[MRG] DOC: shift projects missing in related_projects from the wiki page by dalmia · Pull Request #8297 · scikit-learn/scikit-learn

dalmia · 2017-02-05T21:54:29Z

Reference Issue

What does this implement/fix? Explain your changes.

Links the related projects from the wiki page to related_projects.rst

Any other comments

Am linking everything that is missing as of now. However, in the issue thread it was mentioned that we need include only the important. Please provide feedback as to what all should be included.

jnothman · 2017-02-06T05:32:22Z

doc/related_projects.rst

  wrapper around scikit-learn that makes it easy to run machine learning
  experiments with multiple learners and large feature sets.

+- `sklearn-deap <https://github.com/rsteca/sklearn-deap>`_ Use evolutionary


I don't think this belongs in this subsection, which is about providing API wrappers. PyMC appears to be accidentally, incorrectly, in this section.

The other possible choices seem - Auto-ML and Model Export for production. Might you have any preference?

jnothman · 2017-02-06T05:34:22Z

doc/related_projects.rst

 - `mlxtend <https://github.com/rasbt/mlxtend>`_ Includes model visualization
  utilities.

+- `Fast svmlight / libsvm file loader <https://github.com/mblondel/svmlight-loader>`_


I'm not certain whether this is current, i.e. still much faster than what's in sklearn.datasets. @mblondel?

Yep, it is still significantly faster.

Should be kept included then.

jnothman · 2017-02-06T05:37:16Z

doc/related_projects.rst

+  Caruana et al's Ensemble Selection algorithm in Python, based on scikit-learn
+
+- `random-output-trees <https://github.com/arjoly/random-output-trees>`_
+  Multi-output random forest on randomised output space


I think this needs to be clearer about what it's useful for

jnothman · 2017-02-06T05:38:22Z

doc/related_projects.rst


+- `libOPF <https://github.com/LibOPF/LibOPF>`_ Optimal path forest classifier
+
+- `pyensemble <https://github.com/dclambert/pyensemble>`_ An implementation of


I suspect this is better fit in the Auto-ML section. But you would do well to check that.

I felt this was a trade-off between the two but maybe Auto-ML might be more natural for it. I'll make the change.

jnothman · 2017-02-06T05:40:09Z

doc/related_projects.rst

  K-means and mixture of von Mises Fisher clustering routines for data on the
  unit hypersphere.

+- `pyIPCA <https://github.com/pickle27/pyIPCA>`_ Incremental Principal


How, if at all, does this differ from our IncrementalPCA?

Seems to be a modification of our IncrementalPCA like:

class CCIPCA(BaseEstimator, TransformerMixin): """Candid covariance-free incremental principal component analysis (CCIPCA) Linear dimensionality reduction using an online incremental PCA algorithm. CCIPCA computes the principal components incrementally without estimating the covariance matrix. This algorithm was designed for high dimensional data and converges quickly. This implementation only works for dense arrays. However it should scale well to large data.

jnothman · 2017-02-06T05:40:35Z

doc/related_projects.rst

 - `Deep Learning <http://deeplearning.net/software_links/>`_ A curated list of deep learning
  software libraries.

+- `glm-sklearn <https://github.com/jcrudy/glm-sklearn>`_ scikit-learn


This fits in regression and classification above

jnothman · 2017-02-06T05:41:31Z

doc/related_projects.rst

+- Generating data with `non-parametric Gaussian mixture models <https://gist.github.com/2011426>`_
+  Useful if you need "random" data that should have non-trivial structure.
+
+- `scikit-protopy <https://github.com/dvro/scikit-protopy>`_ scikit-learn


This best lives with decomposition and clustering.

jnothman · 2017-02-06T05:41:55Z

doc/related_projects.rst

 ---------------------

-The `wiki <https://github.com/scikit-learn/scikit-learn/wiki/Third-party-projects-and-code-snippets>`_ has more!
+**Gists**


I think these gists can't really be considered projects, and a Wiki page might best be retained for them.

Yes, I was skeptic about including this here too. Should we then rename the original wiki to just "Code Snippets" ?

But since we are going to include this heading anyways, it might be better to retain the gists in this page and save the trouble of maintaining another wiki page separately?

dalmia · 2017-02-16T15:16:59Z

I have removed the gists as they correctly don't belong here. We should retain the original wiki page for them. If this seems fine, I'll move on to making the change there.

codecov · 2017-02-16T19:01:22Z

Codecov Report

Merging #8297 into master will decrease coverage by 1.44%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #8297      +/-   ##
==========================================
- Coverage   96.19%   94.75%   -1.45%     
==========================================
  Files         348      342       -6     
  Lines       64645    60809    -3836     
==========================================
- Hits        62187    57617    -4570     
- Misses       2458     3192     +734

Impacted Files	Coverage Δ
sklearn/feature_extraction/tests/test_image.py	`4.78% <0%> (-95.22%)`	⬇️
sklearn/feature_extraction/image.py	`58.16% <0%> (-41.84%)`	⬇️
sklearn/utils/arpack.py	`42.5% <0%> (-32.5%)`	⬇️
sklearn/utils/random.py	`59.29% <0%> (-32.2%)`	⬇️
sklearn/datasets/tests/test_kddcup99.py	`29.16% <0%> (-10.84%)`	⬇️
sklearn/manifold/mds.py	`84.46% <0%> (-9.71%)`	⬇️
sklearn/utils/tests/test_estimator_checks.py	`88.23% <0%> (-8.96%)`	⬇️
sklearn/linear_model/tests/test_bayes.py	`83.01% <0%> (-8.16%)`	⬇️
sklearn/utils/tests/test_utils.py	`92.94% <0%> (-7.06%)`	⬇️
sklearn/manifold/spectral_embedding_.py	`85.16% <0%> (-6.25%)`	⬇️
... and 260 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 3e29334...ce6d7d3. Read the comment docs.

jnothman · 2017-02-23T09:22:02Z

doc/related_projects.rst

+- `random-output-trees <https://github.com/arjoly/random-output-trees>`_
+  Randomized output tree for multilabel / multi-output regression tasks
+
+- `fastFM <https://github.com/ibayer/fastFM>`_ Fast factorization machine


"Factorization" makes me pretty sure this belongs with decomposition, not classification/regression

jnothman · 2017-02-23T09:22:17Z

doc/related_projects.rst

+- `fastFM <https://github.com/ibayer/fastFM>`_ Fast factorization machine
+  implementation compatible with scikit-learn
+
+- `glm-sklearn <https://github.com/jcrudy/glm-sklearn>`_ scikit-learn


I'd put this higher up in the list.

jnothman · 2017-02-23T09:22:52Z

doc/related_projects.rst

  software libraries.

+- `sklearn-deap <https://github.com/rsteca/sklearn-deap>`_ Use evolutionary
+   algorithms instead of gridsearch in scikit-learn.


gridsearch -> grid search.

This does not belong here. Perhaps a "model selection and evaluation" section under "other estimators and tasks"

dalmia changed the title ~~DOC: incorporate related projects (except gists)~~ [WIP] DOC: incorporate related projects (except gists) Feb 5, 2017

dalmia changed the title ~~[WIP] DOC: incorporate related projects (except gists)~~ [WIP] DOC: shift projects missing in related_projects from the wiki page Feb 5, 2017

dalmia changed the title ~~[WIP] DOC: shift projects missing in related_projects from the wiki page~~ [MRG] DOC: shift projects missing in related_projects from the wiki page Feb 5, 2017

jnothman reviewed Feb 6, 2017

View reviewed changes

dalmia added 4 commits February 18, 2017 01:16

DOC: incorporate related projects (except gists)

0f82adf

DOC: add gists from wiki page

c10dc97

DOC: change order / description of some projects

1f208ba

DOC: remove gists and rearrange sklearn-deap

ce6d7d3

dalmia force-pushed the 8266 branch from 8757e64 to ce6d7d3 Compare February 17, 2017 19:47

jnothman reviewed Feb 23, 2017

View reviewed changes

jnothman mentioned this pull request Apr 12, 2017

Retire Third-party-projects wiki page? #8266

Closed

jnothman added Easy Well-defined and straightforward way to resolve Stalled good first issue Easy with clear instructions to resolve help wanted labels Feb 6, 2018

cmarmo mentioned this pull request May 5, 2020

[MRG] DOC: shift projects missing in related_projects from the wiki page #17129

Merged

cmarmo removed the help wanted label May 5, 2020

rth closed this in #17129 May 5, 2020


		- `libOPF <https://github.com/LibOPF/LibOPF>`_ Optimal path forest classifier

		- `pyensemble <https://github.com/dclambert/pyensemble>`_ An implementation of

Uh oh!

Conversation

dalmia commented Feb 5, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Reference Issue

What does this implement/fix? Explain your changes.

Any other comments

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dalmia Feb 6, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dalmia commented Feb 16, 2017

Uh oh!

codecov bot commented Feb 16, 2017 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

dalmia commented Feb 5, 2017 •

edited

Loading

dalmia Feb 6, 2017 •

edited

Loading

codecov bot commented Feb 16, 2017 •

edited

Loading