Ted Dunning

t-digest. A new data structure for accurate on-line accumulation of rank-based statistics such as quantiles and trimmed means

MiA. Mahout in Action Example Code

log-synth. Generates more or less realistic log data for testing simple aggregation queries.

anomaly-detection. A simple demonstration of sub-sequence sampling as used for anomaly detection with EKG signals

Plume. Explorations relative to cloning FlumeJava

knn. Large scale k-nn experiments

pig-vector. Mahout vector encoding for pig

bandit-ranking. HTML

feature-extraction. Sample techniques for a variety of feature extraction methods

python-llr. A python implementation of the most commonly used variants of the G-test

open-json. Open JSON - a truly open source JSON implementation

probability-book. A copy of the source for Grinstead and Snell's lovely probability book

k-means-auto-encoder. Some quick exploration of how k-means auto-encoders work

t-digest-benchmark. A simple JMH benchmark for various versions of t-digest

in-memory-cooccurrence. Analyze for significant cooccurrence using Mahout sparse matrices

sequencemodel. The sequence anomaly detector from our second in the Practical Machine Learning Series

pcap-filter. Experiments in PCAP file decoding at speed

Chapter-16. Example server for Chapter 16 of Mahout in Action

go-wspr. A WSPR beacon that uses novel techniques to control an Si5351 for mHz frequency control even into the UHF range.

ponies. Sample recommender flow for search as recommendation

ancient-stats. The ancient C version of the LLR statistics and related utilities. This is for reference only.

mahout-examples. Mahout Examples

parksim. Java

sequence-model. A simple implementation of a probabilistic model for event sequences

TDigest. Native Julia implementation of t-digest

config-print. Prints Hadoop configuration variables

EvenDigits. Program that scans for new entries of https://oeis.org/A068994 using a technique that gives >1000x speedup

h2o-matrix. Demonstration of Mahout compatible matrix and vector types based on h2o

H3Geometry. H3 convenience package

cluster-hinting. R

freezer. A hybrid discrete/continuous simulation of a freezer and its users

graph-demo. Demonstrates use of the multi command in Zookeeper

MicroSpice.jl. A native Julia circuit simulator for passive circuits that is amenable to numerical optimization

stream-lib. Stream summarizer and cardinality estimator.

image-rep. A repo for storing images referenced in other projects

s2cell. Minimal Python S2 Geometry cell ID, token and lat/lon conversion library

hpe-dev-portal. HTML

Pidfile.jl. Utility for creating advisory pidfiles (lock files)

split-search. Quick tests of the basins of attraction in an ERT optimizer

OpenUnits.

SI5351. A microPython port of the AdaFruit CircuitPython module for the SI5351 clock generator

cfgrib.jl. Julia

t-digest-example. Java

meta-ep. Recorded-step meta-mutation implementation and paper

streaming-quantiles. Implements the Karnin-Lang-Liberty (KLL) algorithm in python

G2. Implements the G^2 test for comparing counts

ibm2ieee. Julia package to convert between IBM floating point (aka hexadecimal floating point or HFP) to IEEE floating point

maprdb-tick-example. Java