I assume you mean a small code snippet / doctest using each estimator.

This is one of the reasons I added automatic links back to pertinent
examples from the gallery when showing the compiled API reference, e.g.
http://scikit-learn.org/dev/modules/generated/sklearn.manifold.Isomap.html#examples-using-sklearn-manifold-isomap.
Unfortunately, there was no straightforward way to have these render where
the doctest examples do (and hence they were relegated to the end of the
page which is out of the numpydoc generation).

On 12 November 2014 23:06, Manoj Kumar [email protected] wrote:

Most of the docs in classes lack examples. It would be great to add one
or two examples similar to this,
http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.Lasso.html

—
Reply to this email directly or view it on GitHub
https://github.com/scikit-learn/scikit-learn/issues/3846.

Yes, I had meant probably one or two liners (for example) https://github.com/scikit-learn/scikit-learn/pull/3802/files#diff-1741ad6b05f1eb0fd71af8bad0e001c7R321 just to show the API.

This is one of the reasons I added automatic links back to pertinent examples from the gallery when showing the compiled API reference

This is awesome. I had not noticed it before!

This is awesome. I had not noticed it before!

Only in dev. And hidden away at the bottom of the page :(

Where else do you think it would be better?

Below the class description and above the parameters?

Anywhere before the method description would be fine. Method descriptions can take up a lot of vertical space on the API reference page, so someone is unlikely to scroll past them looking for examples.

Sorry for the late reply. I assume that it is not too easy to do? Else you would have done it yourself?

No, it's even hackier than the current approach!

On 19 November 2014 21:59, Manoj Kumar [email protected] wrote:

Sorry for the late reply. I assume that it is not too easy to do? Else
you would have done it yourself?

—
Reply to this email directly or view it on GitHub
https://github.com/scikit-learn/scikit-learn/issues/3846#issuecomment-63623385
.

How about simply adding examples to the missing classes' docstring?

These are the estimators ( 98 - 10 / 148 ) currently missing an Examples section :-

Done/WIP

ExtraTreesRegressor
BaggingClassifier
BaggingRegressor

4498 -

AdaBoostRegressor

GradientBoostingRegressor

Not needed

ExtraTreeClassifier  # Used only in ensembling
ExtraTreeRegressor # -do-

TODO

AffinityPropagation
AgglomerativeClustering
Binarizer
CheckingClassifier
CountVectorizer
DBSCAN
DPGMM
DictionaryLearning
ElasticNet
ElasticNetCV
EmpiricalCovariance
FactorAnalysis
FastICA
FeatureAgglomeration
GaussianRandomProjection
GenericUnivariateSelect
GraphLasso
GraphLassoCV
HashingVectorizer
Imputer
IncrementalPCA
Isomap
KMeans
KernelCenterer
KernelDensity
KernelPCA
LarsCV
LassoCV
LassoLarsCV
LedoitWolf
LinearRegression
LinearSVC
LinearSVR
LocallyLinearEmbedding
LogOddsEstimator
LogisticRegression
LogisticRegressionCV
MDS
MeanEstimator
MeanShift
MinCovDet
MinMaxScaler
MiniBatchDictionaryLearning
MiniBatchKMeans
MiniBatchSparsePCA
MultiTaskLassoCV
Normalizer
Nystroem
OAS
OneClassSVM
OrthogonalMatchingPursuit
OrthogonalMatchingPursuitCV
PLSSVD
PassiveAggressiveClassifier
PassiveAggressiveRegressor
PatchExtractor
Perceptron
PriorProbabilityEstimator
QuantileEstimator
RANSACRegressor
RBFSampler
RandomForestClassifier
RandomForestRegressor
RidgeCV
RidgeClassifier
RidgeClassifierCV
ScaledLogOddsEstimator
SelectFdr
SelectFpr
SelectFwe
SelectKBest
SelectPercentile
ShrunkCovariance
SkewedChi2Sampler
SparsePCA
SparseRandomProjection
SpectralBiclustering
SpectralClustering
SpectralCoclustering
SpectralEmbedding
StandardScaler
TfidfVectorizer
TheilSenRegressor
VBGMM
Ward
WardAgglomeration
ZeroEstimator

Of course, but don't you think that is slightly on the tedious side? Unless you have a script.

I wanted to learn a little about all these estimators... I think this would be a good way to do that :) Shall I start working on it?

Sure!

I am working on this.

@ltcguthrie which part? There are many models to work on for this.

I'll start with TfidfVectorizer.

Working on RandomForestClassifier and RandomForestRegressor

Working on PassiveAggressiveClassifier, PassiveAggressiveRegressor

Working on LinearSVC, LinearSVR

Working on StandardScaler, MinMaxScaler

Working on ElasticNet, ElasticNetCV

Also, created google doc for list of done/undone objects

https://docs.google.com/spreadsheets/d/19D-RQocsLk4BM7-Xax8hVvIu3XDgwYSUnvja4cMrJww/edit#gid=0

@lodurality you can also create a list here with checkboxes

• [ ] like
• [ ] this

I'm going to mark it as 0.21 and good first issue since I believe it will be helpful to include a small example for every class.

working on sklearn/cluster

Hi, I'm working on adding an example for the Imputer as a first contribution.

Taking up linear regression: http://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html

Working on TfidfVectorizer and CountVectorizer.

Looks like these were worked on in #7961, #8519, and #8525 but not yet finished or merged.

Also working on HashingVectorizer in sklearn/feature_extraction/text.py.

Taking OPTICS. (depends on #11677)

Taking sklearn/feature_selection/univariate_selection.py

I think you can still provide an example for OPTICS even with the bug :)

@qinhanmin2014 do you mean an example w/o an outlier so that it doesn't show the bug? I guess it'd be nice to have an example which shoes how OPTICS detects an outlier, and I suspect we'd forget about changing the example once that bug is fixed. I can write a note in the _bug_ issue to fix the example once the problem is fixed, but then the initial example won't have the outlier. Should I do that?

do you mean an example w/o an outlier so that it doesn't show the bug?

No, I mean one with outlier(s), like DBSCAN

I've not looked into the bug but I'd expect that OPTICS can detect correct outliers most of the time, so you can still construct an example. If this is not the case, I think we need to tag the bug as blocker.

Fair enough, I'll try and see if I can find a simple example where it works fine.

Taking on sklearn/linear_model/ridge.py

taking on sklearn/covariance/graph_lasso_.py

taking sklearn/linear_model/logistic.py

taking preprocessing classes.

taking Perceptron

I have added the example in the Perceptron class but for some unknown reasons it won't work.
Help needed,I am a novice!(ps-I have already created a pull request).

https://github.com/scikit-learn/scikit-learn/pull/11798

taking sklearn/random_projection

[[MRG] Perceptron examples in Perceptron.py to scikit-learn/scikit-learn/linear_model/perceptron.py #11798

Taking 'sklearn/manifold`

I am working on sklearn.feature_extraction.image.PatchExtractor

Andrea and I will be working on sklearn.cluster.Ward and LedoitWolf example. We are updating the Excel tracker to ensure no one is putting on any duplicated efforts.

I am working on sklearn.decomposition.DictionaryLearning

We just updated the google doc from @lodurality here:
https://docs.google.com/spreadsheets/d/19D-RQocsLk4BM7-Xax8hVvIu3XDgwYSUnvja4cMrJww/edit#gid=0

Me and Shimeng will be working on the missing ones.

@andreanr, that doc is kinda old and missing a whole bunch of new classes. A better idea would be to take a module or a folder, and cover all the public classes in it.

@adrinjalali Yes definitely, we will finish the outstanding ones in the current list and then we will update the list base on the updated modules.

@andreanr are you working on the classes in ensemble.gradient_boosting, too? Otherwise, I'd like to work on that.

@daten-kieker go ahead!

Hi,

Is there anything left that I could contribute to?

Thanks.

@srividhyaprakash I only added examples for estimators in ensemble.gradient_boosting. You could have a look at the other classes.

@daten-kieker, Thanks for your reply. This is my first time in a public well-maintained repo. Could you please guide me on a possible starter guide to commit style and format for the example?
Thanks.

@srividhyaprakash , you can start by going through what you find in the developer's guide. You can also ask on gitter if you have questions getting started. I wish you a joyful contribution journey here.

Once you start on this particular issue, you can check some of the pull requests you see on this thread to get an idea of what it involves.

Hello, is there anything left that I would be able to assist with?

If it is still available, could I deal with the LinearRegression docstring?

@adrinjalali apologies, I just read through the different commits and I saw that you already finished the LinearRegression in #11975 . May I deal with the LogisticRegression docstring?

We've kinda lost track of what's left here. One contribution would be to actually check all public classes and list the ones which still need an example, and take it from there.

@adrinjalali, I can compile a list of what's left

@adrinjalali, I got so excited I forgot to ask, is there a specific branch of this project that has the latest updates?

master has the latest updates. On the other hand, one approach might be to
consider the generated API documentation (e.g. at
https://github.com/scikit-learn/scikit-learn.github.io/tree/master/dev/modules/generated)
and grep for files not containing class="rubric">Examples

@jnothman, thank you, the grepping advice was spot on. I will start compiling the list today

Any more help need here to either to update a existing doc or to create a new one..?
It would help me get started with open source too....so any directions or work would be appreciated.

We still need an updated list of classes without examples. Ping @Khayyon1?

@Khayyon1 have you completed the list or do you need some help..?

@adrinjalali, I am going to have to drop it because school resumed shortly after I asked to assist and I have been unable to work on this, my apologies

@adrinjalali so should i see which all are remaining from the list mentioned by @raghavrv

@coderop2 that's a 3-4 year old list, probably better to start a new one!

Sry for the late reply...As soon as my college exams are over i will start compiling a new list while keeping the old list in mind....i hope that will be enough to act as a reference for everyone

$ # list those with examples
$ git grep -p '^    Examples$' sklearn | grep '=class ' | sed 's/[^ ]* //;s/(.*//;s/:.*//' | sort > /tmp/classes_with_examples.txt
$ # rough list of all public classes
$ grep '\.[A-Z][a-zA-Z]\+' doc/modules/classes.rst  > /tmp/classes.txt
$ # classes without examples
$ grep -v -wFf /tmp/classes_with_examples.txt /tmp/classes.txt

...
that was incorrect. See below.
...

Not certain this is accurate

This may be a more accurate one, still has false positives. Also, assuming we don't need examples for Mixin and Warning classes.

BaggingRegressor
BallTree
BaseEstimator
BayesianGaussianMixture
CalibratedClassifierCV
ClassifierChain
ColumnTransformer
CompoundKernel
ConstantKernel
DictionaryLearning
DistanceMetric
DotProduct
DummyClassifier
DummyRegressor
Exponentiation
ExpSineSquared
ExtraTreeClassifier
ExtraTreeRegressor
ExtraTreesClassifier
ExtraTreesRegressor
FunctionTransformer
GaussianMixture
GradientBoostingClassifier
GradientBoostingRegressor
GraphLasso
GraphLassoCV
GroupShuffleSplit
Hyperparameter
Imputer
IsolationForest
IsotonicRegression
IterativeImputer
KDTree
Kernel
KernelDensity
LocalOutlierFactor
Matern
Memory
MiniBatchDictionaryLearning
MLPClassifier
MLPRegressor
MultiOutputClassifier
MultiOutputRegressor
OAS
OneClassSVM
OneVsOneClassifier
OneVsRestClassifier
OPTICS
OutputCodeClassifier
PairwiseKernel
Parallel
Product
RandomizedSearchCV
RandomTreesEmbedding
RationalQuadratic
RBF
RegressorChain
SelectFromModel
SparseCoder
Sum
TfidfTransformer
WhiteKernel

Yes, I had the wrong lookup, giving many false positives. This is better:

$ grep -v -f <(cat /tmp/classes_with_examples.txt | sed 's/.*/\\.&$/')  /tmp/classes.txt
   base.BaseEstimator
   base.BiclusterMixin
   base.ClassifierMixin
   base.ClusterMixin
   base.DensityMixin
   base.RegressorMixin
   base.TransformerMixin
   calibration.CalibratedClassifierCV
   cluster.OPTICS
    compose.ColumnTransformer
   covariance.OAS
   decomposition.DictionaryLearning
   decomposition.MiniBatchDictionaryLearning
   decomposition.SparseCoder
   dummy.DummyClassifier
   dummy.DummyRegressor
   ensemble.BaggingClassifier
   ensemble.BaggingRegressor
   ensemble.ExtraTreesClassifier
   ensemble.ExtraTreesRegressor
   ensemble.GradientBoostingClassifier
   ensemble.GradientBoostingRegressor
   ensemble.IsolationForest
   ensemble.RandomTreesEmbedding
   exceptions.ChangedBehaviorWarning
   exceptions.ConvergenceWarning
   exceptions.DataConversionWarning
   exceptions.DataDimensionalityWarning
   exceptions.EfficiencyWarning
   exceptions.NonBLASDotWarning
   exceptions.UndefinedMetricWarning
   feature_extraction.text.TfidfTransformer
   feature_selection.SelectFromModel
  gaussian_process.kernels.CompoundKernel
  gaussian_process.kernels.ConstantKernel
  gaussian_process.kernels.DotProduct
  gaussian_process.kernels.ExpSineSquared
  gaussian_process.kernels.Exponentiation
  gaussian_process.kernels.Hyperparameter
  gaussian_process.kernels.Kernel
  gaussian_process.kernels.Matern
  gaussian_process.kernels.PairwiseKernel
  gaussian_process.kernels.Product
  gaussian_process.kernels.RBF
  gaussian_process.kernels.RationalQuadratic
  gaussian_process.kernels.Sum
  gaussian_process.kernels.WhiteKernel
   isotonic.IsotonicRegression
   impute.IterativeImputer
   mixture.BayesianGaussianMixture
   mixture.GaussianMixture
   model_selection.GroupShuffleSplit
   model_selection.RandomizedSearchCV
    multiclass.OneVsRestClassifier
    multiclass.OneVsOneClassifier
    multiclass.OutputCodeClassifier
    multioutput.ClassifierChain
    multioutput.MultiOutputRegressor
    multioutput.MultiOutputClassifier
    multioutput.RegressorChain
   neighbors.BallTree
   neighbors.DistanceMetric
   neighbors.KDTree
   neighbors.KernelDensity
   neighbors.LocalOutlierFactor
   neural_network.MLPClassifier
   neural_network.MLPRegressor
   preprocessing.FunctionTransformer
   svm.OneClassSVM
   tree.ExtraTreeClassifier
   tree.ExtraTreeRegressor
   utils.Memory
   utils.Parallel
   covariance.GraphLasso
   covariance.GraphLassoCV
   preprocessing.Imputer

This is 76 while your list is 62: it excludes base.* and exceptions.* which is fair enough, @adrinjalali, while including strange things like Sum.

Sorry, we both include Sum. Silly.

Taking on GroupShuffleSplit

I would work on dummy (DummyClassifier, DummyRegressor).

This is the current list of classes that might need examples:

base.BaseEstimator
base.BiclusterMixin
base.ClassifierMixin
base.ClusterMixin
base.DensityMixin
base.RegressorMixin
base.TransformerMixin
cluster.OPTICS
compose.ColumnTransformer
covariance.OAS
decomposition.DictionaryLearning
decomposition.MiniBatchDictionaryLearning
decomposition.SparseCoder
ensemble.BaggingClassifier
ensemble.BaggingRegressor
ensemble.ExtraTreesClassifier
ensemble.ExtraTreesRegressor
ensemble.GradientBoostingClassifier
ensemble.GradientBoostingRegressor
ensemble.IsolationForest
ensemble.RandomTreesEmbedding
exceptions.ChangedBehaviorWarning
exceptions.ConvergenceWarning
exceptions.DataConversionWarning
exceptions.DataDimensionalityWarning
exceptions.EfficiencyWarning
exceptions.NonBLASDotWarning
exceptions.UndefinedMetricWarning
feature_extraction.text.TfidfTransformer
feature_selection.SelectFromModel
gaussian_process.kernels.CompoundKernel
gaussian_process.kernels.ConstantKernel
gaussian_process.kernels.DotProduct
gaussian_process.kernels.ExpSineSquared
gaussian_process.kernels.Exponentiation
gaussian_process.kernels.Hyperparameter
gaussian_process.kernels.Kernel
gaussian_process.kernels.Matern
gaussian_process.kernels.PairwiseKernel
gaussian_process.kernels.Product
gaussian_process.kernels.RBF
gaussian_process.kernels.RationalQuadratic
gaussian_process.kernels.Sum
gaussian_process.kernels.WhiteKernel
impute.IterativeImputer
inspection.PartialDependenceDisplay
metrics.RocCurveDisplay
mixture.BayesianGaussianMixture
mixture.GaussianMixture
multiclass.OneVsRestClassifier
multiclass.OneVsOneClassifier
multiclass.OutputCodeClassifier
multioutput.ClassifierChain
multioutput.MultiOutputRegressor
multioutput.MultiOutputClassifier
multioutput.RegressorChain
neighbors.BallTree
neighbors.DistanceMetric
neighbors.KDTree
neighbors.KernelDensity
neighbors.LocalOutlierFactor
neural_network.MLPClassifier
neural_network.MLPRegressor
preprocessing.FunctionTransformer
svm.LinearSVC
tree.ExtraTreeClassifier
tree.ExtraTreeRegressor
utils.Memory
utils.Parallel

Hi @pspachtholz @MechCoder , I would like to work on this. I am interested to contribute to scikit-learn and think that this can be a good starting point.

Thanks

@PyExtreme I think you can just go ahead, select one/more classes from the list that you find interesting and add examples. I would then post what youre working on here to avoid duplicate work. For guidance you can take a look at previously merged pull requests.

@pspachtholz , I am picking up _ExtraTreesClassifier_ initially and after making commit, I would like to pick up in batches.

Thanks

Worked on gradient boosting.

Picking up neural_network.MLPClassifier & neural_network.MLPRegressor

I'm on svm.LinearSVC

I'm working on sklearn.multioutput.MultiOutputClassifier

Hi, I'll be working on mixture.BayesianGaussianMixture and mixture.GaussianMixture

I'm on feature_extraction.text.TfidfTransformer. Wish me good luck!

I'm on feature_extraction.text.TfidfTransformer. Wish me good luck!

also working on it - Maybe we can share examples?

picking up ensemble.ExtraTreesClassifier

I'm picking the ensemble.BaggingRegressor

I'm working on feature_selection.SelectFromModel

I am picking neighbors.KernelDensity

I'm on ensemble.IsolationForest

on multiclass.OneVsRestClassifier

Hey, I'll be working on the Gaussian Process kernels, starting with gaussian_process.kernels.RBF and then do the other kernels with @thorbenjensen

Working on ensemble.GradientBoostingClassifier

Working on Perceptron

Working on ensemble.GradientBoostingRegressor

Next up tree.ExtraTreeClassifier

working on ensemble.GradientBoostingRegressor

taking a look at multiclass.OutputCodeClassifier

Looking at neighbors.LocalOutlierFactor.

working on ensemble.RandomTreesEmbedding

picking up ensemble.ExtraTreesClassifier

Hi @jorahn , I have already submitted a PR on that.
Please feel free to pick another module.

I suggest looking through the comments before to see if someone is already working on a class to avoid duplicated work.
For some classes there are already some (possibly stale) mrg requests, where we could ask the authors whether they are still actively working on it.

@flaviomorelli @LBrummer @mschaffenroth I did the ensemble gradient in this PR already #15151

Really cool btw to see that this is picking up speed :-)

picking up ensemble.ExtraTreesClassifier

Hi @jorahn , I have already submitted a PR on that.
Please feel free to pick another module.

oh, didn't see that on this issue. so we now have two PRs for that

Perceptron is already documented.

working on tree.ExtraTreeRegressor

working on impute.IterativeImputer

@pspachtholz , I am picking up _ExtraTreesClassifier_ initially and after making commit, I would like to pick up in batches.

Thanks

@jorahn , I had mentioned about it 1 week ago here only and had already been working on it since 1 week.

Working on OneVsOneClassifier.

I'm going for the metrics.RocCurveDisplay

working on multioutput.ClassifierChain

Working on PriorProbabilityEstimator.

Working on multioutput.MultiOutputRegressor

taking neighbors.BallTree

PriorProbabilityEstimator is deprecated in version 0.21 and will be removed in version 0.23. Not working on this.

Working on covariance.OAS

SelectPercentile already documented.

working on ensemble.ExtraTreesRegressor

taking neighbors.BallTree

there are some examples for BallTree and compose.ColumnTransformer already

RANSACRegressor already documented.

taking decomposition.SparseCoder

SelectKBest already documented.

SpectralClustering already documented

Working on IsolationForest

Working on IsolationForest

Hi @zioalex, I already have a PR for this: #15205. Happy to take suggestions and comments :smiley:

Working in neighbors.DistanceMetric.

working on multioutput.RegressorChain

working on exceptions.ConvergenceWarning

working on exceptions.ChangedBehaviorWarning

working on exceptions.ChangedBehaviorWarning

After taking to @adrinjalali, we concluded it is better not to work on this one since it is release dependent. And it probably will be moved somewhere else.

For your peace of mind be like us, do not work on exceptions.ChangedBehaviorWarning

Working on exceptions.DataDimensionalityWarning

I would like to work on a few. I will start with optics.

Hi! I have just raised a PR after adding an example code snippet to K-means clustering. This is my first open source contribution, so it would be great if someone could look into this and see if it needs any more work!
Thanks and Regards,
Smriti Singh

I went through the list from last October and checked which classes already have merged PRs/examples. I suppose that claims from last year or longer ago, on which nothing further has happened, can be ignored.
So here comes an updated todo list:

don't work on this, it is release dependent: exceptions.ChangedBehaviorWarning

open PRs:

decomposition.SparseCoder #15233

~exceptions.DataDimensionalityWarning #15246~
mixture.BayesianGaussianMixture #15193
mixture.GaussianMixture #15193
multioutput.ClassifierChain #15211
multioutput.RegressorChain #15215

new open prs:
decomposition.DictionaryLearning #16907
~exceptions.EfficiencyWarning #16785~
~exceptions.UndefinedMetricWarning #16784~

Free for the Taking:

(EDIT: those in base are better suited for the developer guide, let's ignore them for now)
~base.BaseEstimator~
~base.BiclusterMixin~
~base.ClassifierMixin~
~base.ClusterMixin~
~base.DensityMixin~
~base.RegressorMixin~
~base.TransformerMixin~

decomposition.MiniBatchDictionaryLearning

~exceptions.NonBLASDotWarning~

~feature_selection.SelectFromModel~

gaussian_process.kernels.CompoundKernel
gaussian_process.kernels.Hyperparameter
~gaussian_process.kernels.Kernel~

~inspection.PartialDependenceDisplay~

~multiclass.OneVsOneClassifier~

~multioutput.MultiOutputClassifier~

~utils.Memory~
~utils.Parallel~

Update. Now merged:
covariance.OAS #16681
multioutput.MultiOutputRegressor #16698
tree.ExtraTreeClassifier #16671

neighbors.DistanceMetric
neighbors.KDTree
neighbors.LocalOutlierFactor

exceptions.DataConversionWarning #16704

multiclass.OneVsOneClassifier #16700

Do we want examples for the base.* classes? I guess it makes more sense to have them better documented in the developer guide. WDYT @jnothman @NicolasHug ?

Agreed it's better for the developer guide. I'm editing the comment

I think it was closed by accident.

multioutput.MultiOutputClassifier already has examples. Should it be updated to include an example of the attribute or should it be taken off the TO DO list?

I think it can be removed from the list. Thanks @marenwestermann

I think it can be removed from the list. Thanks @marenwestermann

Ok, then feature_selection.SelectFromModel can also be removed because it has examples too. (These were added in October last year.)

Regarding utils.Memory and utils.Parallel:
on the scikit-learn website they are said "to be removed in version 0.23". I checked the utils.__init__.py file where these classes lived in and they have been removed. So these can be taken off the list, too.

I'll try and tackle gaussian_process.kernels.Kernel :)

Apart from the above DictionaryLearning/MiniBatchDictionaryLearning PR, I also had a look at the neighbors.* classes: They all already have at least one example. The ones for neighbors.KDTree and neighbors.BallTree are generated from the CLASS_DOC format string which can be found in the _binary_tree.pxi include file..

It is my first time contributing this project. I would like to look into neighbors.* classes including:
neighbors.DistanceMetric
neighbors.KDTree
neighbors.LocalOutlierFactor

I had a look at gaussian_process.kernels.Kernel. According to the documentation it is the "Base class for all kernels". Therefore, all its attributes are read-only property attributes (methods with @property decorators). From my understanding it therefore doesn't make sense to add examples to this class because it can only be used in combination with another class.
However, if you scroll to the bottom of the webpage of gaussian_process.kernels.Kernel there's a link to an example of how this class can be used. This was added in Nov 2019. In this example the class SequenceKernel is created which inherits from the Kernel class. The SequenceKernel class is not a feature of scikit-learn but might be an interesting feature to add.

Please correct me if anything that I've written here is wrong.

Yeah it'd make it too complicated to write an example for it. I'm happy for it to be removed from the list.

@Malesche exceptions.DataDimensionalityWarning is now closed and can therefore be taken off the TODO list.

Regarding inspection.PartialDependenceDisplay:
In the class description it is said, "It is recommended to use plot_partial_dependence to create a PartialDependenceDisplay ". I had a look at this function and inside it a PartialDependenceDisplayobject is created and its plot method is called. There are examples for how to use plot_partial_dependence, and I therefore think that there's no need to add examples to inspection.PartialDependenceDisplay. What do you think, @adrinjalali ?

agreed @marenwestermann

exceptions.NonBLASDotWarning can also be taken off the TODO list because a decision against having examples in exceptions.py has been made (see #17040).

@NicolasHug, @amueller just in case you want to use this issue as a sprint issue, you can find here the list of the classes still missing examples (thanks to Joel script, amended by himself... :) ). I have already removed base and exceptions classes. May I suggest to edit the issue with the list just at the beginning? This will make easier to identify available classes. PR still opened from previous sprints (not only) are linked into the list: I think it will be useful to finalize them before a new event (thanks @thomasjpfan for having started some reviews already).

Thanks for the suggestion @cmarmo, I updated the issue.

HI, I will try to take on gaussian_process.kernels.Hyperparameter.

Hi, I will take on linear_model.*.

• linear_model.PoissonRegressor
• linear_model.TweedieRegressor
• linear_model.GammaRegressor

Hi, We will take ensemble.GradientBoostingClassifier

Hi, We will take ensemble.GradientBoostingClassifier

This one already has an example.

Hey, We are now taking:

• metrics.ConfusionMatrixDisplay
• metrics.PrecisionRecallDisplay

Hey @adrinjalali , Can you please check the PR? Thanks!

@emdupre and I will work on below as part of the data umbrella sprint.

• neighbors.BallTree
• neighbors.DistanceMetric
• neighbors.KDTree

Hello @adrinjalali, Seems like we can already see some examples for below. It's not updated in the todo list. Please suggest.

• neighbors.BallTree
• neighbors.DistanceMetric
• neighbors.KDTree

Hi I will work on:

• decomposition.DictionaryLearning (#16907)
• decomposition.MiniBatchDictionaryLearning (#16907)
• decomposition.SparseCoder (#15233)

Hi, I will work on:

• gaussian_process.kernels.CompoundKernel

Closing this as all the PRs have been merged and there is no more relevant classes without examples.
Thanks @j2heng for helping in triaging during the sprint!

Oh really. The list on the top is out-of-date?

Oh yes I merged the other PR.