PyTransit 2.6.1__tar.gz → 2.6.4__tar.gz

{PyTransit-2.6.1 → pytransit-2.6.4}/MANIFEST.in

@@ -1,4 +1,6 @@
 include pytransit/contamination/data/*
 include pytransit/models/opencl/*.cl
 include pytransit/models/roadrunner/*.cl
+include pytransit/models/roadrunner/*.py
+include pytransit/models/numba/*.py
 include pytransit/stars/data/*.fits

{PyTransit-2.6.1 → pytransit-2.6.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: PyTransit
-Version: 2.6.1
+Version: 2.6.4
 Summary: Fast and painless exoplanet transit light curve modelling.
 Author-email: Hannu Parviainen <hpparvi@gmail.com>
 License: GPLv2
@@ -30,6 +30,7 @@ Requires-Dist: deprecated
 Requires-Dist: uncertainties
 Requires-Dist: pyrr
 Requires-Dist: seaborn
+Requires-Dist: emcee
 
 PyTransit
 =========
@@ -40,14 +41,14 @@ PyTransit
 [![ASCL](https://img.shields.io/badge/ASCL-A1505.024-blue.svg?style=flat)](http://ascl.net/1505.024)
 [![DOI](https://zenodo.org/badge/5871/hpparvi/PyTransit.svg)](https://zenodo.org/badge/latestdoi/5871/hpparvi/PyTransit)
 
-Fast and easy-to-use tools for exoplanet transit light curve modelling with Python. PyTransit offers optimised CPU and 
-GPU implementations of popular exoplanet transit models with a unified interface, and thrives to be *the fastest and the 
-most versatile* tool for transit modelling in Python. 
+*PyTransit: fast and versatile exoplanet transit light curve modelling in Python.* PyTransit provides a set of optimised
+transit models with a unified API that makes modelling complex sets of heterogeneous light curve (nearly) as easy as 
+modelling individual transit light curves. The models are optimised with Numba which allows for model evaluation speeds
+paralleling Fortran and C-implementations but with hassle-free platform-independent multithreading.
+
+The package has been under continuous development since 2009, and is described in [Parviainen (2015)](http://arxiv.org/abs/1504.07433), 
+[Parviainen (2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract), and [Parviainen (2020b)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.3356P/abstract). 
 
-PyTransit makes transit model evaluation trivial whether modelling straightforward single-passband transit light curves or
-more complex science-cases, such as transmission spectroscopy. Further, the model can be evaluated for a large set of
-parameter sets simultaneously in parallel to optimize the evaluation speed with population-based MCMC samplers and 
-optimization methods, such as emcee and  DE-MCMC.
 
 ```Python
 from pytransit import RoadRunnerModel
@@ -57,9 +58,9 @@ tm.set_data(times)
 
 tm.evaluate(k=0.1, ldc=[0.2, 0.1], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
 
-tm.evaluate(k=[0.10, 0.12], ldc=[[0.2, 0.1, 0.5, 0.1]], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
+tm.evaluate(k=[0.10, 0.12], ldc=[[0.2, 0.1], [0.5, 0.1]], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
 
-tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1, 0.5, 0.1],[0.4, 0.2, 0.75, 0.1]],
+tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1], [0.5, 0.1],[0.4, 0.2, 0.75, 0.1]],
             t0=[0.0, 0.01], p=[1, 1], a=[3.0, 2.9], i=[.5*pi, .5*pi])
 ```
 
@@ -67,13 +68,15 @@ tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1, 0.5, 0.1],[0.4, 0.2,
 ![](doc/source/basic_example_2.svg)
 ![](doc/source/basic_example_3.svg)
 
-The package has been used in research since 2010, and is described in [Parviainen (2015)](http://arxiv.org/abs/1504.07433), 
-[Parviainen (2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract),
-and [Parviainen (2020b)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.3356P/abstract).
+
 
   
 ## Examples and tutorials
 
+### EMAC Workshop introduction video
+
+[![EMAC Workshop PyTransit introduction video](video1.png)](https://youtu.be/bLnxkFNrMDQ?si=OTjr4kUGK1kkhkLC)
+
 ### RoadRunner transit model
 
 RoadRunner [(Parviainen, 2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract) is a fast exoplanet transit model that can use any radially symmetric function to model stellar limb darkening 
@@ -98,36 +101,6 @@ transmission spectrum light curves.
 
 Read the docs at [pytransit.readthedocs.io](https://pytransit.readthedocs.io).
 
-## News
-- **21.10.2020, version 2.5**
-    - Version 2.5 makes modelling of TTVs trivial with ``pytransit.RoadRunnerModel`` and ``pytransit.QuadraticModel``.
-    - See the [TTV modelling example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/ttv_model_example.ipynb) for an example of how to evaluate the models for a TTV analysis.
-
-- **14.10.2020, version 2.4**
-    - Version 2.4 adds ``pytransit.EclipseModel`` to model secondary eclipses with as little hassle as possible.
-    - See the [secondary eclipse model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/example_eclipse_model.ipynb)
-    for an example of how to use it.
-
-- **16.9.2020, Version 2.3** 
-    - Version 2.3 adds ``OblateStarModel`` to model transits over gravity-darkened rapidly rotating star as presented by Barnes (2009).
-    - See the [oblate star model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/osmodel_example_1.ipynb)
-    for an example of usage.
-
-- **13.9.2020, Version 2.2**
-    - Version 2.2 brings a significant speedup to the evaluation speed of all models. The normalised planet-star distances are
-    now calculated using a Taylor-series expansion of the planet's (x,y) sky-plane position. The method gives a 2-6 x speedup to the transit
-    model evaluation and is detailed in Parviainen & Korth (2020, submitted to MNRAS)
-
-- **7.7.2020, Version 2.1** 
-    - Version 2.1 introduces the *RoadRunner* transit model (Parviainen, submitted to MNRAS, 2020), a fast exoplanet transit model 
-    that can use any radially symmetric function to model stellar limb darkening while still being faster to evaluate than 
-    the analytical transit model for quadratic limb darkening.
-    - See the [basic example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_1.ipynb),
-   the [custom limb darkening notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_2.ipynb),
-   and the [LDTk limb darkening model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_3.ipynb).
-
-
-
 Installation
 ------------
 ### PyPI
@@ -142,7 +115,7 @@ Clone the repository from github and do the normal python package installation
 
     git clone https://github.com/hpparvi/PyTransit.git
     cd PyTransit
-    python setup.py install
+    pip install .
 
 
 Citing

{PyTransit-2.6.1 → pytransit-2.6.4}/PyTransit.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: PyTransit
-Version: 2.6.1
+Version: 2.6.4
 Summary: Fast and painless exoplanet transit light curve modelling.
 Author-email: Hannu Parviainen <hpparvi@gmail.com>
 License: GPLv2
@@ -30,6 +30,7 @@ Requires-Dist: deprecated
 Requires-Dist: uncertainties
 Requires-Dist: pyrr
 Requires-Dist: seaborn
+Requires-Dist: emcee
 
 PyTransit
 =========
@@ -40,14 +41,14 @@ PyTransit
 [![ASCL](https://img.shields.io/badge/ASCL-A1505.024-blue.svg?style=flat)](http://ascl.net/1505.024)
 [![DOI](https://zenodo.org/badge/5871/hpparvi/PyTransit.svg)](https://zenodo.org/badge/latestdoi/5871/hpparvi/PyTransit)
 
-Fast and easy-to-use tools for exoplanet transit light curve modelling with Python. PyTransit offers optimised CPU and 
-GPU implementations of popular exoplanet transit models with a unified interface, and thrives to be *the fastest and the 
-most versatile* tool for transit modelling in Python. 
+*PyTransit: fast and versatile exoplanet transit light curve modelling in Python.* PyTransit provides a set of optimised
+transit models with a unified API that makes modelling complex sets of heterogeneous light curve (nearly) as easy as 
+modelling individual transit light curves. The models are optimised with Numba which allows for model evaluation speeds
+paralleling Fortran and C-implementations but with hassle-free platform-independent multithreading.
+
+The package has been under continuous development since 2009, and is described in [Parviainen (2015)](http://arxiv.org/abs/1504.07433), 
+[Parviainen (2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract), and [Parviainen (2020b)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.3356P/abstract). 
 
-PyTransit makes transit model evaluation trivial whether modelling straightforward single-passband transit light curves or
-more complex science-cases, such as transmission spectroscopy. Further, the model can be evaluated for a large set of
-parameter sets simultaneously in parallel to optimize the evaluation speed with population-based MCMC samplers and 
-optimization methods, such as emcee and  DE-MCMC.
 
 ```Python
 from pytransit import RoadRunnerModel
@@ -57,9 +58,9 @@ tm.set_data(times)
 
 tm.evaluate(k=0.1, ldc=[0.2, 0.1], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
 
-tm.evaluate(k=[0.10, 0.12], ldc=[[0.2, 0.1, 0.5, 0.1]], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
+tm.evaluate(k=[0.10, 0.12], ldc=[[0.2, 0.1], [0.5, 0.1]], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
 
-tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1, 0.5, 0.1],[0.4, 0.2, 0.75, 0.1]],
+tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1], [0.5, 0.1],[0.4, 0.2, 0.75, 0.1]],
             t0=[0.0, 0.01], p=[1, 1], a=[3.0, 2.9], i=[.5*pi, .5*pi])
 ```
 
@@ -67,13 +68,15 @@ tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1, 0.5, 0.1],[0.4, 0.2,
 ![](doc/source/basic_example_2.svg)
 ![](doc/source/basic_example_3.svg)
 
-The package has been used in research since 2010, and is described in [Parviainen (2015)](http://arxiv.org/abs/1504.07433), 
-[Parviainen (2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract),
-and [Parviainen (2020b)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.3356P/abstract).
+
 
   
 ## Examples and tutorials
 
+### EMAC Workshop introduction video
+
+[![EMAC Workshop PyTransit introduction video](video1.png)](https://youtu.be/bLnxkFNrMDQ?si=OTjr4kUGK1kkhkLC)
+
 ### RoadRunner transit model
 
 RoadRunner [(Parviainen, 2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract) is a fast exoplanet transit model that can use any radially symmetric function to model stellar limb darkening 
@@ -98,36 +101,6 @@ transmission spectrum light curves.
 
 Read the docs at [pytransit.readthedocs.io](https://pytransit.readthedocs.io).
 
-## News
-- **21.10.2020, version 2.5**
-    - Version 2.5 makes modelling of TTVs trivial with ``pytransit.RoadRunnerModel`` and ``pytransit.QuadraticModel``.
-    - See the [TTV modelling example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/ttv_model_example.ipynb) for an example of how to evaluate the models for a TTV analysis.
-
-- **14.10.2020, version 2.4**
-    - Version 2.4 adds ``pytransit.EclipseModel`` to model secondary eclipses with as little hassle as possible.
-    - See the [secondary eclipse model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/example_eclipse_model.ipynb)
-    for an example of how to use it.
-
-- **16.9.2020, Version 2.3** 
-    - Version 2.3 adds ``OblateStarModel`` to model transits over gravity-darkened rapidly rotating star as presented by Barnes (2009).
-    - See the [oblate star model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/osmodel_example_1.ipynb)
-    for an example of usage.
-
-- **13.9.2020, Version 2.2**
-    - Version 2.2 brings a significant speedup to the evaluation speed of all models. The normalised planet-star distances are
-    now calculated using a Taylor-series expansion of the planet's (x,y) sky-plane position. The method gives a 2-6 x speedup to the transit
-    model evaluation and is detailed in Parviainen & Korth (2020, submitted to MNRAS)
-
-- **7.7.2020, Version 2.1** 
-    - Version 2.1 introduces the *RoadRunner* transit model (Parviainen, submitted to MNRAS, 2020), a fast exoplanet transit model 
-    that can use any radially symmetric function to model stellar limb darkening while still being faster to evaluate than 
-    the analytical transit model for quadratic limb darkening.
-    - See the [basic example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_1.ipynb),
-   the [custom limb darkening notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_2.ipynb),
-   and the [LDTk limb darkening model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_3.ipynb).
-
-
-
 Installation
 ------------
 ### PyPI
@@ -142,7 +115,7 @@ Clone the repository from github and do the normal python package installation
 
     git clone https://github.com/hpparvi/PyTransit.git
     cd PyTransit
-    python setup.py install
+    pip install .
 
 
 Citing

{PyTransit-2.6.1 → pytransit-2.6.4}/PyTransit.egg-info/SOURCES.txt

@@ -79,6 +79,7 @@ pytransit/models/numba/__init__.py
 pytransit/models/numba/gdmodel.py
 pytransit/models/numba/general_nb.py
 pytransit/models/numba/ldmodels.py
+pytransit/models/numba/ldtkldm.py
 pytransit/models/numba/ma_chromosphere_nb.py
 pytransit/models/numba/ma_quadratic_nb.py
 pytransit/models/numba/ma_uniform_nb.py
@@ -93,12 +94,16 @@ pytransit/models/opencl/orbits.cl
 pytransit/models/opencl/qpower2.cl
 pytransit/models/roadrunner/__init__.py
 pytransit/models/roadrunner/common.py
+pytransit/models/roadrunner/ecintersection.py
 pytransit/models/roadrunner/model.py
 pytransit/models/roadrunner/model_full.py
 pytransit/models/roadrunner/model_simple.py
 pytransit/models/roadrunner/model_trspec.py
+pytransit/models/roadrunner/opmodel.py
+pytransit/models/roadrunner/opmodel_full.py
 pytransit/models/roadrunner/rrmodel.cl
 pytransit/models/roadrunner/rrmodel.py
+pytransit/models/roadrunner/rrmodel_cl.py
 pytransit/models/roadrunner/tsmodel.py
 pytransit/orbits/__init__.py
 pytransit/orbits/orbits.py

{PyTransit-2.6.1 → pytransit-2.6.4}/PyTransit.egg-info/requires.txt

@@ -12,3 +12,4 @@ deprecated
 uncertainties
 pyrr
 seaborn
+emcee

{PyTransit-2.6.1 → pytransit-2.6.4}/README.md

@@ -7,14 +7,14 @@ PyTransit
 [![ASCL](https://img.shields.io/badge/ASCL-A1505.024-blue.svg?style=flat)](http://ascl.net/1505.024)
 [![DOI](https://zenodo.org/badge/5871/hpparvi/PyTransit.svg)](https://zenodo.org/badge/latestdoi/5871/hpparvi/PyTransit)
 
-Fast and easy-to-use tools for exoplanet transit light curve modelling with Python. PyTransit offers optimised CPU and 
-GPU implementations of popular exoplanet transit models with a unified interface, and thrives to be *the fastest and the 
-most versatile* tool for transit modelling in Python. 
+*PyTransit: fast and versatile exoplanet transit light curve modelling in Python.* PyTransit provides a set of optimised
+transit models with a unified API that makes modelling complex sets of heterogeneous light curve (nearly) as easy as 
+modelling individual transit light curves. The models are optimised with Numba which allows for model evaluation speeds
+paralleling Fortran and C-implementations but with hassle-free platform-independent multithreading.
+
+The package has been under continuous development since 2009, and is described in [Parviainen (2015)](http://arxiv.org/abs/1504.07433), 
+[Parviainen (2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract), and [Parviainen (2020b)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.3356P/abstract). 
 
-PyTransit makes transit model evaluation trivial whether modelling straightforward single-passband transit light curves or
-more complex science-cases, such as transmission spectroscopy. Further, the model can be evaluated for a large set of
-parameter sets simultaneously in parallel to optimize the evaluation speed with population-based MCMC samplers and 
-optimization methods, such as emcee and  DE-MCMC.
 
 ```Python
 from pytransit import RoadRunnerModel
@@ -24,9 +24,9 @@ tm.set_data(times)
 
 tm.evaluate(k=0.1, ldc=[0.2, 0.1], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
 
-tm.evaluate(k=[0.10, 0.12], ldc=[[0.2, 0.1, 0.5, 0.1]], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
+tm.evaluate(k=[0.10, 0.12], ldc=[[0.2, 0.1], [0.5, 0.1]], t0=0.0, p=1.0, a=3.0, i=0.5*pi)
 
-tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1, 0.5, 0.1],[0.4, 0.2, 0.75, 0.1]],
+tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1], [0.5, 0.1],[0.4, 0.2, 0.75, 0.1]],
             t0=[0.0, 0.01], p=[1, 1], a=[3.0, 2.9], i=[.5*pi, .5*pi])
 ```
 
@@ -34,13 +34,15 @@ tm.evaluate(k=[[0.10, 0.12], [0.11, 0.13]], ldc=[[0.2, 0.1, 0.5, 0.1],[0.4, 0.2,
 ![](doc/source/basic_example_2.svg)
 ![](doc/source/basic_example_3.svg)
 
-The package has been used in research since 2010, and is described in [Parviainen (2015)](http://arxiv.org/abs/1504.07433), 
-[Parviainen (2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract),
-and [Parviainen (2020b)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.3356P/abstract).
+
 
   
 ## Examples and tutorials
 
+### EMAC Workshop introduction video
+
+[![EMAC Workshop PyTransit introduction video](video1.png)](https://youtu.be/bLnxkFNrMDQ?si=OTjr4kUGK1kkhkLC)
+
 ### RoadRunner transit model
 
 RoadRunner [(Parviainen, 2020a)](https://ui.adsabs.harvard.edu/abs/2020MNRAS.499.1633P/abstract) is a fast exoplanet transit model that can use any radially symmetric function to model stellar limb darkening 
@@ -65,36 +67,6 @@ transmission spectrum light curves.
 
 Read the docs at [pytransit.readthedocs.io](https://pytransit.readthedocs.io).
 
-## News
-- **21.10.2020, version 2.5**
-    - Version 2.5 makes modelling of TTVs trivial with ``pytransit.RoadRunnerModel`` and ``pytransit.QuadraticModel``.
-    - See the [TTV modelling example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/ttv_model_example.ipynb) for an example of how to evaluate the models for a TTV analysis.
-
-- **14.10.2020, version 2.4**
-    - Version 2.4 adds ``pytransit.EclipseModel`` to model secondary eclipses with as little hassle as possible.
-    - See the [secondary eclipse model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/example_eclipse_model.ipynb)
-    for an example of how to use it.
-
-- **16.9.2020, Version 2.3** 
-    - Version 2.3 adds ``OblateStarModel`` to model transits over gravity-darkened rapidly rotating star as presented by Barnes (2009).
-    - See the [oblate star model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/osmodel_example_1.ipynb)
-    for an example of usage.
-
-- **13.9.2020, Version 2.2**
-    - Version 2.2 brings a significant speedup to the evaluation speed of all models. The normalised planet-star distances are
-    now calculated using a Taylor-series expansion of the planet's (x,y) sky-plane position. The method gives a 2-6 x speedup to the transit
-    model evaluation and is detailed in Parviainen & Korth (2020, submitted to MNRAS)
-
-- **7.7.2020, Version 2.1** 
-    - Version 2.1 introduces the *RoadRunner* transit model (Parviainen, submitted to MNRAS, 2020), a fast exoplanet transit model 
-    that can use any radially symmetric function to model stellar limb darkening while still being faster to evaluate than 
-    the analytical transit model for quadratic limb darkening.
-    - See the [basic example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_1.ipynb),
-   the [custom limb darkening notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_2.ipynb),
-   and the [LDTk limb darkening model example notebook](https://github.com/hpparvi/PyTransit/blob/master/notebooks/roadrunner/roadrunner_model_example_3.ipynb).
-
-
-
 Installation
 ------------
 ### PyPI
@@ -109,7 +81,7 @@ Clone the repository from github and do the normal python package installation
 
     git clone https://github.com/hpparvi/PyTransit.git
     cd PyTransit
-    python setup.py install
+    pip install .
 
 
 Citing

{PyTransit-2.6.1 → pytransit-2.6.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "PyTransit"
-version = "2.6.1"
+version = "2.6.4"
 description = 'Fast and painless exoplanet transit light curve modelling.'
 authors=[{name='Hannu Parviainen', email='hpparvi@gmail.com'}]
 classifiers=[
@@ -14,7 +14,7 @@ classifiers=[
 ]
 keywords = ['astronomy',  'astrophysics',  'exoplanets']
 dependencies = ["numpy", "numba", "scipy", "pandas", "xarray", "tables", "meepmeep", "pyopencl", "ldtk",
-                "semantic_version", "deprecated", "uncertainties", "pyrr", "seaborn"]
+                "semantic_version", "deprecated", "uncertainties", "pyrr", "seaborn", "emcee"]
 requires-python = ">= 3.8"
 readme = "README.md"
 license = {text = "GPLv2"}
@@ -32,7 +32,3 @@ include-package-data = true
 [tool.setuptools.packages.find]
 include = ["pytransit"]
 namespaces = true
-
-#[tool.setuptools.package-data]
-#"pytransit.contamination.data" = ["*.nc"]
-#"pytransit" = ["*.fits", "*.txt", "*.csv", "*.h5", "*.nc", "*.cl"]

{PyTransit-2.6.1 → pytransit-2.6.4}/pytransit/__init__.py

@@ -64,12 +64,13 @@ with warnings.catch_warnings():
     from .models.eclipse_model import EclipseModel
     from .models.ma_chromosphere import ChromosphereModel
     from .models.general import GeneralModel
-    from .models.roadrunner.rrmodel import RoadRunnerModel
-    from .models.roadrunner.tsmodel import TransmissionSpectroscopyModel
     from .models.osmodel import OblateStarModel
     from .models.gdmodel import GravityDarkenedModel
 
+    from .models import RoadRunnerModel, OblatePlanetModel, TransmissionSpectroscopyModel
+
     TSModel = TransmissionSpectroscopyModel
+    OPModel = OblatePlanetModel
     RRModel = RoadRunnerModel
 
     # OpenCL models

{PyTransit-2.6.1 → pytransit-2.6.4}/pytransit/models/__init__.py

@@ -14,4 +14,11 @@
 #  You should have received a copy of the GNU General Public License
 #  along with this program.  If not, see <https://www.gnu.org/licenses/>.
 
-from .transitmodel import TransitModel
+from .transitmodel import TransitModel
+from .roadrunner.rrmodel import RoadRunnerModel
+from .roadrunner.opmodel import OblatePlanetModel
+from .roadrunner.tsmodel import TransmissionSpectroscopyModel
+
+RRModel = RoadRunnerModel
+OPModel = OblatePlanetModel
+TSModel = TransmissionSpectroscopyModel

{PyTransit-2.6.1 → pytransit-2.6.4}/pytransit/models/ldtkldm.py

@@ -22,6 +22,7 @@ from numpy import zeros, interp, pi, ndarray, linspace, meshgrid, transpose
 from scipy.interpolate import interpn, interpnd, RegularGridInterpolator as RGI
 
 from .ldmodel import LDModel
+from .numba.ldtkldm import trilinear_interpolation_set, integrate_profiles_set
 
 try:
     from ldtk import LDPSetCreator
@@ -53,33 +54,31 @@ class LDTkLDModel(LDModel):
         self.profiles = None
         self.rgi = None
 
-    def _init_interpolation(self, mu, nteff, nlogg, nz):
+    def _init_interpolation(self, mu):
         self.mu = mu
         self.nmu = mu.size
         c = self.sc.client
 
-        teffs = linspace(*c.teffl, nteff)
-        loggs = linspace(*c.loggl, nlogg)
-        zs = linspace(*c.zl, nz)
+        teffs = linspace(*c.teffl, self.sc.client.nteff)
+        loggs = linspace(*c.loggl, self.sc.client.nlogg)
+        zs = linspace(*c.zl, self.sc.client.nz)
         teffg, loggg, zg = meshgrid(teffs, loggs, zs)
+        self.teff0, self.dteff, self.nteff = teffs[0], teffs[1]-teffs[0], self.sc.client.nteff
+        self.logg0, self.dlogg, self.nlogg = loggs[0], loggs[1]-loggs[0], self.sc.client.nlogg
+        self.metal0, self.dmetal, self.nmetal = zs[0], zs[1]-zs[0], self.sc.client.nz
 
         self.ps = self.sc.create_profiles(teff=teffg.ravel(), logg=loggg.ravel(), metal=zg.ravel())
         self.ps.resample(mu=self.mu)
-        self.profiles = transpose(self.ps._ldps.copy(), axes=(1, 0, 2)).reshape((nteff, nlogg, nz, self.npb, self.nmu))
-        self.rgi = RGI((teffs, loggs, zs), self.profiles)
+        self.profiles = transpose(self.ps._ldps.copy(), axes=(1, 0, 2)).reshape((self.nteff, self.nlogg, self.nmetal, self.npb, self.nmu))
 
     def __call__(self, mu: ndarray, x: ndarray) -> Tuple[ndarray, ndarray]:
         if self.mu is None or id(mu) != id(self.mu):
-            self._init_interpolation(mu, 5, 3, 3)
-
-        npv = x.shape[0]
-        z = self.ps._z
-        ldp = zeros((npv, self.npb, self.nmu))
-        ldi = zeros((npv, self.npb))
-        for ipv in range(npv):
-            ldp[ipv] = self.rgi(x[ipv])
-            for ipb in range(self.npb):
-                ldi[ipv, ipb] = 2 * pi * ntrapz(z, z * ldp[ipv, ipb, :])
+            self._init_interpolation(mu)
+        ldp = trilinear_interpolation_set(self.profiles, x[:, 0, 0], x[:, 0, 1], x[:, 0, 2],
+                                          self.teff0, self.dteff, self.nteff,
+                                          self.logg0, self.dlogg, self.nlogg,
+                                          self.metal0, self.dmetal, self.nmetal)
+        ldi = integrate_profiles_set(self.mu, ldp)
         return ldp, ldi
 
     def _evaluate(self, mu: ndarray, x: ndarray) -> ndarray:

pytransit-2.6.4/pytransit/models/numba/ldtkldm.py

@@ -0,0 +1,91 @@
+from numba import njit, prange
+from numpy import zeros, floor, sqrt, pi
+
+
+@njit
+def trilinear_interpolation_unit_cube(data, x, y, z):
+    """Trilinear interpolation inside a unit cube.
+
+    Implementation taken from Paul Bourke's notes
+    https://paulbourke.net/miscellaneous/interpolation
+    """
+    n1 = data.shape[3]
+    n2 = data.shape[4]
+    ldp = zeros((n1, n2))
+    rx, ry, rz = (1.0 - x), (1.0 - y), (1.0 - z)
+
+    a1 = rx * ry * rz
+    a2 = x * ry * rz
+    b1 = rx * y * rz
+    b2 = rx * ry * z
+    c1 = x * ry * z
+    c2 = rx * y * z
+    d1 = x * y * rz
+    d2 = x * y * z
+
+    for i1 in range(n1):
+        for i2 in range(n2):
+            ldp[i1, i2] = (data[0, 0, 0, i1, i2] * a1 +
+                           data[1, 0, 0, i1, i2] * a2 +
+                           data[0, 1, 0, i1, i2] * b1 +
+                           data[0, 0, 1, i1, i2] * b2 +
+                           data[1, 0, 1, i1, i2] * c1 +
+                           data[0, 1, 1, i1, i2] * c2 +
+                           data[1, 1, 0, i1, i2] * d1 +
+                           data[1, 1, 1, i1, i2] * d2)
+    return ldp
+
+
+@njit
+def trilinear_interpolation_single(data, x, y, z, x0, dx, nx, y0, dy, ny, z0, dz, nz):
+    x = min(max(x, x0), x0 + nx * dx)
+    y = min(max(y, y0), y0 + ny * dy)
+    z = min(max(z, z0), z0 + nz * dz)
+    ix = int(floor((x - x0) / dx))
+    iy = int(floor((y - y0) / dy))
+    iz = int(floor((z - z0) / dz))
+    ax = (x - x0 - ix * dx) / dx
+    ay = (y - y0 - iy * dy) / dy
+    az = (z - z0 - iz * dz) / dz
+    return trilinear_interpolation_unit_cube(data[ix:ix + 2, iy:iy + 2, iz:iz + 2, :, :], ax, ay, az)
+
+
+@njit(parallel=True)
+def trilinear_interpolation_set(data, xs, ys, zs, x0, dx, nx, y0, dy, ny, z0, dz, nz):
+    npv = xs.shape[0]
+    ldp = zeros((npv, data.shape[3], data.shape[4]))
+    for ipv in prange(npv):
+        ldp[ipv, :, :] = trilinear_interpolation_single(data, xs[ipv], ys[ipv], zs[ipv], x0, dx, nx, y0, dy, ny, z0, dz,
+                                                        nz)
+    return ldp
+
+
+@njit
+def integrate_profiles_single(mu, ldp):
+    nmu = mu.size
+    npb = ldp.shape[0]
+    ldi = zeros(npb)
+
+    z = sqrt(1.0 - mu ** 2)
+    dz = z[1:] - z[:-1]
+    for ipb in range(npb):
+        for i in range(1, nmu):
+            ldi[ipb] += dz[i - 1] * 0.5 * (z[i] * ldp[ipb, i] + z[i - 1] * ldp[ipb, i - 1])
+    return 2.0 * pi * ldi
+
+
+@njit(parallel=True)
+def integrate_profiles_set(mu, ldp):
+    nmu = mu.size
+    npv = ldp.shape[0]
+    npb = ldp.shape[1]
+    ldi = zeros((npv, npb))
+
+    z = sqrt(1.0 - mu ** 2)
+    dz = z[1:] - z[:-1]
+    for j in prange(npv * npb):
+        ipv = j // npb
+        ipb = j % npb
+        for i in range(1, nmu):
+            ldi[ipv, ipb] += dz[i - 1] * 0.5 * (z[i] * ldp[ipv, ipb, i] + z[i - 1] * ldp[ipv, ipb, i - 1])
+    return 2.0 * pi * ldi