A new data structure for accurate on-line accumulation of rank-based statistics such as quantiles and trimmed means
-
Updated
Nov 12, 2026 - Java
A new data structure for accurate on-line accumulation of rank-based statistics such as quantiles and trimmed means
Probabilistic data structures in Go
Routines and data structures for using isarn-sketches idiomatically in Apache Spark
redis module for percentile estimation
Sketching data structures for scala, including t-digest
A modern C++ implementation of a merging t-digest data structure.
Notebook image and notebook for feature reduction talk
A Prototype For Fitting Monotonic Cubic Splines to a Tdigest Sketch
Set of algorithms, used for estimation statistic characteristics on streaming data.
Distinct counts, quantiles and frequencies in bounded memory. AMS Flajolet, NCC-0185.
Demo of 1-pass variable importance using t-digests
Demo notebook and data for Spark Summit Dublin 2017: One-Pass Data Science with Generative T-Digests
A T-Digest is a Probabilistic Data structure (a "sketch") used to Estimate Quantiles and Percentiles from a Large Stream of Data or a Massive Dataset
Benchmarking five approaches to quantiles over large datasets — DDSketch, t-digest, KLL, a fixed-width histogram sketch and an exact parallel-sort baseline. C++17 with CI on Linux and macOS, Python plotting.
Type-classes to interface isarn-sketches with Algebird
Fixed-budget data-stream sketching with data-driven budget allocation
Probabilistic data structures implemented in Go.
To associate your repository with the t-digest topic, visit your repo's landing page and select "manage topics."