scikit-learn
diff --git a/‎doc/related_projects.rst
Lines changed: 92 additions & 55 deletions b/‎doc/related_projects.rst
Lines changed: 92 additions & 55 deletions
@@ -12,19 +12,12 @@ Interoperability and framework enhancements
 These tools adapt scikit-learn for use with other technologies or otherwise
 enhance the functionality of scikit-learn's estimators.
 
-- `ML Frontend <https://github.com/jeff1evesque/machine-learning>`_ provides
-  dataset management and SVM fitting/prediction through
-  `web-based <https://github.com/jeff1evesque/machine-learning#web-interface>`_
-  and `programmatic <https://github.com/jeff1evesque/machine-learning#programmatic-interface>`_
-  interfaces.
+**Data formats**
 
 - `sklearn_pandas <https://github.com/paulgb/sklearn-pandas/>`_ bridge for
   scikit-learn pipelines and pandas data frame with dedicated transformers.
 
-- `Scikit-Learn Laboratory
-  <https://skll.readthedocs.io/en/latest/index.html>`_  A command-line
-  wrapper around scikit-learn that makes it easy to run machine learning
-  experiments with multiple learners and large feature sets.
+**Auto-ML**
 
 - `auto-sklearn <https://github.com/automl/auto-sklearn/>`_
   An automated machine learning toolkit and a drop-in replacement for a
@@ -36,6 +29,37 @@ enhance the functionality of scikit-learn's estimators.
   preprocessors as well as the estimators. Works as a drop-in replacement for a
   scikit-learn estimator.
 
+**Experimentation frameworks**
+
+- `PyMC <http://pymc-devs.github.io/pymc/>`_ Bayesian statistical models and
+  fitting algorithms.
+
+- `REP <https://github.com/yandex/REP>`_ Environment for conducting data-driven
+  research in a consistent and reproducible way
+
+- `ML Frontend <https://github.com/jeff1evesque/machine-learning>`_ provides
+  dataset management and SVM fitting/prediction through
+  `web-based <https://github.com/jeff1evesque/machine-learning#web-interface>`_
+  and `programmatic <https://github.com/jeff1evesque/machine-learning#programmatic-interface>`_
+  interfaces.
+
+- `Scikit-Learn Laboratory
+  <https://skll.readthedocs.io/en/latest/index.html>`_  A command-line
+  wrapper around scikit-learn that makes it easy to run machine learning
+  experiments with multiple learners and large feature sets.
+
+**Model inspection and visualisation**
+
+- `eli5 <https://github.com/TeamHG-Memex/eli5/>`_ A library for
+  debugging/inspecting machine learning models and explaining their
+  predictions.
+
+- `mlxtend <https://github.com/rasbt/mlxtend>`_ Includes model visualization
+  utilities.
+
+
+**Model export for production**
+
 - `sklearn-pmml <https://github.com/alex-pirozhenko/sklearn-pmml>`_
   Serialization of (some) scikit-learn estimators into PMML.
 
@@ -47,6 +71,12 @@ enhance the functionality of scikit-learn's estimators.
 - `sklearn-porter <https://github.com/nok/sklearn-porter>`_
   Transpile trained scikit-learn models to C, Java, Javascript and others.
 
+- `sklearn-compiledtrees <https://github.com/ajtulloch/sklearn-compiledtrees/>`_
+  Generate a C++ implementation of the predict function for decision trees (and
+  ensembles) trained by sklearn. Useful for latency-sensitive production
+  environments.
+
+
 Other estimators and tasks
 --------------------------
 
@@ -55,14 +85,7 @@ project. The following are projects providing interfaces similar to
 scikit-learn for additional learning algorithms, infrastructures
 and tasks.
 
-- `pylearn2 <http://deeplearning.net/software/pylearn2/>`_ A deep learning and
-  neural network library build on theano with scikit-learn like interface.
-
-- `sklearn_theano <http://sklearn-theano.github.io/>`_ scikit-learn compatible
-  estimators, transformers, and datasets which use Theano internally
-
-- `lightning <https://github.com/scikit-learn-contrib/lightning>`_ Fast state-of-the-art
-  linear model solvers (SDCA, AdaGrad, SVRG, SAG, etc...).
+**Structured learning**
 
 - `Seqlearn <https://github.com/larsmans/seqlearn>`_  Sequence classification
   using HMMs or structured perceptron.
@@ -81,25 +104,41 @@ and tasks.
   (`CRFsuite <http://www.chokkan.org/software/crfsuite/>`_ wrapper with
   sklearn-like API).
 
-- `py-earth <https://github.com/scikit-learn-contrib/py-earth>`_ Multivariate adaptive
-  regression splines
+**Deep neural networks etc.**
 
-- `sklearn-compiledtrees <https://github.com/ajtulloch/sklearn-compiledtrees/>`_
-  Generate a C++ implementation of the predict function for decision trees (and
-  ensembles) trained by sklearn. Useful for latency-sensitive production
-  environments.
+- `pylearn2 <http://deeplearning.net/software/pylearn2/>`_ A deep learning and
+  neural network library build on theano with scikit-learn like interface.
 
-- `lda <https://github.com/ariddell/lda/>`_: Fast implementation of latent
-  Dirichlet allocation in Cython which uses `Gibbs sampling
-  <https://en.wikipedia.org/wiki/Gibbs_sampling>`_ to sample from the true
-  posterior distribution. (scikit-learn's
-  :class:`sklearn.decomposition.LatentDirichletAllocation` implementation uses
-  `variational inference
-  <https://en.wikipedia.org/wiki/Variational_Bayesian_methods>`_ to sample from
-  a tractable approximation of a topic model's posterior distribution.)
+- `sklearn_theano <http://sklearn-theano.github.io/>`_ scikit-learn compatible
+  estimators, transformers, and datasets which use Theano internally
 
-- `Sparse Filtering <https://github.com/jmetzen/sparse-filtering>`_
-  Unsupervised feature learning based on sparse-filtering
+- `nolearn <https://github.com/dnouri/nolearn>`_ A n
9E7A
umber of wrappers and
+  abstractions around existing neural network libraries
+
+- `keras <https://github.com/fchollet/keras>`_ Deep Learning library capable of
+  running on top of either TensorFlow or Theano.
+
+- `lasagne <https://github.com/Lasagne/Lasagne>`_ A lightweight library to
+  build and train neural networks in Theano.
+
+**Broad scope**
+
+- `mlxtend <https://github.com/rasbt/mlxtend>`_ Includes a number of additional
+  estimators as well as model visualization utilities.
+
+- `sparkit-learn <https://github.com/lensacom/sparkit-learn>`_ Scikit-learn
+  API and functionality for PySpark's distributed modelling.
+
+**Other regression and classification**
+
+- `xgboost https://github.com/dmlc/xgboost` Optimised gradient boosted decision
+  tree library.
+
+- `lightning <https://github.com/scikit-learn-contrib/lightning>`_ Fast
+  state-of-the-art linear model solvers (SDCA, AdaGrad, SVRG, SAG, etc...).
+
+- `py-earth <https://github.com/scikit-learn-contrib/py-earth>`_ Multivariate
+  adaptive regression splines
 
 - `Kernel Regression <https://github.com/jmetzen/kernel_regression>`_
   Implementation of Nadaraya-Watson kernel regression with automatic bandwidth
@@ -108,28 +147,32 @@ and tasks.
 - `gplearn <https://github.com/trevorstephens/gplearn>`_ Genetic Programming
   for symbolic regression tasks.
 
-- `nolearn <https://github.com/dnouri/nolearn>`_ A number of wrappers and
-  abstractions around existing neural network libraries
+- `multiisotonic <https://github.com/alexfields/multiisotonic>`_ Isotonic
+  regression on multidimensional features.
 
-- `sparkit-learn <https://github.com/lensacom/sparkit-learn>`_ Scikit-learn functionality and API on PySpark.
+**Decomposition and clustering**
 
-- `keras <https://github.com/fchollet/keras>`_ Deep Learning library capable of
-  running on top of either TensorFlow or Theano.
-
-- `mlxtend <https://github.com/rasbt/mlxtend>`_ Includes a number of additional
-  estimators as well as model visualization utilities.
-
-- `kmodes <https://github.com/nicodv/kmodes>`_ k-modes clustering algorithm for categorical data, and
-  several of its variations.
+- `lda <https://github.com/ariddell/lda/>`_: Fast implementation of latent
+  Dirichlet allocation in Cython which uses `Gibbs sampling
+  <https://en.wikipedia.org/wiki/Gibbs_sampling>`_ to sample from the true
+  posterior distribution. (scikit-learn's
+  :class:`sklearn.decomposition.LatentDirichletAllocation` implementation uses
+  `variational inference
+  <https://en.wikipedia.org/wiki/Variational_Bayesian_methods>`_ to sample from
+  a tractable approximation of a topic model's posterior distribution.)
 
-- `hdbscan <https://github.com/lmcinnes/hdbscan>`_ HDBSCAN and Robust Single Linkage clustering algorithms
-  for robust variable density clustering.
+- `Sparse Filtering <https://github.com/jmetzen/sparse-filtering>`_
+  Unsupervised feature learning based on sparse-filtering
 
-- `lasagne <https://github.com/Lasagne/Lasagne>`_ A lightweight library to build and train neural networks in Theano.
+- `kmodes <https://github.com/nicodv/kmodes>`_ k-modes clustering algorithm for
+  categorical data, and several of its variations.
 
-- `multiisotonic <https://github.com/alexfields/multiisotonic>`_ Isotonic regression on multidimensional features.
+- `hdbscan <https://github.com/lmcinnes/hdbscan>`_ HDBSCAN and Robust Single
+  Linkage clustering algorithms for robust variable density clustering.
 
-- `spherecluster <https://github.com/clara-labs/spherecluster>`_ Spherical K-means and mixture of von Mises Fisher clustering routines for data on the unit hypersphere.
+- `spherecluster <https://github.com/clara-labs/spherecluster>`_ Spherical
+  K-means and mixture of von Mises Fisher clustering routines for data on the
+  unit hypersphere.
 
 Statistical learning with Python
 --------------------------------
@@ -145,12 +188,6 @@ Other packages useful for data analysis and machine learning.
   statistical models. More focused on statistical tests and less on prediction
   than scikit-learn.
 
-- `PyMC <http://pymc-devs.github.io/pymc/>`_ Bayesian statistical models and
-  fitting algorithms.
-
-- `REP <https://github.com/yandex/REP>`_ Environment for conducting data-driven
-  research in a consistent and reproducible way
-
 - `Sacred <https://github.com/IDSIA/Sacred>`_ Tool to help you configure,
   organize, log and reproduce experiments