WIP: Brute force method to find the optimal number of harmonics for WaveX, DMWaveX, and CMWaveX

CHANGELOG-unreleased.md

@@ -11,6 +11,9 @@ the released changes.
 ### Changed
 - Moved the events -> TOAs and photon weights code into the function `load_events_weights` within `event_optimize`.
 - Updated the `maxMJD` argument in `event_optimize` to default to the current mjd
+- More validation for correlated noise components in `TimingModel`
+- Moved functions related to `WaveX`, `DMWaveX`, and `CMWaveX` from `pint.utils` to their own namespaces within `pint.models`
+- Moved `plrednoise_from_wavex` and similar functions from `pint.utils` to `pint.noise_analysis`
 - `maskParameter.__repr__()` output now includes the frozen attribute.
 - Changed default value of `FDJUMPLOG` to `Y`
 - Bumped `black` version to 24.x
@@ -19,7 +22,10 @@ the released changes.
 `pint.erfautils`, `pint.fits_utils`, `pint.logging` and `pint.residuals`
 - Doing `model.par = something` will try to assign to `par.quantity` or `par.value` but will give warning
 - `PLChromNoise` component to model chromatic red noise with a power law spectrum
+- Fourier series representation of chromatic noise (`CMWaveX`)
+- `pint.utils.cmwavex_setup` and `pint.utils.plchromnoise_from_cmwavex` functions
 ### Fixed
+- Bug in `DMWaveX.get_indices()` function
 - Explicit type conversion in `woodbury_dot()` function
 - Documentation: Fixed empty descriptions in the timing model components table
 - BIC implementation

requirements.txt

@@ -10,3 +10,4 @@ uncertainties
 loguru
 nestle>=0.2.0
 numdifftools
+joblib

setup.cfg

@@ -44,6 +44,7 @@ install_requires =
     loguru
     nestle>=0.2.0
     numdifftools
+    joblib
 
 [options.packages.find]
 where = src

src/pint/logging.py

@@ -142,7 +142,7 @@ def __init__(
         onlyonce_level: str = "INFO",
     ) -> None:
         r"""
-        Define regexs for messages that will only be seen once.  Use ``\S+`` for a variable that might change.
+        Define regexs for messages that will only be seen once.  Use ``\\S+`` for a variable that might change.
         If a message comes through with a new value for that variable, it will be seen.
 
         Make sure to escape other regex commands like ``()``.

src/pint/models/__init__.py

@@ -26,6 +26,7 @@
 from pint.models.binary_ddk import BinaryDDK
 from pint.models.binary_ell1 import BinaryELL1, BinaryELL1H, BinaryELL1k
 from pint.models.chromatic_model import ChromaticCM
+from pint.models.cmwavex import CMWaveX
 from pint.models.dispersion_model import (
     DispersionDM,
     DispersionDMX,

src/pint/models/binary_ddk.py

@@ -49,14 +49,19 @@ class BinaryDDK(BinaryDD):
     of the system from Earth, the finite size of the system, and the
     interaction of these with the proper motion.
 
-    From Kopeikin (1995) this includes :math:`\Delta_{\pi M}` (Equation 17), the mixed annual-orbital parallax term, which changes :math:`a_1` and :math:`\omega`
-    (:meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_a1_parallax` and :meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_omega_parallax`).
+    From Kopeikin (1995) this includes :math:`\\Delta_{\\pi M}` (Equation 17),
+    the mixed annual-orbital parallax term, which changes :math:`a_1` and :math:`\\omega`
+    (:meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_a1_parallax`
+    and :meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_omega_parallax`).
 
-    It does not include :math:`\Delta_{\pi P}`, the pure pulsar orbital parallax term (Equation 14).
+    It does not include :math:`\\Delta_{\\pi P}`, the pure pulsar orbital parallax term
+    (Equation 14).
 
-    From Kopeikin (1996) this includes apparent changes in :math:`\omega`, :math:`a_1`, and :math:`i` due to the proper motion
-    (:meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_omega_proper_motion`, :meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_a1_proper_motion`,
-    :meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_kin_proper_motion`) (Equations 8, 9, 10).
+    From Kopeikin (1996) this includes apparent changes in :math:`\\omega`, :math:`a_1`, and
+    :math:`i` due to the proper motion (:meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_omega_proper_motion`,
+    :meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_a1_proper_motion`,
+    :meth:`~pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel.delta_kin_proper_motion`)
+    (Equations 8, 9, 10).
 
     The actual calculations for this are done in
     :class:`pint.models.stand_alone_psr_binaries.DDK_model.DDKmodel`.
@@ -67,7 +72,7 @@ class BinaryDDK(BinaryDD):
        KIN
             the inclination angle: :math:`i`
        KOM
-            the longitude of the ascending node, Kopeikin (1995) Eq 9: :math:`\Omega`
+            the longitude of the ascending node, Kopeikin (1995) Eq 9: :math:`\\Omega`
        K96
             flag for Kopeikin binary model proper motion correction
 
@@ -233,19 +238,19 @@ def alternative_solutions(self):
 
         We first define the symmetry point where a1dot is zero (in equatorial coordinates):
 
-        :math:`KOM_0 = \\tan^{-1} (\mu_{\delta} / \mu_{\\alpha})`
+        :math:`KOM_0 = \\tan^{-1} (\\mu_{\\delta} / \\mu_{\\alpha})`
 
         The solutions are then:
 
         :math:`(KIN, KOM)`
 
-        :math:`(KIN, 2KOM_0 - KOM - 180^{\circ})`
+        :math:`(KIN, 2KOM_0 - KOM - 180^{\\circ})`
 
-        :math:`(180^{\circ}-KIN, KOM+180^{\circ})`
+        :math:`(180^{\\circ}-KIN, KOM+180^{\\circ})`
 
-        :math:`(180^{\circ}-KIN, 2KOM_0 - KOM)`
+        :math:`(180^{\\circ}-KIN, 2KOM_0 - KOM)`
 
-        All values will be between 0 and :math:`360^{\circ}`.
+        All values will be between 0 and :math:`360^{\\circ}`.
 
         Returns
         -------