Prune git dir (#1043)

* Handle wrapper in pipeline viz (#753)

* handle wrapper in viz

* fix pipeline viz in notebooks

* remove re import

Former-commit-id: f21972fc6e7aad11014123b20e427a68ca38b6c3

* docs: update DDM docs (#743)



Former-commit-id: b69f066d6780ddf6310cc5033a265fd3dfb7bf21

* Implement one-class SVM (#764)

* add l2 regularisation tests

* add inverse-scaling tests

* no intercept test

* hinge loss test

* split linear models across files

* wip

* implement ConstantThresholder

* implement QuantileThresholder

* implement one-class SVM

* format

* add threshold unit test params

* handle numpy arrays in huber loss

* fix isoup test

* Saulo review

Former-commit-id: 09a24d35c1f548239c54c1244973241bfe5c4edc

* Remove prejudiced terms (#766)

* remove blacklist and whitelist names

* rename master to main

* prejudiced, not controversial

Former-commit-id: 5e1a1fa85254f562ef8fe7006386255f2318a12c

* Use augmented operators (#767)

* use augmented operators

* black

Former-commit-id: 9a0cbcdcf66aeba8c954c8af34903ef5d3d739ab

* fix unit tests badge

Former-commit-id: 0581be76ebdd923b2780e25be5dd9c3c5892e2a4

* Fix OCC flaky test (#769)

* seed params

* flake

Former-commit-id: c8dd4ad5fa8f63f1441bcf2adef3d0981b9e90be

* Fix PyTorchClassifier variable classes (#775)

* FIX issue when last layer of NN is trainable

* REFORMAT

Former-commit-id: 5b814750591c3704cef7a78d651394698e20aa2a

* Refactor tests of Naive Bayes.


Former-commit-id: 70fb830ce3f7bc908c942ba69cd307b9b9a7ba8b

* Update naive bayes tests


Former-commit-id: 2e0b25a2ef2d2ba9ec080cf86a491f7465433b18

* Calculate interactions between feature groups (#777)

* rename PolynomialExtender to PolynomialFeatures

* implement TransformerProduct

* flake

* revert renaming

* handle pipelines in product

* flake

Former-commit-id: be89e15fcce6e4fae68f629f36f719a5e412bdef

* Refactor pipeline order of resolution (#768)

* make pipeline inherit from OrderedDict

* add last_step property

* switch boston dataset to diabetes

* typo

* make unsupervised estimators learn during learn_one

* fix tests

* Add log_method_calls context manager

* add warm_up_mode context manager

* handle *_many methods

* fix pipeline tests

* flake8

* revert ordereddict inheritance

* make WARM_UP property private

* add pure_inference_model context manager

Former-commit-id: 6d5d8ec89eb1f54ba1275452b154df1de38dfdc1

* Add CMU keystroke dataset (#781)

* add snakes

* add keystroke dataset

Former-commit-id: 236cade8d46e128629a4dc12b69e0e2426461506

* Fix typo in matrix factorization for recommender systems part 1 (#782)

Signed-off-by: Masato Naka <masatonaka1989@gmail.com>
Former-commit-id: 82d48a6256496c557bb5ae0f95e0db19fa32c03a

* fix typos



Former-commit-id: 8192c3194b6bea753f5659edf983e4c2f9c398bd

* Add debug_one to FM regressor (#784)

* Add debug_one to FM regressor

* add doctest

Former-commit-id: 34cd53e7eef923332523cb14fb84aa251000166d

* Add OneHotListEncoder to the preprocessing `one_hot` module (#786)

* Initial commit

* Integrated OneHotListEncoder into OneHotEncoder class by checking for an interable. Also added some Doctests

* Removed old reference to OneHotListEncoder

* Removed unused Iterable import and changed Iterable references to list/set. Removed tuple as iterable option.

* Typo fix

Former-commit-id: bd007b682b41f0805cbecc9573b3f5ffac3e4892

* edit release notes



Former-commit-id: cfaf41700305d76c11802ed2eff2f5a7f65d4059

* Finish bandit refactoring (#746)

* typo

* add ranking property to bandit

* rename selection to model_selection

* implement check_regression_selector_performance

* make _unit_test_params a generator

* remove redundant tests

* add burn_in parameter

* finish epsilon greedy docstring

* unit test StackingClassifier

* implement ucb regressor

* lil weng references

* black

* fix docs

* black

* flake8

* format

* docstring

* add rebase command

* WrapperMixin becomes Wrapper

* quick viz fix

* fix wrapping

* fix unit test params yielding

* fix bandit ranking

* scale targets in ucb

* pre-commit

Former-commit-id: 70ac826e9677bc50085e6a64e8ecd735a4654a98

* bump

Former-commit-id: 15a939c807a01888f16ef497bba55c020afde987

* fix notebooks

Former-commit-id: 5fbb236858dea70037f42f9893c71be090756028

* fix pred_clipper tests

Former-commit-id: ab3927fc2e1afeabdf542feb1d7a4dd686e3a9e1

* pre-commit

Former-commit-id: 7637571c8584e37395bf5a72d25abba67308d666

* fix matrix facto imports

Former-commit-id: 036ffd915db719dcc54d2456b9f31b74892142ac

* refactor cibuildwheel



Former-commit-id: ab2a584f2e0575a2b3784367ce02a56000793f8e

* Update CONTRIBUTING.md

Former-commit-id: 6d58b4f318158ece464f5d461de701928e158dcc

* [HOFM] Simplify the latent scalar calculation with numpy.multiply (#788)



Former-commit-id: 8f2a9c38f3cf72f7e9f8c29e8468717c6de0ace2

* Update pypi.yml

Former-commit-id: 27f0d4bd696f114298799698c0ac4f0ca1be2e72

* Add debug_one method to all FMRegressors (#787)

* add debug_one to ffm regressor

* add debug_one to fwfm regressor

* add debug_one to hofm regressor

* [refactor] introduce common private methods for interaction calculation in facto

* use the common private methods in debug_one

* move debug_one to base

* Update docstring

* Fixed docstring and removed unnecessary brackets in sum

* Update unreleased

* Update unreleased.md

Former-commit-id: d6d2f6835723117b0506f34ee9bd8e85420b85d0

* docs: Enable to render equations for instant loading (#792)



Former-commit-id: fd0a115a1b65bf0f1a5086609cd8c952b171be64

* Refactor anomaly detection (#797)

* Change progressive_validation and make AnomalyDetectors working with ROCAUC

* REMOVE wrong type hint

* REMOVE predict_proba_one from AnomalyDetector

* Remove unused import

* Refactor work_with method
Add documentation

Former-commit-id: 5fbe7c99739056b9637051aa4ac3043fc6d68b84

* Remove Lagger and TargetLagger (#799)



Former-commit-id: 2f67236e441e5bd47e49cf0c32bbdf30834c7e67

* Add state property to Agg and TargetAgg (#804)



Former-commit-id: cd3ab91a8082dc102cf4789cab92bd37827fd553

* Refactor the reco module (#807)

* recommenders work

* a lot of refactoring

* remove bar chart

* remove perfplot

* make existing recommenders inherit from base.Regressor

* pre-commit

* don't use star imports

* flake8

* update pre-commit

* add seed comments

* typo

* add type hint

* add Recommended docstring

* add __all__ to utils.random

* isort

Former-commit-id: 3405ed7b0891c98622d3da5ddcdc31ef05a2ea92

* Update bug_report.md

Former-commit-id: 2b0943078459bf10797041884e44b8560d538877

* Upgrade version of numpy to avoid ImportError (#816)



Former-commit-id: 9d4d0eb4fe9b99b2ffbff87a9c6d5846d8a2a19a

* remove surprise


Former-commit-id: b64f14fc18b3dc94ddc490dace3ba5b493bc7406

* Refactored feature renaming (#821)



Former-commit-id: 3732f700da72642afe54095d4b252b05c5018c7d

* Reco ranking (#832)

* expose forecasting metrics

* rename recommender to ranker

* Update content-personalization.ipynb

* activate ranker tests

* Update label_combination_hoeffding_tree.py

* Update label_combination_hoeffding_tree.py

Former-commit-id: d7800b8d8d1ab5d7ccec108d5ad9d886f573df65

* Feature/mini batch support (#827)

* add mini-batch functionality to Select

* add draft of one-hot encoding mini-batch functionality

* add tests (somehow not idempotent with pytest cli in the respective folder)

* add static method decorator

* removed not too relevant comments

* add another test which fails

* add mini-batching with failing test

* fix codestyle

* swap pandas for numpy in some places

* fix codestyle

* add small but important memory opt

* fix doctest

* fix tests

* fix description

* comments

* fix docs

* change example

* formatting

* refactoring

* introduce mini-batch transformer

* Update unreleased.md

* typos

* code quality

* fix tests

Co-authored-by: Max Halford <maxhalford25@gmail.com>
Former-commit-id: b229fadf2809de111a17057b78ead490377a066a

* Refactor multi-output metrics (#833)

* remove _performance_evaluator.py

* remove PairConfusionMatrix

* refactor metrics

* implement jaccard

* Delete progress.log

* style

Former-commit-id: fcb4ad015e564b6df92055dab4942aa9addd0a05

* Fix typo: incremental decision tree user guide (#835)



Former-commit-id: 0125d4b72ac1de3d3cef434c37fe4864e43f9e01

* Create special module (#837)

* Create special module

* special, not utils

* don't inherit from deque

* style

* fix tests

Former-commit-id: a517ae65ff6e5476ae473a769039780e0fdcb31c

* Batch update for Mean/Var/Cov (#838)

* simplify __add__ and __sub__

* implement Mean.update_many

* implement Var.update_many

* don't inherit from SortedWindow

* implement test_add_bivariate

* remove tests, they're already there

* typo

* implement Cov.update_many

* flake

* Update test_var.py

* unweighted

* types

* Update cov.py

* Update iqr.py

* Update unreleased.md

* rename special to misc

Former-commit-id: be2a9047b22d35d51408dcb862c7574972bab2be

* Implement covariance matrix (#839)

* init

* Implement CovMatrix

* typos

* wrong import

* fix confusion

* Update cov.py

* Update confusion.py

* fix confusion

* Update cov.py

Former-commit-id: ec66af84c10b555fc003b819946da1b80aa91850

* 0.10.0


Former-commit-id: d869d8c747fc4a2b9f93ed6da4c5da993ae5a2b2

* handle extra args in progressive validation


Former-commit-id: 56eff4e036e5729408408669bc9535a737803639

* Update release-docs.yml


Former-commit-id: 2149490a1573b54b54096641678d589dbce75e4a

* hack predclipper to handle args


Former-commit-id: 7e337ec65b020bfbf122db56e63f058381533e23

* Update pred_clipper.py


Former-commit-id: 6e6a8db46210b87d71e4d988c328cb57362c1b52

* Update qa.py


Former-commit-id: 12292ec9c9e40354358690b2c2f2733af9d3bb5d

* kwargs, not args


Former-commit-id: 441cc5cfdc3a8bc82faabe15e74a818c3cd65ece

* Update qa.py


Former-commit-id: 5e4a8ea0c63c8236dcee3385cfe2e69769ec6619

* fix qa


Former-commit-id: c96424c6ecf492e61aa9d194acef5400f0e6b7ed

* Update progressive_validation.py


Former-commit-id: b1a073b4117246c235b76f8357abf52ac8d34144

* allow pipeline to accept kwargs


Former-commit-id: 448c82cab33d0b327b0bed2e8e1ca811c86b8b71

* Update qa.py


Former-commit-id: b55df3e9e844aa02733f82ab80e3037b65e36406

* 0.10.1


Former-commit-id: 222021fc3f9cfa5c287bac8a6e08f9ec1abd2cad

* Update CONTRIBUTING.md


Former-commit-id: a5ad5222c980b9667c9841b46cba2c7c9a2e7350

* Update release-docs.yml


Former-commit-id: ef7a10838814c85fe843bbdb19d867a7e993063c

* Update quantile-regression-uncertainty.ipynb


Former-commit-id: eee8ab87ebadd88292a65301c687d1ff60908dbf

* Improve covariance matrix (#840)

* Add unit test for covariance matrix

* Update test_cov_matrix.py

* Update setup.py

* Update cov_matrix.py

* fix some warnings

* Update test_cov_matrix.py

* Update test_cov_matrix.py

Former-commit-id: 6916208f141e02371c32433afe840d940beb1ab1

* Update cov_matrix.py


Former-commit-id: 965d311a41003bd3e6aca12675903d0e2193f64c

* Create FUNDING.yml

Former-commit-id: 4a4881e434ea2529abe12249d97d4fee7d67cb1b

* refactoring code with chain comparison (#842)

* refactoring code with chain comparison

refactoring code with chain comparison which is more pythonic, concise, readable and efficient; how do think this change which has practical value？

* modify syntax error

* fix syntax error

Former-commit-id: ec1cf318310add301afe12160cebc66eaebcec2c

* Update k_means.py


Former-commit-id: 145efc8aad6ea54944d45966675a37459f93cb50

* fix kmeans


Former-commit-id: 89120f216be893714e299c21d51cb744964c865f

* Update k_means.py


Former-commit-id: f568f4e8f33b57f4f99ce15dfb5f011303de9ea0

* simplify isort config


Former-commit-id: d8aeeffa9ec114f0657d629581c234e3a538c6b9

* move clustering metrics


Former-commit-id: 68aa41c32543a77f5aa53895c0c894e63f9ca712

* Update iter_csv.py


Former-commit-id: 7085ae1625095a5104d6e2543bdafb2b445da290

* Update iter_csv.py


Former-commit-id: 4b1eb11caf752ad21b84c4539ac039d5d7885809

* Hotfix dotvecmat (#865)

* Kalman filter optimizer

* Update kalman_filter.py

* General losses

* Update math.py

* Change API

Change API

* Create mini_batch.py

* Update base.py

* Revert Kalman fitler

* Revert "Update math.py"

This reverts commit 19e66d5417c316445513568a8d10d6032f110390.

* Revert Kalman filter

* Adapt to current river master

* Correct index of dotvecmat

* Write tests for dotvecmat with numpy as reference

* Correct doctest

* Reformatting test

* Incorporate reviewer comments

Co-authored-by: Leandro Von Krannichfeldt <vleandro@student.ethz.ch>
Former-commit-id: 993328ee58fb267facda47ab8eb14c365b41c960

* Added parameters to EDDM drift detector (#870)

* Update eddm.py

* Update eddm.py

* Removed eddm_ prefix

* Update eddm.py

Former-commit-id: 7b219039fa420585a2a2ad0ca87472017b1693e0

* Make AMRules show a default rule consequent (#873)

* now the rules' repr show the target mean output
* update release notes
* reduce decimal places

Former-commit-id: a18e8ccb8694943940772823a81ebae9785a7391

* Update on DBSTREAM (#872)

Updates on DBSTREAM to handle empty set of labels, cases when features are not fully overlapping and OverflowError.

Co-authored-by: vsoch <vsoch@users.noreply.github.com>
Co-authored-by: Hoang Anh NGO <50743576+hoanganhngo610@users.noreply.github.com>
Former-commit-id: 395a61ab759938bbaea035a6f944053f509965d5

* implement TimeRolling (#876)



Former-commit-id: 1cf8141955b6adc6f65f0500305d863d91f833e1

* Fix bug in debug_one (rules.AMRules) (#877)



Former-commit-id: 93dd73dfeef8083e11a908c0f58be69148c07bbe

* Add Chebyshev's inequality-based under- and over-sampling methods (#878)



Former-commit-id: d8f16a5c51d6c935b75c17a09e79c087212449f3

* [AMRules] revert default rule consequent (#883)



Former-commit-id: 599ce35733fc94f5d10cb3ec571d99a3fc7c195f

* [AMRules-debug_one] Fix number of decimal places and update test coverage


Former-commit-id: 6b41c499c9fddd7955f375639c42878f820b6e90

* Move black, flake8, and isort to dev extra env in setup.py (#885)

* Move black, flake8, and isort deps to dev env in setup.py

Move pre-commit dependencies (black, flake8, isort) to dev extra
environment in setup.py.

* Run 'pre-commit run --all-files'

Black update caused the following change `a ** b` -> `a**b`

Former-commit-id: df431b2411f7f357b4fa8d2839a27ea14a4aa8ce

* Docstring fix for DenStream (#887)

* Update example in docstring to align with the parameters' feasible ranges.
* Check for the value of beta to be within the range (0,1].
* Update docstring for beta, mu and reason for the implementation of n_samples_init.


Former-commit-id: 4e713f99c70a472fecedf9aeb9426a590e186663

* [DBSTREAM] Allow microcluster centroids' updates to have different terms (#888)

Signed-off-by: vsoch <vsoch@users.noreply.github.com>

Co-authored-by: vsoch <vsoch@users.noreply.github.com>
Former-commit-id: 9ea7707636e825a374cf053d149e3946c746ce2d

* apply l2 reg in linear models


Former-commit-id: f399ccd53014be8b4b31d56d2a5a21b3e985ba20

* Update test_glm.py


Former-commit-id: 8f289a39d9bb6dd04d3dd1462ce73c6c5c3b7f59

* Update test_glm.py


Former-commit-id: 981e18b0c0219a894df7cacbda64814acb80c8dc

* Update Chebyshev samplers to match the original paper and code (#894)



Former-commit-id: 06dcac3cae2f07fbc194dff17bcfee8ec8298810

* [DenStream] Handle the case when p is not merged and o-micro-cluster is empty (#898)



Former-commit-id: 0602eee8b311aa9450963ba9c07b55f83dc530c8

* Implement iter_progressive_val_score (#901)

* return metric rather than metric.get

* implement iter_progressive_val_score

* black and isort

* Update progressive_validation.py

Former-commit-id: ca2aa8ca0b9484b23aeb177adaeb7eabe6441723

* Update mlp.py


Former-commit-id: 066d9ead99e17e40ffc992c03c853317f4a40bad

* Update mlp.py


Former-commit-id: c0c88d38b96c99de3e3646005c3ded1bfee47e47

* Feature/l1 implementation (#893)



Former-commit-id: 332e134ed50bad95775af367a49d00509844dce6

* add entry on l1 (#909)



Former-commit-id: 082d5eb15ab0bc24def4d9a7aead1b6d5d5ef7c7

* refactoring neighbors models to use simple collections queue (#895)



Former-commit-id: 9ce947ebfc012ec7059de0a09c765b2da7fc1d25

* Update KSWIN and remove unnecessary numpy use in drift detectors (#917)

* remove numpy usage in favor of collections.deque. Suppress convergence warning
* release notes
* fix tests

Former-commit-id: 361d7dfc4081be8727361bc0e6c0a03092609312

* Streamline SRP{Classifier,Regressor}, remove unneeded numpy usage and add PeriodicTrigger (#918)



Former-commit-id: 78be35004770b80d53387afdcc619a456596e9e6

* Remove unnecessary numpy usage in ARF{C,R} and HAT{C,R} (#919)



Former-commit-id: c419638334eb8b826ccd067f4d148ea4a317b0ee

* fix ARF code style


Former-commit-id: 6c2ff3186884beec5f95d171e3836aa7d78157fc

* fix ARF tests


Former-commit-id: f8af5db832c36cd6d5b9f96f71cc45cb2706edb8

* fix issue #925 (#926)



Former-commit-id: 9780ea4425f033c7d5c00d694db90f5bba978e3d

* Prettify docs (#920)

* Update unreleased.md

* update font and nav links

* improve docs

* Update why-use-river.md

* Update Makefile

* Update Makefile

* wip

* finished for now

* remove some md files

* remove more md files

* Update .gitignore

Former-commit-id: 79a20cfc6d66db0b2166015050ca21bb96f072e4

* Create CITATION.bib (#932)



Former-commit-id: 7743fa118cf63ed89b4f591da94b62a4a114bc4c

* Polish README (#933)



Former-commit-id: 7d1f281b8104e18b00dbe445b736138513717bfd

* Add base submodules (#934)

* Add base submodules

* Update .pre-commit-config.yaml

* pre-commit

* lint

* add __all__ to base module for docs

* lint

* Update base.py

* Update base.py

* Update base.py

* Update base.py

* Update base.py

* fix docstrings

* improve docstrings

* Update base.py

Former-commit-id: 93497bba53d11d21e862acfd656b3fba7cf05c9b

* Introduce anomaly filters (#936)

* add waterflow dataset

* Update water-flow.csv

* make thresholders return booleans

* implement time_series.iter_evaluate

* Update holt_winters.py

* Update holt_winters.py

* implement gaussian anomaly detector

* remove thresholding methods

* wip

* allow Metrics to be cloned

* add constant filter

* lint

* Update test_evaluate.py

* Update test_estimators.py

* fix tests

* Add protect_anomaly_detector parameter

Former-commit-id: ec32c5f4fb8267c03c5959f74d585e5418583a5e

* Update unreleased.md

Former-commit-id: 836183de3b6132b8d306fdd7b0cac613306a1156

* fix anomaly tests


Former-commit-id: ea236b6696e7b6141b56fb98b76376928076b2bc

* black


Former-commit-id: 4e8e257b4f03db93d37fc3345c3c1da5fb1e360d

* Update unreleased.md


Former-commit-id: 5ddb3875de606eb78dbe2c55248aec24a00b2d1b

* [ARF] Make the wrapper and base models share the same rng (#927)

* Make the wrapper and base models share the same rng

* fix issue #925 (#926)

* cleanup

* housekeeping

* minor change

* use ndimage.shift namespace

* adapt to latest scikit-learn changes

* Update test_glm.py

* lint

* Update adaptive_random_forest.py

* test

* sanity check

* style

* rollback test

Co-authored-by: Leo <haliatus@outlook.com>
Co-authored-by: Max Halford <maxhalford25@gmail.com>
Former-commit-id: 3b2003767a9aef6a22e6f73707cc825438291743

* Python 3.8


Former-commit-id: 73279f0fb95b550ab32aea81decf8eb691b7a9c5

* ensure class order in HT


Former-commit-id: ea4dac12fb74d0485e7e2255d90bf6e1fce84935

* update ci


Former-commit-id: 73e5a49c5dc259818a283d493a8d0b2688384c5a

* 0.11.0


Former-commit-id: c91bc4583f299432e7e7395ae7bc5a14d50ccb4d

* 0.11 tidy up


Former-commit-id: f00a707f9e23e13c6efc0b6e8508a82f87a9f139

* Update branch-tests.yml


Former-commit-id: ab7df4ff20f44f7157aac614458409c457b7227a

* fix mypy and memory measurements errors in HTs


Former-commit-id: ccd7aadf5996a3d4031a2d9676b05f8b7a7e789d

* Fixed links not working in next-steps.md (#940)

* Fixed links in next-steps.md

* Fixed links

Former-commit-id: 44efb7da77a0fdfbcb6ca06c0a048c34a3a8a30e

* Update water_flow.py (#939)

* Update water_flow.py

* Update water_flow.py

Former-commit-id: 60872844e6052b5ef20e4075aea30f9031377136

* Create codeball.yml

Former-commit-id: 5214de18d958888e1756c2569d6e53d32171acbe

* Fix for protect_anomaly_detector logic (#944)

The anomaly_dector should be updated if we are not protecting the detector or if it is not an anomaly.  I.e. the only time it should not be updated is when we are protecting the detector and it is an anomaly.

Former-commit-id: 65dc22b79c878860dc9780f9b26d8a9da56ba9fa

* Use evaluate.iter_progressive_val_score in HT recipe (#951)



Former-commit-id: f0bfde903fdf16ff4b4a0c5e2019e963a29c27a8

* Tracks benchmark (#953)

* wip

* lint

* wip

* wip

* show/hide dataset details

* add wrapped generators

* wip

* wip before the beach

* make metrics generic

* fix conflicts

* add first results

* improve benchmarks

* fix time

* wip

* wip

* add benchmarks to docs

* new results

* add shelve

* Update voting.py

* new benchmark results added

* housekeeping

* more housekeeping

* json needed due to different computer chips + shelve

* add models and datasets details

* refactor run

* Update run.py

* should work

* Update tracks.py

Co-authored-by: smastelini <saulomastelini@gmail.com>
Former-commit-id: a4b17e8880815e4b1843b87edfaa296d21c6a82e

* 0.11.1


Former-commit-id: 77b35622302b3e051cbc423ece1dcb7ae86daa73

* Update CONTRIBUTING.md


Former-commit-id: 2a609eedc39992e2c0e9955ce5526cd7c5200925

* Update render.py


Former-commit-id: d1f75c46c12691ddd9490837600bbb0430484c90

* Log reg benchmarks (#955)

* Update run.py

* Update CONTRIBUTING.md

* remove old log reg benchmarks

Former-commit-id: b37d16d8a88958f5eb412e51f9db17ac35f24b60

* Fix errors, refactor and make the neighbors module future-proof (#956)

* refactoring

* makes sure KNN models are pickable

* fix mypy errors

* Update base.py

* Update river/neighbors/base.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* Update river/neighbors/base.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* Update knn_classifier.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>
Former-commit-id: 0d1665c3c897754e4b4ea418252720537237b362

* Enable mypy (#954)

* Update setup.py

* mypy river/base

* Update transformer.py

* mypy: rules

* remove kwargs from base classes

* lint

* Update amrules.py

* mypy river/utils

* mypy river/linear_model

* lint

* mypy river/misc

* Update cov_matrix.py

* Add typing annotation for update and revert

* Add typing on window attribute

* Fix mypy in facto subpackage

* mypy river/stream

* Add typing on window attribute

* Convert Distribution as Protocol

* Remove the inheritance for TimeRolling

* Remove unused import numbers

* Create CODEOWNERS

* mypy river/evaluate

* Update qa.py

* Add explicit typing for xs argument

* Explicit typing for TimeSeries and mypy

* Fix mypy in reco subpackage

* Update qa.py

* Update qa.py

* Update qa.py

* Add typing for metrics attribute for HorizonMetric

* Add annotations for the attributes y_trues and errors for SNARIMAX

* Fixing import after isort

* Update qa.py

* Update qa.py

* Add import annotations for support list for 3.8

* isort fix import

* Fix annotation for mlp

* fix isort

* Annotation consistency

* Fix typing CI

* Fix CI

* Update setup.py

* mypy river/base

* Update transformer.py

* mypy: rules

* remove kwargs from base classes

* lint

* Update amrules.py

* mypy river/utils

* mypy river/linear_model

* lint

* mypy river/misc

* Update cov_matrix.py

* Add typing annotation for update and revert

* Add typing on window attribute

* Add typing on window attribute

* Convert Distribution as Protocol

* Remove the inheritance for TimeRolling

* Fix mypy in facto subpackage

* mypy river/stream

* Remove unused import numbers

* Add explicit typing for xs argument

* Explicit typing for TimeSeries and mypy

* Add typing for metrics attribute for HorizonMetric

* Add annotations for the attributes y_trues and errors for SNARIMAX

* Fix mypy in reco subpackage

* Create CODEOWNERS

* mypy river/evaluate

* Update qa.py

* Update qa.py

* Update qa.py

* Update qa.py

* Fixing import after isort

* Add import annotations for support list for 3.8

* isort fix import

* Fix annotation for mlp

* Update qa.py

* Update qa.py

* fix isort

* Annotation consistency

* Fix typing CI

* Fix CI

* Update branch-docs.yml

* Update release-docs.yml

* mypy river/feature_selection

* FIx distribution

* fix mypy errors in the tree module

* mypy multiclass

* fix mypy errors in drift

* Update hst.py

* Update hst.py

* Fix mypy in datasets subpackage

* fix mypy errors in ensemble

* fix elementtree

* Update viz.py

* remove codecov

* Fix datasets subpkg tests

* mypy cluster

* mypy model_selection

* Mypy compat module

* Add __future__ annotation to use 3.9 type

* mypy

* mypy

* Compose mypy

* mypy feature_extraction

* Update README.md

* mypy river/metrics

* lint

* Add mypy check in .pre-commit-config.yaml

* Add MiniBatchSupervisedTransformer

* Mypy union

* typo fix

* Update qa.py

* Optim mypy

* fix pre commit

* delete unused import

* mypy checks

* Update qa.py

* fix tuple is not subscrible in CI

Co-authored-by: smastelini <saulomastelini@gmail.com>
Co-authored-by: Adil Zouitine <adilzouitinegm@gmail.com>
Co-authored-by: Geoffrey Bolmier <geoffrey.bolmier@volvocars.com>
Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>
Former-commit-id: b460b5e312f6f9de60e4d2652af31080b55b3316

* Update pytorch.py


Former-commit-id: f6d4f009ef31b88d4e22f7d82f41a56fafd01c49

* Update README.md

Former-commit-id: 30eadc7ccb2c3d2fc4a2805075bafae616b53c90

* Update pytorch.py


Former-commit-id: 9cbc98eca3f6b6fabedf9b2aaa549275262168f7

* Update filter.py


Former-commit-id: 18772e28e3a08db8f5d1cd0e0e12a4165c36b018

* use python 3.9 in dev-docs job


Former-commit-id: d4896b423b4010e54d3ae0578f0fcca48cb16b6c

* Update render.py


Former-commit-id: 419736b4461c9f7774f2aefe07f044c9820c93ae

* move compat.pytorch to river-extra


Former-commit-id: 19c4840c18ae1996c6c17d280bd65a2106abc06c

* Update Makefile


Former-commit-id: 3c70d1cf9d52814b438cc5a986cec657aea9086a

* Update README.md


Former-commit-id: e8088e5affd931a6f0b5d977f045f066fa576030

* Update setup.py


Former-commit-id: b443e6f605814724f49a4d6a14114866251b65a2

* Update setup.py


Former-commit-id: 18c1aff91e4525a876f002ab26eed947491115a2

* Update setup.py


Former-commit-id: b36ddd4973e0d61920a00fdd44e031e0103e88e6

* Update run.py


Former-commit-id: acb1baebe99b41afe90e79eee1277aea92840cd8

* add sklearn, torch, and vowpal wabbit to the benchmarks


Former-commit-id: 3b8d331f0d9e244545775cf5988faa4b7919bd69

* make sure all tracks are exported to json


Former-commit-id: a089af9f13328cfcaead0b7e9f7946f3ddfd740c

* Update render.py


Former-commit-id: 43b1b4e2215bf815fa60a78ae86ce8fb33146db9

* Move root `__init__` imports to a new `api` module (#957)

* Move root __init__ imports to a new api module

* Fix import errors

* Fix import errors

* Fix mypy in qo_splitter.py

* Fix import errors

* Fix import errors

* Remove anomaly dependency from utils/inspect.py

* Fix imports in concept_drift_stream.py

* Complete isanomalydetector docstring in utils/inspect.py

* Make utils only dependent of base and compose of base and utils

* Import synth in datasets

* Remove optim dependency from metrics

* Fix context managers imports in tests

* Fix cross_entropy.py missing attribute

* Update changelog

* Apply suggestions from code review

Co-authored-by: Max Halford <maxhalford25@gmail.com>

Co-authored-by: Max Halford <maxhalford25@gmail.com>
Former-commit-id: dab3f0eb045f1b7b06645a008795f4d117df97ef

* Update run.py


Former-commit-id: 31f7e190d40b93dc207adc115ae0441bb26b9cb4

* Trim trailing whitespace in inspect.py (#958)



Former-commit-id: 91144750c9e60485ffaea7155cc00a6465e9b205

* Update CONTRIBUTING.md (#960)



Former-commit-id: 0f74ad1d6ce7183499453fb88d7b8ce61bb37e41

* Update logo.svg (#959)

* Update logo.svg

* Update README.md

* Update README.md

* Update logo.svg

* Update logo.svg

* Revert "Update logo.svg"

This reverts commit 35539a11344a2b4380134f82335a767d01c19dc9 [formerly beb261882cc006aa884f010aaac8efd5c884bd0a].

* Revert "Update logo.svg"

This reverts commit 5043528eff0067445ee801cadc5493411ea83d3e [formerly 6e138ccff6b05f6677ac6b684726827c32630d90].

* Update logo.svg

* Update logo.svg

* c'est top hein

* Update home.html

Former-commit-id: 363e699857d7232a2caf7162e94227fa7306bb96

* Update README.md

Former-commit-id: 4549c3e30760360d8841cd6d329f249e65169aed

* Update README.md

Former-commit-id: 1fc5950e10424f73a341d4722c0ae3c13e5898b5

* Add black and isort config, remove coverage (#961)

* Add black config

* Add a build-system section in pyproject.toml

* Revert "Add a build-system section in pyproject.toml"

This reverts commit 668d73cbd69403c1abd98651b0b5326655e08b0a.

* Debug setup.py CI failure

* Debug setup.py CI failure

* Debug setup.py CI failure

* Fix setup.py CI failure

* Move mypy config to pyproject.toml

* Remove coverage

* Move pytest config to pyproject.toml

* Fix mypy config

Former-commit-id: fa1453a94ed8fd3ca6379294cad086b7f67546ee

* Remove river_extra version constraint in setup.py (#962)



Former-commit-id: 91ee38b7d26687f00f3d47ca4c57eb0d235e3989

* Postpone pandas import in base module (#966)

* quote pandas types in base

* add typing.TYPE_CHECKING

* add typing import

* lint

Former-commit-id: 66de2ccedeae821e8cebf24c4026c344fb0099cc

* Optimized append sorted window (#968)

* Replace the linear search with a binary search in sorted window

* Precise comment

Former-commit-id: 0bd03b4ffa5ade1df351a7417edb0d4a2773c96a

* update release notes


Former-commit-id: 46f8e874e328d7f37e8461eb700a428fb98ab1a7

* aesthetical changes to the Hoeffding Trees' guide


Former-commit-id: 52f07f20601085302b87ecd1e5f89c38dbebfaf9

* Generic rolling (#967)

* make iter_array handle text

* add class

* remove RollingMean

* remove rolling stuff in proba

* remove rolling stuff in metrics module

* Update __init__.py

* Create test_rolling.py

* lint

* Update test_rolling.py

* add docstrings

* lint

* ignore rollable type issues for now

* fix tests

* add notebook

* Update rolling.py

* add docstring params

* fix rolling usage in anomaly

Former-commit-id: 6eea5688e4ed8b8269d6723c2748df58a3d8d61f

* Implement average forecast metric (#969)

* Update home.html

* Implement HorizonAggMetric

* black

* Update metrics.py

Former-commit-id: 8d4d62958e5b43357d11f086f9eb4ec8eb3eb83b

* Introduce mutable attributes (#964)

* allow accessing pipeline steps by position

* Add edit method

* Remove river_extra version constraint in setup.py (#962)

* black

* add check_edit_can_be_idempotent unit test for all estimators

* Update base.py

* Update common.py

* rename edit to mutate

* finish renaming

* rename edit to mutate in compose

* rename edit to mutate (thought I already did)

* Update .pages

* Update river/checks/common.py

Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>

* Update river/base/test_base.py

Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>

* Update cloning-and-mutating.ipynb

* rename check_edit_can_be_idempotent to check_mutate_can_be_idempotent

Co-authored-by: Geoffrey Bolmier <geoffrey.bolmier@volvocars.com>
Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>
Former-commit-id: ffbb7fd85f06455b688adcf0a4bbb6472c3ae4a0

* Update home.html

Former-commit-id: d70d4837d9293ccd2892ec6977db277ef39ad898

* Update home.html

Former-commit-id: befb7ede5048da38d05d4c3b32abd600cebeb106

* Update unreleased.md

Former-commit-id: 33b2987725ede1a7372a6519474fe923534ea48f

* Drift refactor (+ HAT refactor) (#963)

* change drift API

* update usage in the codebase

* update example

* move drift intro

* Update river/base/drift_detector.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* Make reset method private

- Ensure that every active detector calls the _reset method after a drift is encountered
- Update docs guideline

* ongoing refactoring of HAT algorithms

* lint

* add TODO notes

* lint

* unify rng usage

* fix HATC

* update comments

* fix drift tests

* update default leaf prediction strategy

* also update ensembles

* correct HATR test to switch between subtrees

* prepare to rename misleading parameters and standardize usage

* rename split_confidence to delta

* rename tie_threshold to tau

* rename switch_delta to switch_signif

* use full parameter name

* add release notes and update recipe

* redesign DDM

* improve ADWIN documentation

* default params in trees and rules

* update DDM test

* refactor EDDM

* refactor Page-Hinkley and add support to two-sided tests

* fix typo in PH docs

* update Chebyshev tests to account for the default parameters changed in AMRules

* refactor HDDM_A

* update getting started entry

* hddm_a docs

* refactor and fix bug in HDMM_W

* fix documentation

* update release notes

* release notes

* fix docs

* add detail in HDDM_A documentation

Co-authored-by: Max Halford <maxhalford25@gmail.com>
Former-commit-id: 13d9fb811c2fc98fed7d685af075bd9ec81809a0

* Update unit-tests.yml


Former-commit-id: a980b8d884b252a39d6a8242d6ee8749aa617298

* update benchmarks


Former-commit-id: eaa049ac2cf412543d6b99eefe7a6088bed74f8d

* Some light style change to docs (#973)

* use rich when executing notebooks for release docs

* use __repr__in pipeline viz instead of pprint

Former-commit-id: af749bb9eb1caeffc24a5d05571d2a25fe7f1197

* Update viz.py


Former-commit-id: c2469894b1379756a11ba18fa323610578b1405b

* take more samples in synth datasets comparison tests


Former-commit-id: 8ede49454540814fc70a10d890de262ed72e6268

* init related-projects.md (#975)



Former-commit-id: 2ea955437b759763d268bfee4cbb0140260e6637

* Fix SNARIMAX formula for seasonal features (#980)

* fix lag formula for seasonal components

* add snarimax unit tests for lag features creation

* update figures in snarimax example

* lint

* Update unreleased.md

Former-commit-id: a55b51b519858fdb1977b234a7095a3029d6e96f

* Protect AdaptiveStandardScaler against numerical inconsistencies (#981)



Former-commit-id: 220a3fa4f20973f7aac9e8b5210e4451e754f624

* fix heading (#983)



Former-commit-id: a039f917c9c55cba302d13136f1b406e26164d3d

* Update CODEOWNERS (#985)



Former-commit-id: 25988d6887e1b00da87a52cb87ceae3af7d599c6

* remove self.n (#986)



Former-commit-id: 92bb8995c94476bf42daaaf42e63c9a9b722ff3e

* Fix time series differencing (#988)

* add sympy to dev dependencies

* do some work on snarimax

Former-commit-id: 8ee84d1c13736404651586f3a0dd5958b21212cf

* Update unreleased.md


Former-commit-id: 827f95188195d0ce2764e9ec408429c50a1f7b69

* lint


Former-commit-id: 6f390b51b25f87989b463e11bd0460a9bf83e069

* ensure TargetAgg is treat as supervised (#991)



Former-commit-id: e2fb00cee288e6e92a94ff8c300f41f9caee35ca

* Some more SNARIMAX fixing 😁 (#992)

* add differencing unit test with figures

* add undiff with test

* fix undifferencing

Former-commit-id: 7e4bd2b9eecc655389676eb971d8ebc226320077

* Update test_snarimax.py


Former-commit-id: b38fa74e7705035bc76d51965c3609dde5d03dc1

* q instead p in snarimax error history


Former-commit-id: dad36d63da3dc66ec997586fdd7618a666ff4a20

* store differenced values in separate deque


Former-commit-id: 925b4d65ff250b64fc9bbc3b9b8c1764fd1f73f6

* fix memory management


Former-commit-id: 1c137096a1bd0357431b83954307dd26ce13a09b

* Refactor covariance matrix (#984)

* add utils.math to docstrings where it's missing

* init InvCovMatrix

* Update inv_cov_matrix.py

* simplify and improve stats.Cov update formula

* reimplement online covariance matrix

* implement update_many

* implement revert

* implement EmpiricalPrecision.update

* wip

* wip

* drop RollingCov

* drop RollingPearsonCorr

* wip

* add some weighted univariate tests for mean

* sanity commit

* lint

* remove precision matrix

* Update river/covariance/test_emp.py

Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>

* Update river/covariance/test_emp.py

Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>

* Update river/covariance/test_emp.py

Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>

* add an include_attributes parameter to Base.clone

* add include_attributes to inherited clone methods

* Update cov.py

* add n property to var and cov

* Update cov.py

* use _from_state instead of _iadd

* lint

* remove Mean._iadd

* Update test_stats.py

* Update test_base.py

Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>
Former-commit-id: 3a2d9aa3bf9fe29144899b2e912a721d001ee1e6

* Update pypi.yml (#1000)



Former-commit-id: 8cbbef3ca5b1c1b35258bd2e65a707d474b76913

* Re-add wheels for Python 3.7 (#1001)

* Update pypi.yml

* Update unreleased.md

Former-commit-id: 1d5aa061a06bb27e12ae3a428cdb2b35e716ffb4

* Precision matrix (#999)

* init

* implement mini-batch update to precision matrix

* add unit tests

* Update emp.py

* fix kurtosis and skew tests

Former-commit-id: 40c3190c9d05671ae4c2dc8b76c163ea53a45fb0

* Update setup.py (#1002)



Former-commit-id: b7ff5549f9137f26f13b32f2d7cb5a4f9010733a

* Update test_metrics.py


Former-commit-id: 86c071b2ef44f1f14f456d002ee025b5046eb243

* Update test_metrics.py


Former-commit-id: ae42fbec6cc44c30294ab3f54b8696a245b2e8b1

* Update test_metrics.py


Former-commit-id: 045ade751507868d575449808ddfae45dae38c43

* Add the newest tutorial on Online Clustering using River at KDD'22


Former-commit-id: da7de8b928ab7383599db31ade0642ceb155b3a9

* Twitter stream (#1007)

* nit

* implement TwitterLiveStream

* Update unreleased.md

* Update tweet_stream.py

* Update tweet_stream.py

* Update pyproject.toml

* Update tweet_stream.py

* Update tweet_stream.py

* Update tweet_stream.py

* Update tweet_stream.py

* Update tweet_stream.py

* Update tweet_stream.py

Former-commit-id: 36545b9b706a6064a9d2cb2db29769e5fd19684e

* add tests for py37 and py311


Former-commit-id: 7a4ad7ad6de1191718c73123322da431237bf72b

* adapt code for py37


Former-commit-id: db8c709ef9d7fe4c30443dd50d68c833f08f0cb6

* Update README.md


Former-commit-id: 01ab61db6d1b76d6bc225fa541db6fdb840c27ba

* revert


Former-commit-id: 3a42acf7282b52615997e4d9cc504fcbb9ee5ee5

* Update tweet_stream.py


Former-commit-id: 88ba5859ef74fc2a159158752e8045d706e8da75

* refactor docs requirements


Former-commit-id: d30bd629f2e407a95519777c67e5924ef22a5bc3

* Update setup.py


Former-commit-id: 34e8e685ca2f927ffd79a1141b80dc559710849b

* Update pypi.yml

Former-commit-id: a39ba7c0c4b66f6eced40cadd89d2eed673b425e

* Add stop_words parameter (#1008)

* introduce stop_words parameter

* Create test_vectorize.py

* Update unreleased.md

* lint

* Update test_vectorize.py

Former-commit-id: 0c922f847ab8b8f7f885c752e2f88bc50a7461d4

* Update streamkmeans.py


Former-commit-id: 1e9a9ddd408fcfa405b5e3b50cee786d3328717c

* Update unreleased.md

Former-commit-id: 0561a798aa8e69ee4e350e65067619e01580aa13

* Bayesian linear regression (#1009)

* wip

* Update emp.py

* implement BayesianLinearRegression

* lint

* Update newton.py

* fix tests

* Update math.py

* Update bayesian_lin_reg.py

Former-commit-id: 79aebbe53c04fd745651ca55a6ea22079f60d867

* Update bayesian_lin_reg.py

Former-commit-id: 7b219922de0285ef9ef344f4db6295985d9507c3

* Update bayesian_lin_reg.py

Former-commit-id: 36d754803d9e626436836071be57d2137bdcc96a

* Update bayesian_lin_reg.py

Former-commit-id: 62e832dbcf5076b335b642ef23d4dd13ebc315a6

* Update bayesian_lin_reg.py


Former-commit-id: 8b27fdb80ddaef188623d6036f023b5cad359569

* 0.12


Former-commit-id: 5c2df5169b236c3902b0157b33913b4f0aaa30dc

* Add more mutable properties (#1012)

* trees

* reco

* cluster

* rules

* rules and ensembles

* revert n_models

Former-commit-id: 6805de7baf998212fb4aa05f99d8b20626f66a51

* bumb


Former-commit-id: 433cc330b54436a9f034f15e3ce383328e2647cc

* Update release-docs.yml


Former-commit-id: 73da5f0aa9873c8d5b7c9a0cee4a8ed7eb4ab374

* Update release-docs.yml


Former-commit-id: f9e6ec05a3939f55cd34e8d3867eef2058c22b27

* Update release-docs.yml


Former-commit-id: 1338cfbdd7ab21a6dc0c499a14f4fef77ce55702

* Update release-docs.yml


Former-commit-id: 694554606e3d2c4043031c86dba53ceec3c3a2c0

* Update release-docs.yml


Former-commit-id: 25fc0bdd6a9c2b06fbf7be445678a01fa3f943e5

* Update bayesian_lin_reg.py


Former-commit-id: e8d1fdfb885c9578f79a7f395c7ab0df34412f93

* fix positional args handling in clone


Former-commit-id: a7abc2e5e70746f95a0a4a7dd051f7ab2f1f5f85

* lint


Former-commit-id: 5bd080b0d5961f13bf31dd8860c95a0dbd16d000

* disable logical test


Former-commit-id: a1004a422a4d0426bb6c3501689b1c9a74aaf9a6

* Update test_datasets.py


Former-commit-id: bfd442519fbd2b19dc14e3d1cd8004559fb5880f

* run notebooks


Former-commit-id: 07594635930ce417acfd89d894b09f997fb6f2f8

* Update __version__.py


Former-commit-id: 421edda7593cf45ff1df6ff9efd04f6e7a614643

* fix the art of using pipelines


Former-commit-id: 49c6a0b4f572ae2d3ee4c58997f82a4d7964f638

* add ci workflow to test notebooks weekly


Former-commit-id: 9740f0d495ac67e65070b7ee193d25d8f4d696d0

* fix mini-batching example


Former-commit-id: 341667cca425ddc143810129067fea4e69ded40d

* Check mutable attributes exist (#1013)

* add ci workflow to test notebooks weekly

* check mutable attributes exist

* Revert "add ci workflow to test notebooks weekly"

This reverts commit a369c94576d03c293fa1b6b1a764555283252ac6 [formerly 9740f0d495ac67e65070b7ee193d25d8f4d696d0].

* rename check_mutable_attributes_exist

* Update biased_mf.py

Former-commit-id: 48c9c070705f41f3e8e8212906dafdd65a3d7f0a

* Update release-docs.yml


Former-commit-id: 9764fd557b96d202e6c521397746af79adac2a2e

* done for the night


Former-commit-id: 7aae18f38ac5bfbb9bac00da67dcb39827f1efed

* Fix agg state doc (#1027)

* access transformer union steps by index

* move Agg.state example

* edit release note

Former-commit-id: 3348eb777fc1e7b14595af9778f03fadc74acdea

* Refactor code and use Welford's algorithm to CluStream (#1023)

* Refactor code and use Welford's algorithm to CluStream

* Move incremental K-Means on micro-clusters to learn_one, executed after each time_gap

* Revert "Move incremental K-Means on micro-clusters to learn_one, executed after each time_gap"

This reverts commit e61330ce9cb9cb3166b83f1cd3055dc0bc21d5b0.

* Update predict_one of CluStream with re_cluster binary argument

* Update CluStream after running pre-commit checks

* Revert to the use of time_gap (commit e61330ce9)

* Miscellaneous fixes suggested by Max from the previous comments in the PR

Former-commit-id: b1306656c2e6b4b8b69007c095450a72117d3e86

* Twitch chat messages stream (#1026)

Co-authored-by: Konstantin <miontq@gmail.com>
Former-commit-id: cb88c57c4eef2db60f4aee7a5137d47faadf4bd5

* Style nitpicks in TwitchChatStream (#1028)

* style nitpicks in TwitchChatStream

* isort

Former-commit-id: c30927fe4f0ac105b5e1b94c044ff0032a6942e7

* Update binary-classification.ipynb (#1032)



Former-commit-id: 8e1ad115c931889377c3a615c8795221a3977fe3

* Lossy counting (#1029)

* lossy counting

* add reference

* update release notes

* per review comments

* add dataset to actions

* fix typo

Former-commit-id: 108c900ad9286cc43aad66be9f194c61bd7b6d65

* Computationally intensive statistics in Rust (#1025)

* Work in progress first milestone

* WIP add binding

* Fix IQR get

* Wip stable

* Bump online statistics version

* fix iqr test

* Fix check

* Ewvar is pickleable

* All candidate statistics are pickleable

* Delete deadcode

* Setup properly

* add stub _rust_stats.pyi

* Add zip false in setup

* move rust module inside stats module

* isort

* debug pipy

* debug ci wheel with rust

* Disable windows for the moment

* fix build

* fix syntax

* try to fix before build for linux

* Fix mullinux wheel

* arm support

* fix arm

* fix

* fix

* fix

* fix

* Add 3.9 3.10 3.11

* Add rust support

* build wheel for tags instead of each commit

* fix rust support

* fix unit test

* Add rust

* Add equality test for pickle

* mypy fix

* Replace stat by the rust implementation

* Fix doctest

* Upgrade online-statistics

* Typo

* Removes unused modules

* add Cargo.lock in gitignore

* Add fix me comment on _mutable_attributes

* Add rust part

* Unit test only on main

* Change online statistics to watermill

* Add PeaktoPeak

* nitpicks

* Delete env_wheel.sh

* rename self.is_updated

* Update scale.py

* fix docs

* Update pypi.yml

* Update pypi.yml

Co-authored-by: Max Halford <maxhalford25@gmail.com>
Former-commit-id: 416b9aa87cdcd01ded3688dc7e7699f0ea96e579

* 0.13.0 with Adil


Former-commit-id: 3c220e198786dcb0b8ea9ca77a39e60b7bb86fb3

* fix dataset downloads


Former-commit-id: b1c6356c3acde4e259ec346b2f4b8978973f1152

* fix lossy rendering


Former-commit-id: d006e7b0c5f85afc10d7b55ea15f8a947c4b8305

* Handle AttributeError in predict_one of CluStream when there is no micro-cluster centers


Former-commit-id: e1ed23f73761b175b5545ba0dcd34da6beb6e932

* Add the Count-Min Sketch and introduce the new sketch module (#1034)

* add return self to lossy counting

* fix __repr__

* add CMS

* pre-commit

* fix typo

* Update river/stats/cms.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* move cms

* CMS now in misc

* update example

* update example

* trim extra space

* Update river/misc/cms.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* add base.Base as mother class

* move LossyCount

* return self lives...for now

* housekeeping

* update LossyCount

* update LossyCount example

* functional changes

* introducing river.collections

* housekeeping

* update docstring

* ensure all collections members inherit from base.Base

* nitpick

* Update river/collections/counter.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* merging

* update release notes

* let's keep some things in misc

* lint

* typo

* nitpick

Co-authored-by: Max Halford <maxhalford25@gmail.com>
Former-commit-id: e54ebac2ffb8149d738a7d5c42b3a1587b6ab297

* Update sdft.py

Former-commit-id: 983abdd29ad909fc6aab22de21517174eb6f2bd9

* Textclust (#1010)

* integrating textclust

* formatting

* update

* update

* update

* fix micro-cluster string

* update

* change dict to typing.Dict

* formatting

* fix style issues and implementing single-linkage clustering

* Remove the option to store document IDs in textclust

* update documentation

Former-commit-id: 1faa67d4d3716cc9f4f00ab0286701c5b5bd0c14

* Bloom filter to provide approximate set-like operations (#1041)

* add sketch.Set

* lint

* typo

* Update river/sketch/set.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* Update river/sketch/set.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* Update river/sketch/set.py

Co-authored-by: Max Halford <maxhalford25@gmail.com>

* redo release notes

* aesthetic

Co-authored-by: Max Halford <maxhalford25@gmail.com>
Former-commit-id: 9f02fe40511a301768a17f355d04c0e6792a03d9

Signed-off-by: Masato Naka <masatonaka1989@gmail.com>
Co-authored-by: tnwei <12769364+tnwei@users.noreply.github.com>
Co-authored-by: Cedric Kulbach <42866082+kulbachcedric@users.noreply.github.com>
Co-authored-by: raphaelsty <raphael.sourty@gmail.com>
Co-authored-by: Naka Masato <masatonaka1989@gmail.com>
Co-authored-by: NCC Group / Fox-IT Data Science <90677781+fox-ds@users.noreply.github.com>
Co-authored-by: Adil Zouitine <adilzouitinegm@gmail.com>
Co-authored-by: Damien Wojtowicz <damienwojtowicz@users.noreply.github.com>
Co-authored-by: Alexey C <54956904+ColdTeapot273K@users.noreply.github.com>
Co-authored-by: al <arisliang@users.noreply.github.com>
Co-authored-by: zzj <29055749+zjzh@users.noreply.github.com>
Co-authored-by: Leo-VK <61838765+Leo-VK@users.noreply.github.com>
Co-authored-by: Leandro Von Krannichfeldt <vleandro@student.ethz.ch>
Co-authored-by: TawabG <tghorbandi@gmail.com>
Co-authored-by: Saulo Martiello Mastelini <saulomastelini@gmail.com>
Co-authored-by: Vanessasaurus <814322+vsoch@users.noreply.github.com>
Co-authored-by: vsoch <vsoch@users.noreply.github.com>
Co-authored-by: Hoang Anh NGO <50743576+hoanganhngo610@users.noreply.github.com>
Co-authored-by: Geoffrey Bolmier <geoffrey.bolmier@volvocars.com>
Co-authored-by: Leo <haliatus@outlook.com>
Co-authored-by: rcvalerio <r.valerio@campus.fct.unl.pt>
Co-authored-by: dberardo-com <65530457+dberardo-com@users.noreply.github.com>
Co-authored-by: friskin <friskin@gmail.com>
Co-authored-by: Saulo Martiello Mastelini <mastelini@usp.br>
Co-authored-by: abonte <6319051+abonte@users.noreply.github.com>
Co-authored-by: Konstantin Ostrovskiy <necrolph@yahoo.com>
Co-authored-by: Konstantin <miontq@gmail.com>
Co-authored-by: Dennis <Dennis1989@users.noreply.github.com>