Apache Spark itself
Spark originally came out of Berkeley AMPLab and even today AMPLab projects, even though they are not in Apache Spark Foundation, enjoy a status a bit over your everyday github project.
Spark's own MLLib forms the bottom layer of the three-layer ML Base, with MLI being the middle layer and ML Optimizer being the most abstract layer.
3. ML Optimizer (aka Ghostface)
Ghostware was described in 2014 but never released. Of the 39 machine learning libraries, this is the only one that is vaporware, and is included only due to its AMPLab and ML Base status.
Other than ML Base
A recent project from June, 2015, this set of stochastic learning algorithms claims 25x - 75x faster performance than Spark MLlib on Stochastic Gradient Descent (SGD). Plus it's an AMPLab project that begins with the letters "sp", so it's worth watching.
5. Keystone ML
Brought machine learning pipelines to Spark, but pipelines have matured in recent versions of Spark. Also promises some computer vision capability, but there are limitations I previously blogged about.
A server to manage a large collection of machine learning models.
Faster machine learning on Spark by optimizing communication patterns and shuffles, as described in the paperCommunication-Efficient Distributed Dual Coordinate Ascent
I previously blogged DeepLearning4j Adds Spark GPU Support
Brand new and frankly why I started this list for this blog post. Provides an interface to Keras.
Parameter server for model-parallel rather than data-parallel (as Spark's MLlib is).
From Airbnb, used in their automated pricing
Logistic regression, LDA, Factorization machines, Neural Network, Restricted Boltzmann Machines
Similar to Spark DataFrames, but agnostic to engine (i.e. will run on engines other than Spark in the future). Includes cross-validation and interfaces to external machine learning libraries.
Interfaces to other Machine Learning systems
Wraps Stanford CoreNLP.
Interface to Python's Scikit-learn
16. Sparkling Water
Interface to H2O
Wraps Hivemall, machine learning in Hive
Export PMML, an industry standard XML format for transporting machine learning models.
Add-ons that enhance MLlib's existing algorithms
Adds dropout capability to Spark MLLib, based on the paper Dropout: A simple way to prevent neural networks from overfitting.
Adds arbitrary distance functions to K-Means
Visualize the Streaming Machine Learning algorithms built into Spark MLlib
Recursive Neural Networks (RNNs)
SVM based on the performant Spark communication framework CoCoA listed above.
25. Sparkling Ferns
Matrix Factorization Recommendation System
40x faster clustering than Spark MLlib K-Means
K-Means that produces more uniformly-sized clusters, based on A Comparison of Document Clustering Techniques
Build graphs using k-nearest-neighbors and locality sensitive hashing (LSH)
Online Latent Dirichlet Allocation (LDA), Gibbs Sampling LDA, Online Hierarchical Dirichlet Process (HDP)
Algorithm building blocks
Adaboost and MP-Boost
Port to Spark of TFOCS: Templates for First-Order Conic Solvers. If your machine learning cost function happens to be convex, then TFOCS can solve it.
Linear algebra operators to work with Spark MLlib's linalg package
Information-theoretic basis for feature selection, based on Conditional likelihood maximisation: a unifying framework for information theoretic feature selection
Given labeled data, "discretize" one of the continuous numeric dimensions such that each bin is relatively homogenous in terms of data classes. This is a foundational idea CART and ID3 algorithms to generate decision trees. Based on Multi-interval discretization of continuous-valued attributes for classification learning.
Distributed t-Distributed Stochastic Neighbor Embedding (t-SNE) for dimensionality reduction.
Sparse feature vectors
K-Means, Regression, and Statistics
39. Twitter data