---
---
<div class="dropdown2">
<span style="background-color: DodgerBlue; color: White; border:5px
solid DodgerBlue">Contents</span>
<div class="dropdown-content">
| Contents |
| :-: |
| 1. [History](#history) |
| 2. [Setup](#setup) |
| 3. [Structure](#structure) |
| 4. [Status](#status) |
| 5. [Versions](#versions) |
</div>
</div>
[exlink]: https://github.com/dokester/BayesicFitting/tree/master/BayesicFitting/examples
[srclink]: https://github.com/dokester/BayesicFitting/tree/master/BayesicFitting/source
[kllink]: https://github.com/dokester/BayesicFitting/tree/master/BayesicFitting/source/kernels
# Readme.
BayesicFitting is a package for model fitting and Bayesian evidence calculation.
In case you are wondering what that is about take a quick look at
the examples in the side bars.
We have a paper out in "Astronomy and Computing" about BayesicFitting.
[Kester and Mueller (2021)](./docs/references.md/#kester8).
Citation index for the BayesicFitting package:
DOI: 10.5281/zenodo.2597200
## What's new.
+ 05 Nov 2024 version 3.2.2<br>
Adapt to numpy-2 and pertaining Scipy, Astropy and MatplotLib
+ 27 March 2024 version 3.2.1<br>
A lot of fairly small stuff.
* \_\_status\_\_ changed to Alpha for some newer additions
* Priors:
- Adapt to input arrays
- Made limited and/or circular
- Proper integral when limited
* Engines:
- Add bestCheck and bestBoost
- Minor restructoring
* Models:
- Minor restructoring and renaming
- setLimits() replaced by setPrior() in RepeatingModel
- provide for more dimensions of outputs in derivative
* Problems:
- toString() method restructured
* Walker and WalkerList:
- Add logPrior attribute
- Change in inheritance reporting
* PhantomCollection:
- getList() method
* NestedSampler:
- Add repiter attribute: report every repiter when verbose=2
- Add bestBoost
* ModelDistribution:
- put internal sample() method in try-except block
* Tools:
- subclassof() method
- printclass finetuning
* Test and documentation. Update.
<a name="history"></a>
## 1. History
The BayesicFitting package is a python version of the the fitter classes
in Herschel Common Science System (HCSS). HCSS was the all encompassing
software system for the operations and analysis of the ESA satelite
Herschel. The HCSS version of the fitting software was written
in JAVA, mostly by me. I encoded features and classes that were requested
by my Herschel colleagues or that I remembered having used myself during
my lifelong career as data analyst for earlier satellites as IRAS, ISO
and AKARI. So most of the stuff in here was needed and used at a certain
moment in time. Later, the package was developing in directions that
were needed by my work for the James Webb Space Telescope (JWST).
The HCSS system is in the public domain under GPL3. It was used by the 3
instrument groups of the Herschel satellite to write calibration and
analysis software. Since the end of the mission HCSS is not being
maintained
I used a customized version of java2python (j2py on github) to translate
the JAVA classes to python. However, the actual code needed serious
pythonization. Every line has been inspected. Every construct has been
revised.
The documentation got most profit from the automated conversion. Also
the structure into classes, the inheritance, methods and dependencies
are largely the same as in the original HCSS.
<a name="setup"> </a>
## 2. Setup
The package is written in python3 although I am not aware of using any
specific python3 features. It uses numpy (>= 1.9) for its array
structure, scipy (>=1.0) for linear algebra and other stuff and astropy
(>=2.0) for units. Matplotlib (>=2.0) is used for plotting.
Download and unpack the BayesicFitting zip file from github. Move into
the BayesicFitting-master directory and run:
python setup.py install
where python is python3. Or install it as :
pip install BayesicFitting
<a name="structure"> </a>
## 3. Structure
### source
The BayesicFitting package consists of over 100 classes, each class in
its own file. These classes can be divided into 3 broad categories:
models, fitters and nested sampling. About 50 models, 10 fitters and the
remainder is needed to run the nested sampling algorithm. All these
classes are in a directory [BayesicFitting/source][srclink].
A special type of functions are found in [BayesicFitting/source/kernels][kllink].
They can be used to construct a model.
### examples
In [BayesicFitting/examples][exlink] a number of scripts can be
found to exercise the classes. They are in the form of jupyter
notebooks. Some are using real data; others have synthetic data
specially constructed to make some point.
All examples can be inspected by clicking on them. They will fold out in
the browser.
To actually exercise the examples and maybe adapt then, start a jupyter
notebook in your examples directory.
jupyter notebook
The program will open a list in your webbrowser where you can select a
notebook file (.ipynb), which can be run.
### documentation
In the [documenation](./docs)
directory a number of documents can be found.
+ [Manual](./docs/manual.md):
The manual for the package.
+ [Troubles](./docs/troubles.md):
A list of troublesome situations and what to do about it.
+ [Glossary](./docs/glossary.md):
A list of the terms used throughout this package, with explanations.
+ [Design](./docs/design.md):
An architectural design document, displaying the relationships between
the classes.
+ [Style](./docs/convention.md):
A few notes on my style of code and documenation.
+ [References](./docs/references.md):
A list of external references for BayesicFitting.
### test
Almost all classes have a test harness. These are located in
BayesicFitting/test. They can be execised as:
python -m unittest <file>
where python refers to python3 and file refers to one of the files in
BayesicFitting/test.<br>
As most functionality is tested in a test harness, examples on how to
use the classes can be found there too.
<a name="status"> </a>
## 4. Status
A package like this is never finished. Always more classes and/or
functionalities can be added. I present it now as it is in the hope it
will be usefull and it will generate feedback.
According to Wikipedia -> "Software release life cycle" it is called
"Perpetual Beta". It continues to be in a beta-release because new
classes and features can be added.
Some of the newer additions will be indicated as having and "Alpha"
status and keep that until they matured somewhat further.
These classes are more prone to change in their interfaces, methods etc.
More work needs to be done in:
* Introduction of more Classes: NeuralNetModel, Evolving Models, Filters (maybe) ...
<a name="versions"></a>
## 5. Versions
+ 4 Jan 2018 version 0.9.0.<br>
* Initial upload to github.
+ 26 Jan 2018 version <br>
+ 5 Mar 2018 version 1.0.1 <br>
* Package on pypi.com.
* Restructured all import statement to comply with PYPI package.
+ 14 Mar 2018 version 1.0.2 <br>
* Added Dynamic Models
* Added piping of models
+ 23 Mar 2018 version 1.0.3 <br>
* Some issues with ErrorDistributions and map fitting
* 2-d fitting examples added
* All examples revisited
* Links in README.md updated
+ 28 May 2018 version 1.0.4 <br>
* New classes: CircularUniformPrior, PseudoVoigtModel
* VoigtModel uses scipy.special.wozf() and has partials now,
* Refactoring Priors to the BaseModel
* Restructuring Dynamic
* Threading optional in NestedSampler.
* New classes: UniformErrorDistribution, FreeShapeModel and kernels/Tophat
* added to testharnesses and examples
+ 27 June 2018 version 1.0.5<br>
* New classes: RadialVelocityModel and MixedErrorDistribution
* testharnesses and examples
* documentation updates
+ 28 June 2018 version 1.0.6
* longdescription set to markdown (Still not OK on pypi.org)
+ 28 July 2018 version 1.0.7
* small compilation error in 1.0.6
+ 11 October 2018 version 1.0.8
* refactoring the setting of attributes in Models
* documentation (manual, design, etc.) updated.
+ 28 December 2018 version 2.0.0
* Introduction of Problem Classes:
- Problem. <br>
Base class for problems to be handled by NestedSampler.
- ClassicProblem. <br>
Common class for everything that was possible in version 1.
ClassicProblem is transparant as all interfaces to NestedSampler have remained
the same as they were in version 1.0, even though behind the scenes a
ClassicProblem has been invoked.
- ErrorsInXandYProblem. <br>
Problem that have errors in the xdata and in the ydata.
- ... more to come.
* Introduction of Walker and WalkerList to represent the internal ensemble
in NestedSampler.
* Adaptations in NestedSampler, ErrorDistributions, Engines, Sample, SampleList.
* Better separation of responsibilities of ErrorDistribution and Problem. <br>
Consequently ErrorDistribution has a new initialisation, which is incompatible
with previous versions. In most cases this has no effect on the calling
sequences of NestedSampler.
* Rename GenGaussErrorDistribution into ExponentialErrorDistribution.
* New testharnesses and examples.
* Adaptations of documentation: manual and design.
+ 16 Jan 2019 version 2.1.0
* MultipleOutputProblem.
Problems with more dimensional outputs
* StellarOrbitModel.
A 2 dim output model to calculate the orbit of a double star
* Keppler2ndLaw.
To calculate the radius and true anomaly according to Kepplers 2nd law.
(and derivatives)
* RadialVelocityModel: adapted to Kepplers2ndLaw. A slight change in the
order of the parameters.
* NestedSampler: some improvements in output layout.
* New tests, examples and updates for documentation.
+ 7 Feb 2019 version 2.2.0
* ChordEngine. Implementation of the POLYCHORD engine, developed
by Handley etal. (2015) MNRAS
* OrthogonalBasis. Helper class fot ChordEngine.
* Tests and examples
+ 19 Feb 2019 version 2.2.1
* AmoebaFitter still mentioned GenGaussErrorDistribution; replaced
by ExponentialErrorDistribution
* Some documentation issues repaired.
+ 20 Jun 2019 version 2.3.0
* Add LogisticModel and SampleMovie
* Periodic residuals in Problem
* Small issues repaired
* Rerun all examples
* Pictures moved to documentation/images
* Some documentation issues repaired.
+ 14 Nov 2019 version 2.4.0 to 2.4.2
* New Classes:
- DecisionTreeModel
A DecisionTree Model (DTM) is mostly defined on multiple input dimensions (axes).
It splits the data in 2 parts, according low and high values on a certain input axis.
The splitting can continue along other axes.
- Modifiable
Interface to define modifiable behaviour of some Models.
- StructureEngine
Engine to modify Models that implement Modifiable
* Introduce Table from astrolib as (multidimensional) xdata
* Some restructering necessitated by the classes above.
* Testcases and examples for the classes above
+ 3 Feb 2020 version 2.4.3
* Clean up and unification of the python doc strings.
* Reran all examples and test harnasses in python 3.7.
* Add random seed to several examples to make them more stable.
+ 17 Mar 2020 version 2.4.4
* Moved BayesicFitting/BayesicFitting/documentation to BayesicFitting/docs
* Added a references.md file which collects (external) references.
* Updated the docs files.
* Handling of weight in accordance with the definition in the Glossary.
* Add keyword tail= to formatter to display last items of an array.
+ 4 Jun 2020 version 2.5.0
* Add new models: BasicSplinesModel and SplinesDynamicModel
* Option for constraints on the likelihood
* Option for slow engines (working every slow-th iteration in NestedSampler)
* Restructure growPrior setting
* Print formatting in NestedSampler
* Adapt to SplinesDynamicModel
* Homogenized and improved plotoptions in test harnesses
* Three more examples added
+ 5 Jun 2020 versions 2.5.1
* Comment out NeuralNetModel (not yet available) and some typos.
+ 6 Jun 2020 versions 2.5.2
* Two more bugs smashed (in StartEngine and Prior)
+ 29 Jun 2020 versions 2.5.3
* Averaging of circular variables
* Update of static class attributes
* Attribute and printing issues.
+ 23 Oct 2020 version 2.6.0
* New class: PhantomSampler; adaptations in Engines, Explorer, WalkerList
* Restructured NestedSampler to accommodate PhantomSampler
* Test harnass for PhantomSampler
* Option: fix parameters in BasicSplinesModel
* Confusing __str__ method in compound models improved
+ 6 Nov 2020 versions 2.6.1
* Avoid infinities in unbound Priors
* mcycles in initialization of MonteCarlo
* convert xdata, ydata, weights using numpy.asarray
+ 11 Dec 2020 versions 2.6.2
* Add limits and circular to Priors
* Finetune Engines
+ 18 Feb 2021 version 2.7.0
* New class: EvidenceProblem & ModelDistribution; adaptations in NestedSampler and tests.
* Change in constrain method definition
* decay in ExpModel
* some seldom errors, clean-up & new test harnasses.
+ 19 April 2021 version 2.7.1
* remove CrossEngine completely
+ 20 April 2021 version 2.7.2
* put some tests on hold
* few minor issues/errors
+ 29 Oct 2021 version 2.8.0
* New class: BernoulliErrorDistribution and SoftMaxModel, tests, examples and data
* Adaptations To BernoulliED in some other classes
* Multi-dim input and output issues
* Updated some other tests and examples
* Documentation and other small issues.
+ 25 Nov 2021 version 2.8.1
* Cleanup in Plotter
* Documentation issues; Replaced style.md by code-style.md
* Correcting error on Windows systme
+ 07 Feb 2022 version 3.0.0
* New classes: AstropyModel and UserModel
* New class: NeuralNetUtilities
* New classes: NestedSolver, OrderProblem, SalesmanProblem, DistanceCostFunction
* New classes: OrderEngine, MoveEngine, SwitchEngine, LoopEngine, ShuffleEngine
* New classes: ReverseEngine, NearEngine, StartOrderEngine
* Make pipe work for more dimensional output | input
* Test harnesses for the new classes
* New examples for AstropyModel, UserModel and SalesmanProblem
* Update existing examples to improve coverage of pytest
* update Manual
+ 05 Apr 2022 version 3.0.1
* Addressing issue #18: UserModel does not work for multiple dimensions.
+ 19 Nov 2022 version 3.1.0
* Gauss2dErrorDistribution: New Class to handle correlated errors in X and Y
* ErrorDistribution and GaussErrorDistribution : adaptation for covariant errors.
* Small updates and corrections and removal of unused methods
* Added / corrected version information and documentation issues
* Rerun all examples and added tests
+ 18 Jan 2023 version 3.1.0 (still working on the same update)
* Implementing accuracy in fitters and samplers
* Update of documentation
* More tests
+ 18 Jan 2023 version 3.1.1
* remove GaussPriorNew from __init__.py
+ 31 Aug 2023 version 3.2.0
* The new class, PhantomCollection is part of NestedSampler. It contains a sorted
WalkerList, in which all valid positions are collected, proper walkers
and phantoms. Each iteration the phantoms with log likelihood lower than the
low limit for that iteration, are removed. The PhantomCollection is used
to get a better estimate on the size of the bounding box for the walkers and
to obtain starting positions for new walkers. In general the PhantomCollection
contains an order of magnitude more items than the WalkerList itself.
* NestedSampler has a new stopping criterion. It also stops when the log of the
relative contribution to the logZ (evidence) integral is less than -tolerance (=12).
* FootballModel: new class. Model to estimate strengths of football teams in
several key parameters.
* Address PhantomCollection and add **kwargs in all Engines
* Quadratic (in stead of linear) interpolation on edge in GalileanEngine
* Unnormalized Gauss prior changed into a normalized one
* Some documentation issues
* More dimensional arrays in LogFactorial
* __str__() method in NestedSolver and PhantomSampler
* Avoid numeric instabilities in sqrt in SampleList
* New tools in Tools
* More construction options in WalkerList
* New example: Uefa2022.ipynb
* New tests: TestPhantomCollection, TestFootballModel
* Adaptations in existing tests
* Reran all tests and examples
+ 20 March 2024 version 3.2.1
* See above in Whats new
<br><br><br><br>
Raw data
{
"_id": null,
"home_page": "https://www.bayesicfitting.nl",
"name": "BayesicFitting",
"maintainer": null,
"docs_url": null,
"requires_python": null,
"maintainer_email": null,
"keywords": "Bayesian, modeling, evidence, statistics, analysis, regression, inference, optimization, nested sampling, fitting",
"author": "Do Kester",
"author_email": "dokester@home.nl",
"download_url": "https://files.pythonhosted.org/packages/4d/95/59754ef6aaeca908c16bf29c53609a95082c7a0c130ccd8f1657a60cc13b/BayesicFitting-3.2.2.tar.gz",
"platform": null,
"description": "---\n---\n\n<div class=\"dropdown2\">\n <span style=\"background-color: DodgerBlue; color: White; border:5px \nsolid DodgerBlue\">Contents</span>\n <div class=\"dropdown-content\">\n\n| Contents |\n| :-: | \n| 1. [History](#history) |\n| 2. [Setup](#setup) | \n| 3. [Structure](#structure) | \n| 4. [Status](#status) | \n| 5. [Versions](#versions) |\n\n </div>\n</div>\n\n\n[exlink]: https://github.com/dokester/BayesicFitting/tree/master/BayesicFitting/examples\n[srclink]: https://github.com/dokester/BayesicFitting/tree/master/BayesicFitting/source\n[kllink]: https://github.com/dokester/BayesicFitting/tree/master/BayesicFitting/source/kernels\n\n \n\n# Readme.\n\nBayesicFitting is a package for model fitting and Bayesian evidence calculation.\n\nIn case you are wondering what that is about take a quick look at \nthe examples in the side bars.\n\nWe have a paper out in \"Astronomy and Computing\" about BayesicFitting.\n[Kester and Mueller (2021)](./docs/references.md/#kester8).\n\nCitation index for the BayesicFitting package:\nDOI: 10.5281/zenodo.2597200\n\n\n## What's new.\n\n + 05 Nov 2024 version 3.2.2<br>\n Adapt to numpy-2 and pertaining Scipy, Astropy and MatplotLib\n\n + 27 March 2024 version 3.2.1<br>\n A lot of fairly small stuff.\n * \\_\\_status\\_\\_ changed to Alpha for some newer additions\n * Priors: \n - Adapt to input arrays \n - Made limited and/or circular\n - Proper integral when limited\n * Engines:\n - Add bestCheck and bestBoost\n - Minor restructoring\n * Models:\n - Minor restructoring and renaming\n - setLimits() replaced by setPrior() in RepeatingModel\n - provide for more dimensions of outputs in derivative\n * Problems:\n - toString() method restructured\n * Walker and WalkerList:\n - Add logPrior attribute\n - Change in inheritance reporting\n * PhantomCollection:\n - getList() method\n * NestedSampler:\n - Add repiter attribute: report every repiter when verbose=2\n - Add bestBoost\n * ModelDistribution:\n - put internal sample() method in try-except block\n * Tools:\n - subclassof() method\n - printclass finetuning\n * Test and documentation. Update.\n\n\n<a name=\"history\"></a>\n## 1. History \n\nThe BayesicFitting package is a python version of the the fitter classes\nin Herschel Common Science System (HCSS). HCSS was the all encompassing\nsoftware system for the operations and analysis of the ESA satelite \nHerschel. The HCSS version of the fitting software was written\nin JAVA, mostly by me. I encoded features and classes that were requested\nby my Herschel colleagues or that I remembered having used myself during\nmy lifelong career as data analyst for earlier satellites as IRAS, ISO\nand AKARI. So most of the stuff in here was needed and used at a certain\nmoment in time. Later, the package was developing in directions that\nwere needed by my work for the James Webb Space Telescope (JWST). \n\nThe HCSS system is in the public domain under GPL3. It was used by the 3\ninstrument groups of the Herschel satellite to write calibration and\nanalysis software. Since the end of the mission HCSS is not being\nmaintained\n\nI used a customized version of java2python (j2py on github) to translate\nthe JAVA classes to python. However, the actual code needed serious\npythonization. Every line has been inspected. Every construct has been \nrevised.\n\nThe documentation got most profit from the automated conversion. Also\nthe structure into classes, the inheritance, methods and dependencies\nare largely the same as in the original HCSS.\n\n<a name=\"setup\"> </a>\n## 2. Setup \n\nThe package is written in python3 although I am not aware of using any\nspecific python3 features. It uses numpy (>= 1.9) for its array\nstructure, scipy (>=1.0) for linear algebra and other stuff and astropy\n(>=2.0) for units. Matplotlib (>=2.0) is used for plotting.\n\nDownload and unpack the BayesicFitting zip file from github. Move into \nthe BayesicFitting-master directory and run:\n\n python setup.py install\n\nwhere python is python3. Or install it as :\n\n pip install BayesicFitting\n\n<a name=\"structure\"> </a>\n## 3. Structure \n\n### source\n\nThe BayesicFitting package consists of over 100 classes, each class in\nits own file. These classes can be divided into 3 broad categories:\nmodels, fitters and nested sampling. About 50 models, 10 fitters and the\nremainder is needed to run the nested sampling algorithm. All these\nclasses are in a directory [BayesicFitting/source][srclink]. \nA special type of functions are found in [BayesicFitting/source/kernels][kllink]. \nThey can be used to construct a model.\n\n\n### examples\n\nIn [BayesicFitting/examples][exlink] a number of scripts can be\nfound to exercise the classes. They are in the form of jupyter\nnotebooks. Some are using real data; others have synthetic data\nspecially constructed to make some point. \n\nAll examples can be inspected by clicking on them. They will fold out in\nthe browser.\n\nTo actually exercise the examples and maybe adapt then, start a jupyter\nnotebook in your examples directory.\n\n jupyter notebook\n\nThe program will open a list in your webbrowser where you can select a\nnotebook file (.ipynb), which can be run.\n\n\n### documentation\n\nIn the [documenation](./docs)\ndirectory a number of documents can be found. \n\n+ [Manual](./docs/manual.md):\nThe manual for the package.\n\n+ [Troubles](./docs/troubles.md):\nA list of troublesome situations and what to do about it.\n\n+ [Glossary](./docs/glossary.md):\nA list of the terms used throughout this package, with explanations.\n\n+ [Design](./docs/design.md):\nAn architectural design document, displaying the relationships between \nthe classes. \n\n+ [Style](./docs/convention.md):\nA few notes on my style of code and documenation.\n\n+ [References](./docs/references.md):\nA list of external references for BayesicFitting.\n\n### test\n\nAlmost all classes have a test harness. These are located in\nBayesicFitting/test. They can be execised as:\n\n python -m unittest <file>\n\nwhere python refers to python3 and file refers to one of the files in\nBayesicFitting/test.<br>\nAs most functionality is tested in a test harness, examples on how to\nuse the classes can be found there too.\n\n<a name=\"status\"> </a> \n## 4. Status \n\nA package like this is never finished. Always more classes and/or\nfunctionalities can be added. I present it now as it is in the hope it\nwill be usefull and it will generate feedback.\n\nAccording to Wikipedia -> \"Software release life cycle\" it is called \n\"Perpetual Beta\". It continues to be in a beta-release because new \nclasses and features can be added.\n\nSome of the newer additions will be indicated as having and \"Alpha\" \nstatus and keep that until they matured somewhat further. \nThese classes are more prone to change in their interfaces, methods etc.\n\nMore work needs to be done in:\n\n * Introduction of more Classes: NeuralNetModel, Evolving Models, Filters (maybe) ...\n\n<a name=\"versions\"></a>\n## 5. Versions\n\n + 4 Jan 2018 version 0.9.0.<br>\n * Initial upload to github.\n\n + 26 Jan 2018 version <br>\n\n + 5 Mar 2018 version 1.0.1 <br>\n * Package on pypi.com. \n * Restructured all import statement to comply with PYPI package.\n\n + 14 Mar 2018 version 1.0.2 <br>\n * Added Dynamic Models \n * Added piping of models\n\n + 23 Mar 2018 version 1.0.3 <br>\n * Some issues with ErrorDistributions and map fitting\n * 2-d fitting examples added\n * All examples revisited\n * Links in README.md updated\n\n + 28 May 2018 version 1.0.4 <br>\n * New classes: CircularUniformPrior, PseudoVoigtModel\n * VoigtModel uses scipy.special.wozf() and has partials now,\n * Refactoring Priors to the BaseModel\n * Restructuring Dynamic\n * Threading optional in NestedSampler.\n * New classes: UniformErrorDistribution, FreeShapeModel and kernels/Tophat\n * added to testharnesses and examples\n\n + 27 June 2018 version 1.0.5<br>\n * New classes: RadialVelocityModel and MixedErrorDistribution\n * testharnesses and examples\n * documentation updates\n\n + 28 June 2018 version 1.0.6\n * longdescription set to markdown (Still not OK on pypi.org)\n\n + 28 July 2018 version 1.0.7\n * small compilation error in 1.0.6\n\n + 11 October 2018 version 1.0.8\n * refactoring the setting of attributes in Models\n * documentation (manual, design, etc.) updated.\n\n + 28 December 2018 version 2.0.0\n * Introduction of Problem Classes: \n - Problem. <br>\n\tBase class for problems to be handled by NestedSampler.\n - ClassicProblem. <br>\n\tCommon class for everything that was possible in version 1.\n\tClassicProblem is transparant as all interfaces to NestedSampler have remained \n\tthe same as they were in version 1.0, even though behind the scenes a \n\tClassicProblem has been invoked.\n - ErrorsInXandYProblem. <br>\n\tProblem that have errors in the xdata and in the ydata.\n - ... more to come.\n * Introduction of Walker and WalkerList to represent the internal ensemble\n in NestedSampler. \n * Adaptations in NestedSampler, ErrorDistributions, Engines, Sample, SampleList.\n * Better separation of responsibilities of ErrorDistribution and Problem. <br>\n Consequently ErrorDistribution has a new initialisation, which is incompatible \n with previous versions. In most cases this has no effect on the calling \n sequences of NestedSampler.\n * Rename GenGaussErrorDistribution into ExponentialErrorDistribution.\n * New testharnesses and examples.\n * Adaptations of documentation: manual and design.\n\n + 16 Jan 2019 version 2.1.0\n * MultipleOutputProblem.\n Problems with more dimensional outputs \n * StellarOrbitModel. \n A 2 dim output model to calculate the orbit of a double star\n * Keppler2ndLaw.\n To calculate the radius and true anomaly according to Kepplers 2nd law. \n (and derivatives)\n * RadialVelocityModel: adapted to Kepplers2ndLaw. A slight change in the \n order of the parameters.\n * NestedSampler: some improvements in output layout.\n * New tests, examples and updates for documentation.\n\n + 7 Feb 2019 version 2.2.0\n * ChordEngine. Implementation of the POLYCHORD engine, developed \n by Handley etal. (2015) MNRAS \n * OrthogonalBasis. Helper class fot ChordEngine.\n * Tests and examples\n\n + 19 Feb 2019 version 2.2.1 \n * AmoebaFitter still mentioned GenGaussErrorDistribution; replaced \n by ExponentialErrorDistribution\n * Some documentation issues repaired.\n\n + 20 Jun 2019 version 2.3.0 \n * Add LogisticModel and SampleMovie\n * Periodic residuals in Problem\n * Small issues repaired\n * Rerun all examples\n * Pictures moved to documentation/images\n * Some documentation issues repaired.\n\n + 14 Nov 2019 version 2.4.0 to 2.4.2\n * New Classes:\n - DecisionTreeModel\n\tA DecisionTree Model (DTM) is mostly defined on multiple input dimensions (axes).\n \tIt splits the data in 2 parts, according low and high values on a certain input axis.\n \tThe splitting can continue along other axes.\n - Modifiable\n\tInterface to define modifiable behaviour of some Models.\n - StructureEngine\n\tEngine to modify Models that implement Modifiable\n * Introduce Table from astrolib as (multidimensional) xdata\n * Some restructering necessitated by the classes above.\n * Testcases and examples for the classes above\n\n + 3 Feb 2020 version 2.4.3\n * Clean up and unification of the python doc strings.\n * Reran all examples and test harnasses in python 3.7.\n * Add random seed to several examples to make them more stable.\n\n + 17 Mar 2020 version 2.4.4\n * Moved BayesicFitting/BayesicFitting/documentation to BayesicFitting/docs\n * Added a references.md file which collects (external) references.\n * Updated the docs files.\n * Handling of weight in accordance with the definition in the Glossary.\n * Add keyword tail= to formatter to display last items of an array.\n\n + 4 Jun 2020 version 2.5.0\n * Add new models: BasicSplinesModel and SplinesDynamicModel\n * Option for constraints on the likelihood\n * Option for slow engines (working every slow-th iteration in NestedSampler)\n * Restructure growPrior setting\n * Print formatting in NestedSampler\n * Adapt to SplinesDynamicModel\n * Homogenized and improved plotoptions in test harnesses\n * Three more examples added\n\n + 5 Jun 2020 versions 2.5.1\n * Comment out NeuralNetModel (not yet available) and some typos.\n\n + 6 Jun 2020 versions 2.5.2\n * Two more bugs smashed (in StartEngine and Prior)\n\n + 29 Jun 2020 versions 2.5.3\n * Averaging of circular variables\n * Update of static class attributes\n * Attribute and printing issues.\n\n + 23 Oct 2020 version 2.6.0\n * New class: PhantomSampler; adaptations in Engines, Explorer, WalkerList\n * Restructured NestedSampler to accommodate PhantomSampler\n * Test harnass for PhantomSampler\n * Option: fix parameters in BasicSplinesModel\n * Confusing __str__ method in compound models improved\n\n + 6 Nov 2020 versions 2.6.1\n * Avoid infinities in unbound Priors\n * mcycles in initialization of MonteCarlo\n * convert xdata, ydata, weights using numpy.asarray\n\n + 11 Dec 2020 versions 2.6.2\n * Add limits and circular to Priors\n * Finetune Engines\n\n + 18 Feb 2021 version 2.7.0\n * New class: EvidenceProblem & ModelDistribution; adaptations in NestedSampler and tests.\n * Change in constrain method definition\n * decay in ExpModel\n * some seldom errors, clean-up & new test harnasses. \n\n + 19 April 2021 version 2.7.1\n * remove CrossEngine completely\n\n + 20 April 2021 version 2.7.2\n * put some tests on hold\n * few minor issues/errors\n\n + 29 Oct 2021 version 2.8.0\n * New class: BernoulliErrorDistribution and SoftMaxModel, tests, examples and data\n * Adaptations To BernoulliED in some other classes\n * Multi-dim input and output issues\n * Updated some other tests and examples\n * Documentation and other small issues.\n\n + 25 Nov 2021 version 2.8.1\n * Cleanup in Plotter\n * Documentation issues; Replaced style.md by code-style.md\n * Correcting error on Windows systme\n\n + 07 Feb 2022 version 3.0.0\n * New classes: AstropyModel and UserModel\n * New class: NeuralNetUtilities\n * New classes: NestedSolver, OrderProblem, SalesmanProblem, DistanceCostFunction\n * New classes: OrderEngine, MoveEngine, SwitchEngine, LoopEngine, ShuffleEngine\n * New classes: ReverseEngine, NearEngine, StartOrderEngine\n * Make pipe work for more dimensional output | input\n * Test harnesses for the new classes\n * New examples for AstropyModel, UserModel and SalesmanProblem\n * Update existing examples to improve coverage of pytest\n * update Manual\n\n + 05 Apr 2022 version 3.0.1\n * Addressing issue #18: UserModel does not work for multiple dimensions.\n\n + 19 Nov 2022 version 3.1.0\n * Gauss2dErrorDistribution: New Class to handle correlated errors in X and Y\n * ErrorDistribution and GaussErrorDistribution : adaptation for covariant errors.\n * Small updates and corrections and removal of unused methods\n * Added / corrected version information and documentation issues\n * Rerun all examples and added tests\n\n + 18 Jan 2023 version 3.1.0 (still working on the same update)\n * Implementing accuracy in fitters and samplers\n * Update of documentation\n * More tests\n\n + 18 Jan 2023 version 3.1.1\n * remove GaussPriorNew from __init__.py\n\n + 31 Aug 2023 version 3.2.0\n * The new class, PhantomCollection is part of NestedSampler. It contains a sorted \n WalkerList, in which all valid positions are collected, proper walkers \n and phantoms. Each iteration the phantoms with log likelihood lower than the \n low limit for that iteration, are removed. The PhantomCollection is used \n to get a better estimate on the size of the bounding box for the walkers and\n to obtain starting positions for new walkers. In general the PhantomCollection \n contains an order of magnitude more items than the WalkerList itself. \n * NestedSampler has a new stopping criterion. It also stops when the log of the \n relative contribution to the logZ (evidence) integral is less than -tolerance (=12). \n * FootballModel: new class. Model to estimate strengths of football teams in \n several key parameters. \n * Address PhantomCollection and add **kwargs in all Engines\n * Quadratic (in stead of linear) interpolation on edge in GalileanEngine\n * Unnormalized Gauss prior changed into a normalized one\n * Some documentation issues\n * More dimensional arrays in LogFactorial\n * __str__() method in NestedSolver and PhantomSampler\n * Avoid numeric instabilities in sqrt in SampleList\n * New tools in Tools\n * More construction options in WalkerList\n * New example: Uefa2022.ipynb\n * New tests: TestPhantomCollection, TestFootballModel\n * Adaptations in existing tests\n * Reran all tests and examples\n\n + 20 March 2024 version 3.2.1\n * See above in Whats new\n\n\n\n\n<br><br><br><br>\n\n",
"bugtrack_url": null,
"license": "LICENSE.txt",
"summary": "A Python Toolbox for Bayesian fitting.",
"version": "3.2.2",
"project_urls": {
"Homepage": "https://www.bayesicfitting.nl"
},
"split_keywords": [
"bayesian",
" modeling",
" evidence",
" statistics",
" analysis",
" regression",
" inference",
" optimization",
" nested sampling",
" fitting"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "973fcae41e814f18c1f9110a55dfa0070deb433a343a463f50618970b4284bbf",
"md5": "091418c29c0d88b8018c9e205937508c",
"sha256": "84b1e69a2dcf217ebd0986d7433036a65fc6bb3fc78c5bc69da54464914393a8"
},
"downloads": -1,
"filename": "BayesicFitting-3.2.2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "091418c29c0d88b8018c9e205937508c",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 5670903,
"upload_time": "2024-11-07T17:13:21",
"upload_time_iso_8601": "2024-11-07T17:13:21.100964Z",
"url": "https://files.pythonhosted.org/packages/97/3f/cae41e814f18c1f9110a55dfa0070deb433a343a463f50618970b4284bbf/BayesicFitting-3.2.2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "4d9559754ef6aaeca908c16bf29c53609a95082c7a0c130ccd8f1657a60cc13b",
"md5": "79522a4426019359629372b903994cf1",
"sha256": "f0ea915abaca6259c6fc445ad1b642ae9258dd0ae47d3a6a12afb3211d786420"
},
"downloads": -1,
"filename": "BayesicFitting-3.2.2.tar.gz",
"has_sig": false,
"md5_digest": "79522a4426019359629372b903994cf1",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 7370280,
"upload_time": "2024-11-07T17:13:24",
"upload_time_iso_8601": "2024-11-07T17:13:24.206268Z",
"url": "https://files.pythonhosted.org/packages/4d/95/59754ef6aaeca908c16bf29c53609a95082c7a0c130ccd8f1657a60cc13b/BayesicFitting-3.2.2.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-11-07 17:13:24",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "bayesicfitting"
}
JSON