \cos(n\theta) + \hat{u} \sin(n\theta)\right)\). However, this can be computed more efficiently by noting that \(q^n = \exp(n \ln(q))\).
Parameters: - q ((…, 4)
numpy.ndarray
) – Array of quaternions. - n ((..) np.arrray) – Scalars to exponentiate quaternions with.
Returns: Powers of
q
.Return type: (…)
numpy.ndarray
Example:
>>> rowan.power([1, 0, 0, 0], 5) array([1., 0., 0., 0.])
- q ((…, 4)
-
rowan.
reflect
(q, v)¶ Reflect a list of vectors by a corresponding set of quaternions.
For help constructing a mirror plane, see
from_mirror_plane()
.Parameters: - q ((…, 4)
numpy.ndarray
) – Array of quaternions. - v ((…, 3)
numpy.ndarray
) – Array of vectors.
Returns: The result of reflecting
v
usingq
.Return type: (…, 4)
numpy.ndarray
Example:
>>> rowan.reflect([1, 0, 0, 0], [1, 1, 1]) array([1., 1., 1.])
- q ((…, 4)
-
rowan.
rotate
(q, v)¶ Rotate a list of vectors by a corresponding set of quaternions.
Parameters: - q ((…, 4)
numpy.ndarray
) – Array of quaternions. - v ((…, 3)
numpy.ndarray
) – Array of vectors.
Returns: The result of rotating
v
usingq
.Return type: (…, 4)
numpy.ndarray
Example:
>>> rowan.rotate([1, 0, 0, 0], [1, 1, 1]) array([1., 1., 1.])
- q ((…, 4)
-
rowan.
to_axis_angle
(q)¶ Convert the quaternions in q to axis-angle representations.
The output angles are counterclockwise rotations about the axis.
Parameters: q ((…, 4) numpy.ndarray
) – An array of quaternions.Returns: The axes and the angles (in radians). Return type: tuple[(…, 3) numpy.ndarray
, (…, 1)numpy.ndarray
]Example:
>>> rowan.to_axis_angle([[1, 0, 0, 0]]) (array([[0., 0., 0.]]), array([0.]))
-
rowan.
to_euler
(q, convention='zyx', axis_type='intrinsic')¶ Convert quaternions to Euler angles.
Euler angles are returned in the sequence provided, so in, e.g., the default case (‘zyx’), the angles returned are for a rotation \(Z(\alpha) Y(\beta) X(\gamma)\).
Note
In all cases, the \(\alpha\) and \(\gamma\) angles are between \(\pm \pi\). For proper Euler angles, \(\beta\) is between \(0\) and \(pi\) degrees. For Tait-Bryan angles, \(\beta\) lies between \(\pm\pi/2\).
For simplicity, quaternions are converted to matrices, which are then converted to their Euler angle representations. All equations for rotations are derived by considering compositions of the three elemental rotations about the three Cartesian axes:
Extrinsic rotations are represented by matrix multiplications in the proper order, so \(z-y-x\) is represented by the multiplication \(XYZ\) so that the system is rotated first about \(Z\), then about \(Y\), then finally \(X\). For intrinsic rotations, the order of rotations is reversed, meaning that it matches the order in which the matrices actually appear i.e. the \(z-y'-x''\) convention (yaw, pitch, roll) corresponds to the multiplication of matrices \(ZYX\). For proof of the relationship between intrinsic and extrinsic rotations, see the Wikipedia page on Davenport chained rotations.
For more information, see the Wikipedia page for Euler angles (specifically the section on converting between representations).
Warning
Euler angles are a highly problematic representation for a number of reasons, not least of which is the large number of possible conventions and their relative imprecision when compared to using quaternions (or axis-angle representations). If possible, you should avoid Euler angles and work with quaternions instead. If Euler angles are required, note that they are susceptible to gimbal lock, which leads to ambiguity in the representation of a given rotation. To address this issue, in cases where gimbal lock arises,
to_euler()
adopts the convention that \(\gamma=0\) and represents the rotation entirely in terms of \(\beta\) and \(\alpha\).Parameters: - q ((…, 4)
numpy.ndarray
) – Quaternions to transform. - convention (str) – One of the 6 valid conventions zxz, xyx, yzy, zyz, xzx, yxy.
- axes (str) – Whether to use extrinsic or intrinsic.
Returns: Euler angles \((\alpha, \beta, \gamma)\) corresponding to
q
.Return type: (…, 3)
numpy.ndarray
Example:
>>> import numpy as np >>> rands = np.random.rand(100, 3) >>> alpha, beta, gamma = rands.T >>> ql = rowan.from_euler(alpha, beta, gamma) >>> alpha_return, beta_return, gamma_return = np.split( ... rowan.to_euler(ql), 3, axis = 1) >>> assert(np.allclose(alpha_return.flatten(), alpha)) >>> assert(np.allclose(beta_return.flatten(), beta)) >>> assert(np.allclose(gamma_return.flatten(), gamma))
- q ((…, 4)
-
rowan.
to_matrix
(q, require_unit=True)¶ Convert quaternions into rotation matrices.
Uses the conversion described on Wikipedia.
Parameters: q ((…, 4) numpy.ndarray
) – An array of quaternions.Returns: The corresponding rotation matrices. Return type: (…, 3, 3) numpy.ndarray
Example:
>>> rowan.to_matrix([1, 0, 0, 0]) array([[1., 0., 0.], [0., 1., 0.], [0., 0., 1.]])
-
rowan.
vector_vector_rotation
(v1, v2)¶ Find the quaternion to rotate one vector onto another.
Note
Vector-vector rotation is underspecified, with one degree of freedom possible in the resulting quaternion. This method chooses to rotate by \(\pi\) around the vector bisecting v1 and v2.
Parameters: - v1 ((…, 3)
numpy.ndarray
) – Array of vectors to rotate. - v2 ((…, 3)
numpy.ndarray
) – Array of vector to rotate onto.
Returns: Quaternions that rotate
v1
ontov2
.Return type: (…, 4)
numpy.ndarray
Example:
>>> rowan.vector_vector_rotation([1, 0, 0], [0, 1, 0]) array([6.12323400e-17, 7.07106781e-01, 7.07106781e-01, 0.00000000e+00])
- v1 ((…, 3)
calculus¶
Overview
rowan.calculus.derivative |
Compute the instantaneous derivative of unit quaternions. |
rowan.calculus.integrate |
Integrate unit quaternions by angular velocity. |
Details
Compute derivatives and integrals of quaternions.
-
rowan.calculus.
derivative
(q, v)¶ Compute the instantaneous derivative of unit quaternions.
Derivatives of quaternions are defined by the equation:
\[\dot{q} = \frac{1}{2} \boldsymbol{v} q\]A derivation is provided here. For a more thorough explanation, see this page.
Parameters: - q ((…, 4)
numpy.ndarray
) – Array of quaternions. - v ((…, 3)
numpy.ndarray
) – Array of angular velocities.
Returns: Derivatives of
q
.Return type: (…, 4)
numpy.ndarray
Example:
>>> rowan.calculus.derivative([1, 0, 0, 0], [1, 0, 0]) array([0. , 0.5, 0. , 0. ])
- q ((…, 4)
-
rowan.calculus.
integrate
(q, v, dt)¶ Integrate unit quaternions by angular velocity.
The integral uses the following equation:
\[\dot{q} = \exp\left(\frac{1}{2} \boldsymbol{v} dt\right) q\]Note that this formula uses the quaternion exponential, so the argument to the exponential (which appears to be a vector) is promoted to a quaternion with scalar part 0 before the exponential is taken. A concise derivation is provided in this paper. This webpage contains a more thorough explanation.
Parameters: - q ((…, 4)
numpy.ndarray
) – Array of quaternions. - v ((…, 3)
numpy.ndarray
) – Array of angular velocities. - dt ((…)
numpy.ndarray
) – Array of timesteps.
Returns: Integrals of
q
.Return type: (…, 4)
numpy.ndarray
Example:
>>> rowan.calculus.integrate([1, 0, 0, 0], [0, 0, 1e-2], 1) array([0.9999875 , 0. , 0. , 0.00499998])
- q ((…, 4)
geometry¶
Overview
rowan.geometry.distance |
Determine the distance between quaternions p and q. |
rowan.geometry.sym_distance |
Determine the distance between quaternions p and q. |
rowan.geometry.riemann_exp_map |
Compute the exponential map on the Riemannian manifold \(\mathbb{H}^*\). |
rowan.geometry.riemann_log_map |
Compute the log map on the Riemannian manifold \(\mathbb{H}^*\). |
rowan.geometry.intrinsic_distance |
Compute the intrinsic distance between quaternions. |
rowan.geometry.sym_intrinsic_distance |
Compute the symmetrized intrinsic distance between quaternions. |
rowan.geometry.angle |
Compute the angle of rotation of a quaternion. |
Details
Various tools for working with the geometric representation of quaternions.
A particular focus is computing the distance between quaternions. These distance computations can be complicated, particularly good metrics for distance on the Riemannian manifold representing quaternions do not necessarily coincide with good metrics for similarities between rotations. An overview of distance measurements can be found in this paper.
-
rowan.geometry.
distance
(p, q)¶ Determine the distance between quaternions p and q.
This is the most basic distance that can be defined on the space of quaternions; it is the metric induced by the norm on this vector space \(\rho(p, q) = \lvert\lvert p - q \rvert\rvert\).
When applied to unit quaternions, this function produces values in the range \([0, 2]\).
Parameters: - p ((…, 4)
numpy.ndarray
) – First array of quaternions. - q ((…, 4)
numpy.ndarray
) – Second array of quaternions.
Returns: Distances between
p
andq
.Return type: (…)
numpy.ndarray
Example:
>>> rowan.geometry.distance([1, 0, 0, 0], [1, 0, 0, 0]) 0.0
- p ((…, 4)
-
rowan.geometry.
sym_distance
(p, q)¶ Determine the distance between quaternions p and q.
This is a symmetrized version of
distance()
that accounts for the fact that \(p\) and \(-p\) represent identical rotations. This makes it a useful measure of rotation similarity.Parameters: - p ((…, 4)
numpy.ndarray
) – First array of quaternions. - q ((…, 4)
numpy.ndarray
) – Second array of quaternions.
When applied to unit quaternions, this function produces values in the range \([0, \sqrt{2}]\).
Returns: Symmetrized distances between p
andq
.Return type: (…) numpy.ndarray
Example:
>>> rowan.geometry.sym_distance([1, 0, 0, 0], [-1, 0, 0, 0]) 0.0
- p ((…, 4)
-
rowan.geometry.
riemann_exp_map
(p, v)¶ Compute the exponential map on the Riemannian manifold \(\mathbb{H}^*\).
The nonzero quaternions form a Lie algebra \(\mathbb{H}^*\) that is also a Riemannian manifold. In general, given a point \(p\) on a Riemannian manifold \(\mathcal{M}\) and an element of the tangent space at \(p\), \(v \in T_p\mathcal{M}\), the Riemannian exponential map is defined by the geodesic starting at \(p\) and tracing out an arc of length \(v\) in the direction of \(v\). This function computes the endpoint of that path (which is itself a quaternion).
Explicitly, we define the exponential map as
\[\begin{equation} \textrm{Exp}_p(v) = p\exp(v) \end{equation}\]Parameters: - p ((…, 4)
numpy.ndarray
) – Points on the manifold of quaternions. - v ((…, 4)
numpy.ndarray
) – Tangent vectors to traverse.
Returns: The endpoints of the geodesic starting from \(p\) and traveling a distance \(\lvert\lvert v\rvert\rvert\) in the direction of \(v\).
Return type: (…, 4)
numpy.ndarray
Example:
rowan.geometry.riemann_exp_map([1, 0, 0, 0], [-1, 0, 0, 0])
- p ((…, 4)
-
rowan.geometry.
riemann_log_map
(p, q)¶ Compute the log map on the Riemannian manifold \(\mathbb{H}^*\).
This function inverts
riemann_exp_map()
. See that function for more details. In brief, given two quaternions p and q, this method returns a third quaternion parameterizing the geodesic passing from p to q. It is therefore an important measure of the distance between the two input quaternions.Parameters: - p ((…, 4)
numpy.ndarray
) – Starting points (quaternions). - q ((…, 4)
numpy.ndarray
) – Endpoints (quaternions).
Returns: The quaternions pointing from \(p\) to \(q\) with magnitudes equal to the length of the geodesics joining these quaternions.
Return type: (…, 4)
numpy.ndarray
Example:
>>> rowan.geometry.riemann_log_map([1, 0, 0, 0], [-1, 0, 0, 0]) array([0., 0., 0., 0.])
- p ((…, 4)
-
rowan.geometry.
intrinsic_distance
(p, q)¶ Compute the intrinsic distance between quaternions.
The quaternion distance is determined as the length of the quaternion joining the two quaternions (see
riemann_log_map()
). Rather than computing this directly, however, as shown in [Huynh09] we can compute this distance using the following equivalence:\[\begin{equation} \lvert\lvert \log(p q^{-1}) \rvert\rvert = 2\cos(\lvert\langle p, q \rangle\rvert) \end{equation}\]When applied to unit quaternions, this function produces values in the range \([0, \pi]\).
[Huynh09] Huynh DQ (2009) Metrics for 3D rotations: comparison and analysis. J Math Imaging Vis 35(2):155-164 Parameters: - p ((…, 4)
numpy.ndarray
) – First array of quaternions. - q ((…, 4)
numpy.ndarray
) – Second array of quaternions.
Returns: Intrinsic distances between
p
andq
.Return type: (…)
numpy.ndarray
Example:
rowan.geometry.intrinsic_distance([1, 0, 0, 0], [-1, 0, 0, 0])
- p ((…, 4)
-
rowan.geometry.
sym_intrinsic_distance
(p, q)¶ Compute the symmetrized intrinsic distance between quaternions.
This is a symmetrized version of
intrinsic_distance()
that accounts for the double cover \(SU(2)\rightarrow SO(3)\), making it a more useful metric for rotation similarity.When applied to unit quaternions, this function produces values in the range \([0, \frac{\pi}{2}]\).
Parameters: - p ((…, 4)
numpy.ndarray
) – First array of quaternions. - q ((…, 4)
numpy.ndarray
) – Second array of quaternions.
Returns: Symmetrized intrinsic distances between
p
andq
.Return type: (…)
numpy.ndarray
Example:
>>> rowan.geometry.sym_intrinsic_distance([1, 0, 0, 0], [-1, 0, 0, 0]) array(0.)
- p ((…, 4)
-
rowan.geometry.
angle
(p)¶ Compute the angle of rotation of a quaternion.
Note that this is identical to
2*intrinsic_distance(p, np.array([1, 0, 0, 0]))
.Parameters: p ((…, 4) numpy.ndarray
) – Array of quaternions.Returns: Angles traced out by the rotations p
.Return type: (…) numpy.ndarray
Example:
>>> rowan.geometry.angle([1, 0, 0, 0]) 0.0
interpolate¶
Overview
rowan.interpolate.slerp |
Spherical linear interpolation between p and q. |
rowan.interpolate.slerp_prime |
Compute the derivative of slerp. |
rowan.interpolate.squad |
Cubically interpolate between p and q. |
Details
Interpolate between pairs of quaternions.
The rowan package provides a simple interface to slerp, the standard method of quaternion interpolation for two quaternions.
-
rowan.interpolate.
slerp
(q0, q1, t, ensure_shortest=True)¶ Spherical linear interpolation between p and q.
The slerp formula can be easily expressed in terms of the quaternion exponential (see
rowan.exp()
).Parameters: - q0 ((…, 4)
numpy.ndarray
) – First array of quaternions. - q1 ((…, 4)
numpy.ndarray
) – Second array of quaternions. - t ((…)
numpy.ndarray
) – Interpolation parameter \(\in [0, 1]\) - ensure_shortest (bool) – Flip quaternions to ensure we traverse the geodesic in the shorter (\(<180^{\circ}\)) direction.
Note
Given inputs such that \(t\notin [0, 1]\), the values outside the range are simply assumed to be 0 or 1 (depending on which side of the interval they fall on).
Returns: Interpolations between p
andq
.Return type: (…, 4) numpy.ndarray
Example:
>>> import numpy as np >>> rowan.interpolate.slerp( ... [[1, 0, 0, 0]], [[np.sqrt(2)/2, np.sqrt(2)/2, 0, 0]], 0.5) array([[0.92387953, 0.38268343, 0. , 0. ]])
- q0 ((…, 4)
-
rowan.interpolate.
slerp_prime
(q0, q1, t, ensure_shortest=True)¶ Compute the derivative of slerp.
Parameters: - q0 ((…, 4)
numpy.ndarray
) – First set of quaternions. - q1 ((…, 4)
numpy.ndarray
) – Second set of quaternions. - t ((…)
numpy.ndarray
) – Interpolation parameter \(\in [0, 1]\) - ensure_shortest (bool) – Flip quaternions to ensure we traverse the geodesic in the shorter (\(<180^{\circ}\)) direction
Returns: The derivative of the interpolations between
p
andq
.Return type: (…, 4)
numpy.ndarray
Example:
import numpy as np q_slerp_prime rowan.interpolate.slerp_prime( [[1, 0, 0, 0]], [[np.sqrt(2)/2, np.sqrt(2)/2, 0, 0]], 0.5)
- q0 ((…, 4)
-
rowan.interpolate.
squad
(p, a, b, q, t)¶ Cubically interpolate between p and q.
The SQUAD formula is just a repeated application of Slerp between multiple quaternions as originally derived in [Shoemake85]:
\[\begin{equation} \textrm{squad}(p, a, b, q, t) = \textrm{slerp}(p, q, t) \left(\textrm{slerp}(p, q, t)^{-1}\textrm{slerp}(a, b, t) \right)^{2t(1-t)} \end{equation}\][Shoemake85] Ken Shoemake. Animating rotation with quaternion curves. SIGGRAPH Comput. Graph., 19(3):245-254, July 1985. Parameters: - p ((…, 4)
numpy.ndarray
) – First endpoint of interpolation. - a ((…, 4)
numpy.ndarray
) – First control point of interpolation. - b ((…, 4)
numpy.ndarray
) – Second control point of interpolation. - q ((…, 4)
numpy.ndarray
) – Second endpoint of interpolation. - t ((…)
numpy.ndarray
) – Interpolation parameter \(t \in [0, 1]\).
Returns: Interpolations between
p
andq
.Return type: (…, 4)
numpy.ndarray
Example:
>>> import numpy as np >>> rowan.interpolate.squad( ... [1, 0, 0, 0], [np.sqrt(2)/2, np.sqrt(2)/2, 0, 0], ... [0, np.sqrt(2)/2, np.sqrt(2)/2, 0], ... [0, 0, np.sqrt(2)/2, np.sqrt(2)/2], 0.5) array([[0.64550177, 0.47254009, 0.52564058, 0.28937053]])
- p ((…, 4)
mapping¶
Overview
rowan.mapping.kabsch |
Find the optimal rotation and translation to map between two sets of points. |
rowan.mapping.davenport |
Find the optimal rotation and translation to map between two sets of points. |
rowan.mapping.procrustes |
Solve the orthogonal Procrustes problem with algorithmic options. |
rowan.mapping.icp |
Find best mapping using the Iterative Closest Point algorithm. |
Details
Solve Procrustes-type problems and perform basic point-cloud registration.
The general space of problems that this subpackage addresses is a small subset of the broader space of point set registration, which attempts to optimally align two sets of points. In general, this mapping can be nonlinear. The restriction of this superposition to linear transformations composed of translation, rotation, and scaling is the study of Procrustes superposition, the first step in the field of Procrustes analysis, which performs the superposition in order to compare two (or more) shapes.
If points in the two sets have a known correspondence, the problem is much simpler. Various precise formulations exist that admit analytical formulations, such as the orthogonal Procrustes problem searching for an orthogonal transformation
or, if a pure rotation is desired, Wahba’s problem
Numerous algorithms to solve this problem exist, particularly in the field of aerospace engineering and robotics where this problem must be solved on embedded systems with limited processing. Since that constraint does not apply here, this package simply implements some of the most stable known methods irrespective of cost. In particular, this package contains the Kabsch algorithm, which solves Wahba’s problem using an SVD in the vein of Peter Schonemann’s original solution to the orthogonal Procrustes problem. Additionally this package contains the Davenport q method, which works directly with quaternions. The most popular algorithms for Wahba’s problem are variants of the q method that are faster at the cost of some stability; we omit these here.
In addition, rowan.mapping
also includes some functionality for
more general point set registration. If a point cloud has a set of known
symmetries, these can be tested explicitly by rowan.mapping
to
find the smallest rotation required for optimal mapping. If no such
correspondence is known at all, then the iterative closest point algorithm can
be used to approximate the mapping.
-
rowan.mapping.
kabsch
(X, Y, require_rotation=True)¶ Find the optimal rotation and translation to map between two sets of points.
This function implements the Kabsch algorithm, which minimizes the RMSD between two sets of points. One benefit of this approach is that the SVD works in dimensions > 3.
Parameters: - X ((N, m)
numpy.ndarray
) – First set of N points. - Y ((N, m)
numpy.ndarray
) – Second set of N points. - require_rotation (bool) – If false, the returned quaternion.
Returns: The rotation matrix and translation vector to map
X
ontoY
.Return type: tuple[(m, m)
numpy.ndarray
, (m, )numpy.ndarray
]Example:
>>> import numpy as np >>> # Create some random points, then make a random transformation of >>> # these points >>> points = np.random.rand(10, 3) >>> rotation = rowan.random.rand(1) >>> translation = np.random.rand(1, 3) >>> transformed_points = rowan.rotate(rotation, points) + translation >>> # Recover the rotation and check >>> R, t = rowan.mapping.kabsch(points, transformed_points) >>> q = rowan.from_matrix(R) >>> assert np.logical_or( ... np.allclose(rotation, q), np.allclose(rotation, -q)) >>> assert np.allclose(translation, t)
- X ((N, m)
-
rowan.mapping.
davenport
(X, Y)¶ Find the optimal rotation and translation to map between two sets of points.
This function implements the Davenport q-method, the most robust method and basis of most modern solvers. It involves the construction of a particular matrix, the Davenport K-matrix, which is then diagonalized to find the appropriate eigenvalues. More modern algorithms aim to solve the characteristic equation directly rather than diagonalizing, which can provide speed benefits at the potential cost of robustness. The implementation in
rowan
does not do this, instead simply computing the spectral decomposition.Parameters: - X ((N, 3)
numpy.ndarray
) – First set of N points. - Y ((N, 3)
numpy.ndarray
) – Second set of N points.
Returns: The quaternion and translation vector to map
X
ontoY
.Return type: tuple[(4, )
numpy.ndarray
, (m, )numpy.ndarray
]Example:
>>> import numpy as np >>> # Create some random points, then make a random transformation of >>> # these points >>> points = np.random.rand(10, 3) >>> rotation = rowan.random.rand(1) >>> translation = np.random.rand(1, 3) >>> transformed_points = rowan.rotate(rotation, points) + translation >>> # Recover the rotation and check >>> q, t = rowan.mapping.davenport(points, transformed_points) >>> assert np.logical_or( ... np.allclose(rotation, q), np.allclose(rotation, -q)) >>> assert np.allclose(translation, t)
- X ((N, 3)
-
rowan.mapping.
procrustes
(X, Y, method='best', equivalent_quaternions=None)¶ Solve the orthogonal Procrustes problem with algorithmic options.
Parameters: - X ((N, m)
numpy.ndarray
) – First set of N points. - Y ((N, m)
numpy.ndarray
) – Second set of N points. - method (str) – A method to use. Options are ‘kabsch’, ‘davenport’ and ‘horn’. The default is to select the best option (‘best’).
- equivalent_quaternions (array-like) – If the precise correspondence is not known, but the points are known to be part of a body with specific symmetries, the set of quaternions generating symmetry-equivalent configurations can be provided. These quaternions will be tested exhaustively to find the smallest symmetry-equivalent rotation.
Returns: The quaternion and translation vector to map
X
ontoY
.Return type: tuple[(4, )
numpy.ndarray
, (m, )numpy.ndarray
]Example:
>>> import numpy as np >>> # Create some random points, then make a random transformation of >>> # these points >>> points = np.random.rand(10, 3) >>> rotation = rowan.random.rand(1) >>> translation = np.random.rand(1, 3) >>> transformed_points = rowan.rotate(rotation, points) + translation >>> # Recover the rotation and check >>> q, t = rowan.mapping.procrustes( ... points, transformed_points, method='horn') >>> assert np.logical_or( ... np.allclose(rotation, q), np.allclose(rotation, -q)) >>> assert np.allclose(translation, t)
- X ((N, m)
-
rowan.mapping.
icp
(X, Y, method='best', unique_match=True, max_iterations=20, tolerance=0.001, return_indices=False)¶ Find best mapping using the Iterative Closest Point algorithm.
Parameters: - X ((N, m)
numpy.ndarray
) – First set of N points. - Y ((N, m)
numpy.ndarray
) – Second set of N points. - method (str) – A method to use for each alignment. Options are ‘kabsch’, ‘davenport’ and ‘horn’. The default is to select the best option (‘best’).
- unique_match (bool) – Whether to require nearest neighbors to be unique.
- max_iterations (int) – Number of iterations to attempt.
- tolerance (float) – Indicates convergence.
- return_indices (bool) – Whether to return indices.
Returns: The quaternion and translation vector to map
X
ontoY
. The (optional) last return value is an array of indices mapping points inX
to points inY
.Return type: tuple[(4, )
numpy.ndarray
, (m, )numpy.ndarray`[, (N, ) :class:`numpy.ndarray
]]Example:
>>> import numpy as np >>> # Create some random points >>> points = np.random.rand(10, 3) >>> # Only works for small rotations >>> rotation = rowan.from_axis_angle((1, 0, 0), 0.01) >>> # Apply a random translation and permutation >>> translation = np.random.rand(1, 3) >>> permutation = np.random.permutation(10) >>> transformed_points = rowan.rotate( ... rotation, points[permutation]) + translation >>> # Recover the rotation and check >>> R, t, indices = rowan.mapping.icp(points, transformed_points, ... return_indices=True) >>> q = rowan.from_matrix(R) >>> assert np.logical_or( ... np.allclose(rotation, q), np.allclose(rotation, -q)) >>> assert np.allclose(translation, t) >>> assert np.array_equal(permutation, indices)
- X ((N, m)
random¶
Overview
rowan.random.rand |
Generate random rotations uniformly distributed on a unit sphere. |
rowan.random.random_sample |
Generate random rotations uniformly. |
Details
Various functions for generating random sets of rotation quaternions.
Random quaternions in general can be generated by simply randomly sampling 4-vectors, and they can be converted into rotation quaternions by normalizing them. This package is strictly focused on generating uniform samples on \(SO(3)\), ensuring that rotations are uniformly sampled rather than just the space of unit quaternions.
-
rowan.random.
rand
(*args)¶ Generate random rotations uniformly distributed on a unit sphere.
This is a convenience function a la
np.random.rand
. If you want a function that takes a tuple as input, userandom_sample()
instead.Parameters: shape (tuple) – The shape of the array to generate. Returns: Random quaternions of the shape provided with an additional axis of length 4. Return type: numpy.ndarray
Example:
>>> rowan.random.rand(3, 3, 2)
-
rowan.random.
random_sample
(size=None)¶ Generate random rotations uniformly.
In general, sampling from the space of all quaternions will not generate uniform rotations. What we want is a distribution that accounts for the density of rotations, i.e., a distribution that is uniform with respect to the appropriate measure. The algorithm used here is detailed in [Shoe92].
[Shoe92] Shoemake, K.: Uniform random rotations. In: D. Kirk, editor, Graphics Gems III, pages 124-132. Academic, New York, 1992. Parameters: size (tuple) – The shape of the array to generate. Returns: Random quaternions of the shape provided with an additional axis of length 4. Return type: numpy.ndarray
Example:
>>> rowan.random.random_sample((3, 3, 2))
Development Guide¶
All contributions to rowan are welcome! Developers are invited to contribute to the framework by pull request to the package repository on github, and all users are welcome to provide contributions in the form of user feedback and bug reports. We recommend discussing new features in form of a proposal on the issue tracker for the appropriate project prior to development.
Design Philosophy and Code Guidelines¶
The goal of rowan is to provide a flexible, easy-to-use, and scalable approach to dealing with rotation representations. To best serve these goals, rowan operates entirely on NumPy arrays (the de facto standard for efficient multi-dimensional arrays in Python) and supports NumPy broadcasting wherever possible. Use of broadcasting ensures that rowan can take full advantage of NumPy performance, and in general all operations are very fast (benchmarks are included in the code base). Furthermore, to remain lightweight and easy to install, rowan is written in pure Python and has no hard dependencies aside from NumPy.
Code contributions should keep these ideals in mind and adhere to the following guidelines:
- Use the OneFlow model of development: - Both new features and bug fixes should be developed in branches based on
master
. - Hotfixes (critical bugs that need to be released fast) should be developed in a branch based on the latest tagged release.- All code must be compatible with all supported versions of Python (listed in the package
setup.py
file).- Avoid external dependencies where possible, and avoid introducing any hard dependencies. Soft dependencies are allowed for specific functionality, but such dependencies cannot impede the installation of rowan or the use of any other features.
- All code should adhere to the source code and documentation conventions discussed below.
- Create unit tests and integration tests as part of development.
- Preserve backwards-compatibility whenever possible. Make clear if something must change, and notify package maintainers that merging such changes will require a major release.
- Enable broadcasting if at all possible. Functions for which broadcasting is not available must be documented as such.
To provide a reasonable balance between a high level of backwards compatibility and a reasonable maintenance burden, rowan has adopted NEP 29 to limit the Python and NumPy versions that will be supported.
Tip
During continuous integration, the code is checked automatically with Flake8. To ensure your code is compliant before committing, you can set up a pre-commit hook using pre-commit (recommended) or use flake8’s built-in hooks installation:
flake8 --install-hook git
git config --bool flake8.strict true
Note
Please see the individual package documentation for detailed guidelines on how to contribute to a specific package.
Source Code Conventions¶
The rowan package adheres to a relatively strict set of style guidelines. All code in rowan should be formatted using black; a notable consequence of this is that the recommended max line length is 88, not the more common 80. Imports should be formatted using isort. For guidance on the style, see PEP 8 and the Google Python Style Guide, but any ambiguities should be resolved automatically by running black. All code should also follow the principles in PEP 20. In particular, always prefer simple, explicit code where possible, avoiding unnecessary convolution or complicated code that could be written more simply. Avoid writing code in a manner that will be difficult for others to understand.
Tip
Developers should format their code using black and isort locally using the commands:
black --exclude "coxeter/[polytri|bentley_ottman]" coxeter/ tests/
isort -rc coxeter/ tests/
Documentation¶
API documentation should be written as part of the docstrings of the package in the Google style. There is one notable exception to the guide: class properties should be documented in the getters functions, not as class attributes, to allow for more useful help messages and inheritance of docstrings. Docstrings may be validated using pydocstyle (or using the flake8-docstrings plugin as documented above). The official documentation is generated from the docstrings using Sphinx.
Unit Tests¶
All code should include a set of unit tests which test for correct behavior.
All tests should be placed in the tests
folder at the root of the project.
These tests should be as simple as possible, testing a single function each, and they should be kept as short as possible.
Tests should also be entirely deterministic: if you are using a random set of objects for testing, they should either be generated once and then stored in the tests/files
folder, or the random number generator in use should be seeded explicitly (e.g, numpy.random.seed
or random.seed
).
Tests should be written in the style of the standard Python unittest framework.
At all times, tests should be executable by simply running python -m unittest discover tests
from the root of the project.
Release Guide¶
To make a new release of rowan, follow the following steps:
- Make a new branch off of master based on the expected new version, e.g. release-2.3.1.
- Make any final changes as desired on this branch. Push the changes and ensure all tests are passing as expected on the new branch.
- Once the branch is completely finalized, run bumpversion with the appropriate type (patch, minor, major) so that the version now matches the version number in the branch name.
- Merge the branch back into master, then push master and push tags. The tagged commit will automatically trigger generation of binaries and upload to PyPI and conda-forge.
- Delete the release branch both locally and on the remote.
License¶
rowan BSD-3 Clause Open Source Software License
Copyright 2017-2020 The Regents of the University of Michigan
All rights reserved.
rowan may contain modifications ("Contributions") provided, and to which
copyright is held, by various Contributors who have granted The Regents of the
University of Michigan the right to modify and/or distribute such Contributions.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
3. Neither the name of the copyright holder nor the names of its contributors
may be used to endorse or promote products derived from this software without
specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR
ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Changelog¶
The format is based on Keep a Changelog. This project adheres to Semantic Versioning.
v1.3.0 - 2020-06-18¶
Added¶
- Extensive new validation using numerous flake8 plugins (including black for code style and pydocstyle for docstrings, among others).
Changed¶
- Drop support for all Python versions earlier than 3.6 and all NumPy versions before 1.15.
Fixed¶
- Docstring of geometry.angle was missing a factor of 2 in the comparison to intrinsic_distance.
- Docstrings of functions using support1d decorator were losing their docstring (fixed with functools.wraps).
- Docstrings of return types of all functions.
v1.2.2 - 2019-09-11¶
Added¶
- Mapping indices can be returned upon request from mapping.icp.
Fixed¶
- Euler angle calculations when cos(beta) = 0.
v1.2.1 - 2019-05-30¶
Added¶
- Official ContributorAgreement.
Fixed¶
- Broadcasting for nD arrays of quaternions in to_axis_angle is fixed.
- Providing equivalent quaternions to mapping.procrustes properly performs rotations.
v1.1.7 - 2019-01-23¶
Changed¶
- Stop requiring unit quaternions for rotation and reflection (allows scaling).
v1.1.1 - 2018-10-18¶
Added¶
- Automated deployment using CircleCI.
- Added PDF of paper to the repository.
Fixed¶
- Added missing factor of 2 in angle calculation.
- Fixed issue where method was not respected in rowan.mapping.
- Disabled equivalent quaternion feature and test of rowan.mapping, which has a known bug.
- Added missing negative in failing unit test.
v1.1.0 - 2018-07-30¶
Added¶
- Included benchmarks including comparison to alternatives.
- Installation instructions in the Sphinx documentation.
- More examples for rowan.mapping.
Changed¶
- All examples in docstrings now use the full paths of subpackages.
- All examples in docstrings import all needed packages aside from rowan.
Fixed¶
- Instability in vector_vector_rotation for antiparallel vectors.
- Various code style issues.
- Broken example in the Sphinx documentation.
v0.6.0 - 2018-04-20¶
Added¶
- Derivatives and integrals of quaternions.
- Point set registration methods and Procrustes analysis.
v0.5.0 - 2018-04-12¶
Added¶
- Various distance metrics on quaternion space.
- Quaternion interpolation.
Fixed¶
- Update empty __all__ variable in geometry to export functions.
v0.4.2 - 2018-04-09¶
Added¶
- Support for Read The Docs and Codecov.
- Simplify CircleCI testing suite.
- Minor changes to README.
- Properly update this document.
v0.4.0 - 2018-04-08¶
Added¶
- Add functions relating to exponentiation: exp, expb, exp10, log, logb, log10, power.
- Add core comparison functions for equality, closeness, finiteness.
v0.3.0 - 2018-03-31¶
Added¶
- Broadcasting works for all methods.
- Quaternion reflections.
- Random quaternion generation.
Changed¶
- Converting from Euler now takes alpha, beta, and gamma as separate args.
- Ensure more complete coverage.
Credits¶
The following people contributed to the rowan
package.
Vyas Ramasubramani <vramasub@umich.edu>, University of Michigan - Lead developer.
- Initial design.
- Wrote quaternion operations.
- Wrote calculus subpackage.
- Wrote geometry subpackage.
- Wrote interpolate subpackage.
- Wrote mapping subpackage.
- Wrote random subpackage.
- Wrote documentation.
Bradley Dice <bdice@bradleydice.com>, University of Michigan
- Code review.
- JOSS paper review.
Getting Started¶
Installation¶
The recommended methods for installing rowan are using pip or conda. To install the package from PyPI, execute:
$ pip install rowan --user
To install the package from conda, first add the conda-forge channel and then install rowan:
$ conda config --add channels conda-forge
$ conda install rowan
If you wish, you may also install rowan by cloning the repository and running the setup script:
$ git clone https://github.com/glotzerlab/rowan.git
$ cd rowan
$ python setup.py install --user
The minimum requirements for using rowan are:
- Python >= 3.6
- NumPy >= 1.15
Quickstart¶
This library can be used to work with quaternions by simply instantiating the appropriate NumPy arrays and passing them to the required functions. For example:
import rowan
import numpy as np
one = np.array([10, 0, 0, 0])
one_unit = rowan.normalize(one)
assert(np.all(one_unit == np.array([1, 0, 0, 0])))
if not np.all(one_unit == rowan.multiply(one_unit, one_unit)):
raise RuntimeError("Multiplication failed!")
one_vec = np.array([1, 0, 0])
rotated_vector = rowan.rotate(one_unit, one_vec)
mat = np.eye(3)
quat_rotate = rowan.from_matrix(mat)
alpha, beta, gamma = rowan.to_euler(quat_rotate)
quat_rotate_returned = rowan.from_euler(alpha, beta, gamma)
identity = rowan.to_matrix(quat_rotate_returned)
Running Tests¶
The package is currently tested for Python >= 3.6 on Unix-like systems. Continuous integrated testing is performed using CircleCI on these Python versions with NumPy versions 1.15 and above.
To run the packaged unit tests, execute the following line from the root of the repository:
python -m unittest discover tests
To check test coverage, make sure the coverage module is installed:
pip install coverage
and then run the packaged unit tests with the coverage module:
coverage run -m unittest discover tests
Running Benchmarks¶
Benchmarks for the package are contained in a Jupyter notebook in the benchmarks
folder in the root of the repository.
If you do not have or do not wish to use the notebook format, an equivalent Benchmarks.py script is also included.
The benchmarks compare rowan to two alternative packages, so you will need to install pyquaternion
and numpy_quaternion
if you wish to see those comparisons.
Building Documentation¶
You can also build this documentation from source if you clone the repository. The documentation is written in reStructuredText and compiled using Sphinx. To build from source, first install Sphinx:
pip install sphinx sphinx_rtd_theme
You can then use Sphinx to create the actual documentation in either PDF or HTML form by running the following commands in the rowan root directory:
cd doc
make html # For html output
make latexpdf # For a LaTeX compiled PDF file
open build/html/index.html
Support and Contribution¶
This package is hosted on GitHub. Please report any bugs or problems that you find on the issue tracker.
All contributions to rowan are welcomed via pull requests! Please see the development guide for more information on requirements for new code.