add changelog entries from https://github.com/python-adaptive/adaptive/releases

Author: basnijholt

CHANGELOG.md

@@ -5,3 +5,205 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [unreleased]
+
+
+## version 0.9.0
+
+Since [0.8.0](https://github.com/python-adaptive/adaptive/tree/v0.8.0) we fixed the following [issues](https://github.com/python-adaptive/adaptive/issues):
+
+- [#217](https://github.com/python-adaptive/adaptive/issues/217) Command-line tool
+- [#211](https://github.com/python-adaptive/adaptive/issues/211) Defining inside main() in multiprocess will report error
+- [#208](https://github.com/python-adaptive/adaptive/issues/208) Inquiry on implementation of parallelism on the cluster
+- [#207](https://github.com/python-adaptive/adaptive/issues/207) PyYAML yaml.load(input) Deprecation
+- [#203](https://github.com/python-adaptive/adaptive/issues/203) jupyter-sphinx update Documentation enhancement
+- [#199](https://github.com/python-adaptive/adaptive/issues/199) jupyter-sphinx is pinned to non-existing branch
+
+and merged the following [Pull requests](https://github.com/python-adaptive/adaptive/pulls):
+- [#219](https://github.com/python-adaptive/adaptive/pull/219) pass value_scale to the LearnerND's loss_per_simplex function
+- [#209](https://github.com/python-adaptive/adaptive/pull/209) remove MPI4PY_MAX_WORKERS where it's not used
+- [#204](https://github.com/python-adaptive/adaptive/pull/204) use jupyter_sphinx v0.2.0 from conda instead of my branch
+- [#200](https://github.com/python-adaptive/adaptive/pull/200) ensure atomic writes when saving a file
+- [#193](https://github.com/python-adaptive/adaptive/pull/193) Add a SequenceLearner
+- [#188](https://github.com/python-adaptive/adaptive/pull/188) BalancingLearner: add a "cycle" strategy, sampling the learners one by one
+- [#202](https://github.com/python-adaptive/adaptive/pull/202) Authors
+- [#201](https://github.com/python-adaptive/adaptive/pull/201) Update tutorial.parallelism.rst
+- [#197](https://github.com/python-adaptive/adaptive/pull/197) Add option to display a progress bar when loading a BalancingLearner
+- [#195](https://github.com/python-adaptive/adaptive/pull/195) don't treat the no data case differently in the Learner1D  Learner1D
+- [#194](https://github.com/python-adaptive/adaptive/pull/194) pin everything in the docs/environment.yml file
+
+
+## version 0.8.0
+
+Since [0.7.0](https://github.com/python-adaptive/adaptive/tree/v0.7.0) we fixed the following [issues](https://github.com/python-adaptive/adaptive/issues):
+* [#7](https://github.com/python-adaptive/adaptive/issues/7) suggested points lie outside of domain Learner2D
+* [#39](https://github.com/python-adaptive/adaptive/issues/39) What should learners do when fed the same point twice
+* [#159](https://github.com/python-adaptive/adaptive/issues/159) BalancingLearner puts all points in the first child-learner when asking for points with no data present
+* [#148](https://github.com/python-adaptive/adaptive/issues/148) Loading data file with no data results in an error for the BalancingLearner
+* [#145](https://github.com/python-adaptive/adaptive/issues/145) Returning np.nan breaks the 1D learner
+* [#54](https://github.com/python-adaptive/adaptive/issues/54) Make learnerND datastructures immutable where possible
+* [gitlab:#134](https://gitlab.kwant-project.org/qt/adaptive/issues/134) Learner1D.load throws exception when file is empty
+* [#166](https://github.com/python-adaptive/adaptive/issues/166) live_plot broken with latest holoviews and bokeh
+* [#156](https://github.com/python-adaptive/adaptive/issues/156) Runner errors for Python 3.7 when done
+* [#159](https://github.com/python-adaptive/adaptive/issues/159) BalancingLearner puts all points in the first child-learner when asking for points with no data present
+* [#171](https://github.com/python-adaptive/adaptive/issues/171) default loss of LearnerND changed?
+* [#163](https://github.com/python-adaptive/adaptive/issues/163) Add a page to the documentation of papers where adaptive is used
+* [#179](https://github.com/python-adaptive/adaptive/issues/179) set python_requires in setup.py
+* [#175](https://github.com/python-adaptive/adaptive/issues/175) Underlying algorithm and MATLAB integration
+
+
+and merged the following [Pull requests](https://github.com/python-adaptive/adaptive/pulls):
+* [gitlab:!141](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/141): change the simplex_queue to a SortedKeyList
+* [gitlab:!142](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/142): make methods private in the LearnerND, closes #54
+* [#162](https://github.com/python-adaptive/adaptive/pull/162) test flat bands in the LearnerND
+* [#161](https://github.com/python-adaptive/adaptive/pull/161) import Iterable and Sized from collections.abc
+* [#160](https://github.com/python-adaptive/adaptive/pull/160) Distribute first points in a BalancingLearner
+* [#153](https://github.com/python-adaptive/adaptive/pull/153) invoke conda directly in CI
+* [#152](https://github.com/python-adaptive/adaptive/pull/152) fix bug in curvature_loss Learner1D bug
+* [#151](https://github.com/python-adaptive/adaptive/pull/151) handle NaN losses and add a test, closes #145
+* [#150](https://github.com/python-adaptive/adaptive/pull/150) fix `_get_data` for the BalancingLearner
+* [#149](https://github.com/python-adaptive/adaptive/pull/149) handle empty data files when loading, closes #148
+* [#147](https://github.com/python-adaptive/adaptive/pull/147) remove `_deepcopy_fix` and depend on sortedcollections >= 1.1
+* [#168](https://github.com/python-adaptive/adaptive/pull/168) Temporarily fix docs
+* [#167](https://github.com/python-adaptive/adaptive/pull/167) fix live_plot
+* [#164](https://github.com/python-adaptive/adaptive/pull/164) do not force shutdown the executor in the cleanup
+* [#172](https://github.com/python-adaptive/adaptive/issues/172) LearnerND: change the required loss to 1e-3 because the loss definition changed
+* [#177](https://github.com/python-adaptive/adaptive/pull/177) use the repo code in docs execute
+* [#176](https://github.com/python-adaptive/adaptive/pull/176) do not inline the HoloViews JS
+* [#174](https://github.com/python-adaptive/adaptive/pull/174) add a gallery page of Adaptive uses in scientific works
+* [#170](https://github.com/python-adaptive/adaptive/pull/170) Add logo to the documentation
+* [#180](https://github.com/python-adaptive/adaptive/pull/180) use setup(..., python_requires='>=3.6'), closes #179
+* [#182](https://github.com/python-adaptive/adaptive/pull/182) 2D: do not return points outside the bounds, closes #181 bug
+* [#185](https://github.com/python-adaptive/adaptive/pull/185) Add support for neighbours in loss computation in LearnerND
+* [#186](https://github.com/python-adaptive/adaptive/pull/186) renormalize the plots value axis on every update
+* [#189](https://github.com/python-adaptive/adaptive/pull/189) use pytest rather than py.test
+* [#190](https://github.com/python-adaptive/adaptive/pull/190) add support for mpi4py
+
+
+## version 0.7.0
+
+Since [0.6.0](https://gitlab.kwant-project.org/qt/adaptive/tree/v0.6.0) we fixed the following [issues](https://gitlab.kwant-project.org/qt/adaptive/issues):
+* [#122](https://gitlab.kwant-project.org/qt/adaptive/issues/122): Remove public `fname` learner attribute
+* [#119](https://gitlab.kwant-project.org/qt/adaptive/issues/119): (Learner1D) add possibility to use the direct neighbors in the loss
+* [#114](https://gitlab.kwant-project.org/qt/adaptive/issues/114): (LearnerND) allow any convex hull as domain
+* [#121](https://gitlab.kwant-project.org/qt/adaptive/issues/121): How to handle NaN?
+* [#107](https://gitlab.kwant-project.org/qt/adaptive/issues/107): Make BaseRunner an abstract base class
+* [#112](https://gitlab.kwant-project.org/qt/adaptive/issues/112): (LearnerND) add iso-surface plot feature
+* [#56](https://gitlab.kwant-project.org/qt/adaptive/issues/56): Improve plotting for learners
+* [#118](https://gitlab.kwant-project.org/qt/adaptive/issues/118): widgets don't show up on adaptive.readthedocs.io
+* [#91](https://gitlab.kwant-project.org/qt/adaptive/issues/91): Set up documentation
+* [#62](https://gitlab.kwant-project.org/qt/adaptive/issues/62): AverageLearner math domain error
+* [#113](https://gitlab.kwant-project.org/qt/adaptive/issues/113): make BalancingLearner work with the live_plot
+* [#111](https://gitlab.kwant-project.org/qt/adaptive/issues/111): (LearnerND) flat simplices are sometimes added on the surface of the triangulation
+* [#103](https://gitlab.kwant-project.org/qt/adaptive/issues/103): (BalancingLearner) make new balancinglearner that looks at the total loss rather than loss improvement
+* [#110](https://gitlab.kwant-project.org/qt/adaptive/issues/110): LearnerND triangulation incomplete
+* [#127](https://gitlab.kwant-project.org/qt/adaptive/issues/127): Typo in documentation for` adaptive.learner.learner2D.uniform_loss(ip)`
+* [#126](https://gitlab.kwant-project.org/qt/adaptive/issues/126): (Learner1D) improve time complexity
+* [#104](https://gitlab.kwant-project.org/qt/adaptive/issues/104): Learner1D could in some situations return -inf as loss improvement, which would make balancinglearner never choose to improve
+* [#128](https://gitlab.kwant-project.org/qt/adaptive/issues/128): (LearnerND) fix plotting of scaled domains
+* [#78](https://gitlab.kwant-project.org/qt/adaptive/issues/78): (LearnerND) scale y-values
+
+and merged the following [Merge Requests](https://gitlab.kwant-project.org/qt/adaptive/merge_requests):
+* [!131](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/131): Resolve "(Learner1D) add possibility to use the direct neighbors in the loss"
+* [!137](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/137): adhere to PEP008 by using absolute imports
+* [!135](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/135): test all the different loss functions in each test
+* [!133](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/133): make 'fname' a parameter to 'save' and 'load' only
+* [!136](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/136): build the Dockerimage used in CI
+* [!134](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/134): change resolution_loss to a factory function
+* [!118](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/118): add 'save' and 'load' to the learners and periodic saving to the Runner
+* [!127](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/127): Resolve "(LearnerND) allow any convex hull as domain"
+* [!130](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/130): save execution time on futures inside runners
+* [!111](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/111): Resolve "Make BaseRunner an abstract base class"
+* [!124](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/124): Resolve "(LearnerND) add iso-surface plot feature"
+* [!108](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/108): exponentially decay message frequency in live_info
+* [!129](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/129): add tutorials
+* [!120](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/120): add documentation
+* [!125](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/125): update to the latest miniver
+* [!126](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/126): add check_whitespace
+* [!123](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/123): add an option to plot a HoloMap with the BalancingLearner
+* [!122](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/122): implement 'npoints' strategy for the 'BalancingLearner'
+* [!119](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/119): (learnerND) no more (almost) flat simplices in the triangulation
+* [!109](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/109): make a BalancingLearner strategy that compares the total loss rather than loss improvement
+* [!117](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/117): Cache loss and display it in the live_info widget
+* [!121](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/121): 2D: add loss that minimizes the area of the triangle in 3D
+* [!139](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/139): Resolve "(Learner1D) improve time complexity"
+* [!140](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/140): Resolve "(LearnerND) fix plotting of scaled domains"
+* [!128](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/128): LearnerND scale output values before computing loss
+
+
+## version 0.6.0
+
+Since [0.5.0](https://gitlab.kwant-project.org/qt/adaptive/tree/v0.5.0) we fixed the following [issues](https://gitlab.kwant-project.org/qt/adaptive/issues):
+* [#66](https://gitlab.kwant-project.org/qt/adaptive/issues/66): (refactor) learner.tell(x, None) might be renamed to learner.tell_pending(x)
+* [#92](https://gitlab.kwant-project.org/qt/adaptive/issues/92): DeprecationWarning: sorted_dict.iloc is deprecated. Use SortedDict.keys() instead.
+* [#94](https://gitlab.kwant-project.org/qt/adaptive/issues/94): Learner1D breaks if right bound is added before the left bound
+* [#95](https://gitlab.kwant-project.org/qt/adaptive/issues/95): Learner1D's bound check algo in self.ask doesn't take self.data or self.pending_points
+* [#96](https://gitlab.kwant-project.org/qt/adaptive/issues/96): Learner1D fails when function returns a list instead of a numpy.array
+* [#97](https://gitlab.kwant-project.org/qt/adaptive/issues/97): Learner1D fails when a point (x, None) is added when x already exists
+* [#98](https://gitlab.kwant-project.org/qt/adaptive/issues/98): Learner1D.ask breaks when adding points in some order
+* [#99](https://gitlab.kwant-project.org/qt/adaptive/issues/99): Learner1D doesn't correctly set the interpolated loss when a point is added
+* [#101](https://gitlab.kwant-project.org/qt/adaptive/issues/101): How should learners handle data that is outside of the domain
+* [#102](https://gitlab.kwant-project.org/qt/adaptive/issues/102): No tests for the 'BalancingLearner'
+* [#105](https://gitlab.kwant-project.org/qt/adaptive/issues/105): LearnerND fails for BalancingLearner test
+* [#108](https://gitlab.kwant-project.org/qt/adaptive/issues/108): (BalancingLearner) loss is cached incorrectly
+* [#109](https://gitlab.kwant-project.org/qt/adaptive/issues/109): Learner2D suggests same point twice
+
+and merged the following [Merge Requests](https://gitlab.kwant-project.org/qt/adaptive/merge_requests):
+* [!93](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/93): add a release guide
+* [!94](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/94): add runner.max_retries
+* [!95](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/95): 1D: fix the rare case where the right boundary point exists before the left bound
+* [!96](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/96): More efficient 'tell_many'
+* [!97](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/97): Fix #97 and #98
+* [!98](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/98): Resolve "DeprecationWarning: sorted_dict.iloc is deprecated. Use SortedDict.keys() instead."
+* [!99](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/99): Resolve "Learner1D's bound check algo in self.ask doesn't take self.data or self.pending_points"
+* [!100](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/100): Resolve "Learner1D doesn't correctly set the interpolated loss when a point is added"
+* [!101](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/101): Resolve "Learner1D fails when function returns a list instead of a numpy.array"
+* [!102](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/102): introduce 'runner.retries' and 'runner.raise_if_retries_exceeded'
+* [!103](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/103): 2D: rename `learner._interp` to `learner.pending_points` as in other learners
+* [!104](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/104): Make the AverageLearner only return new points ...
+* [!105](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/105): move specific tests for a particular learner to separate files
+* [!107](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/107): Introduce `tell_pending` which replaces `tell(x, None)`
+* [!112](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/112): Resolve "LearnerND fails for BalancingLearner test"
+* [!113](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/113): Resolve "(BalancingLearner) loss is cached incorrectly"
+* [!114](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/114): update release guide to add a `dev` tag on top of regular tags
+* [!115](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/115): Resolve "How should learners handle data that is outside of the domain"
+* [!116](https://gitlab.kwant-project.org/qt/adaptive/merge_requests/116): 2D: fix #109
+
+
+New features
+* add `learner.tell_pending` which replaces `learner.tell(x, None)`
+* add error-handling with `runner.retries` and `runner.raise_if_retries_exceeded`
+* make `learner.pending_points` and `runner.pending_points` public API
+* rename `learner.ask(n, add_data)` -> `learner.ask(n, tell_pending)`
+* added the `overhead` method to the `BlockingRunner`
+
+
+## version 0.5.0
+
+* Introduce `LearnerND` (beta)
+* Add `BalancingLearner.from_product` (see `learner.ipynb` or doc-string for useage example)
+* `runner.live_info()` now shows the learner's efficiency
+* `runner.task.print_stack()` now displays the full traceback
+* Introduced `learner.tell_many` instead of having `learner.tell` figure out whether multiple points are added ([#59](https://gitlab.kwant-project.org/qt/adaptive/issues/59))
+* Fixed a [bug](https://gitlab.kwant-project.org/qt/adaptive/issues/61) that occured when a `Learner1D` had extremely narrow features
+
+And more bugs, see https://github.com/python-adaptive/adaptive/compare/v0.4.1...v0.5.0
+
+
+## version 0.4.0
+
+Rename `choose_points` -> `ask` and `add_point`, `add_data` -> `tell` and
+* several small bug fixes
+* add Jorn Hoofwijk as an author
+
+
+## version 0.2.0
+
+Release with correct licensing information
+Previously Christoph Groth was not properly attributed for his contributions.
+
+This release also contains a bugfix for Windows users
+
+
+## version 0.1.0
+
+Initial interface and algorithm proposal