Info

ANN-Benchmarks is a benchmarking environment for approximate nearest neighbor algorithms search. This website contains the benchmarking results for the SISAP19 submission Benchmarking Nearest Neighbor Search: Influence of Local Intrinsic Dimensionality which is available here. Please visit http://github.com/maumueller/ann-benchmarks-sisap19/ to see the source code and get an overview over used algorithms and datasets.

Benchmarking Results

Results are split by distance measure and dataset. In the bottom, you can find an overview of an algorithm's performance on all datasets. Each dataset is annoted by (k = ...), the number of nearest neighbors an algorithm was supposed to return. The plot shown depicts Recall (the fraction of true nearest neighbors found, on average over all queries) against Queries per second. Clicking on a plot reveils detailled interactive plots, including approximate recall, index size, and build time.

Machine Details

All experiments were run in Docker containers on 2x 14-core Intel Xeon E5-2690v4 (2.60GHz) with 512GB of RAM using Ubuntu 16.10 with kernel 4.4.0.

Raw Data & Configuration

Please find the raw experimental data here. The query sets and ground truth is available with each dataset. All datasets can be found here. The algorithms used the following parameter choices in the experiments.

Benchmarks for Single Queries

Results by Dataset

Distance: Angular

glove-100-angular-diverse (k = 10)

glove-100-angular-easy (k = 10)

glove-100-angular-hard (k = 10)

glove-100-angular-middle (k = 10)

glove-2m-300-angular-diverse (k = 10)

glove-2m-300-angular-easy (k = 10)

glove-2m-300-angular-hard (k = 10)

glove-2m-300-angular-middle (k = 10)

gnews-300-angular-diverse (k = 10)

gnews-300-angular-easy (k = 10)

gnews-300-angular-hard (k = 10)

gnews-300-angular-middle (k = 10)

Distance: Euclidean

fashion-mnist-784-euclidean-diverse (k = 10)

fashion-mnist-784-euclidean-easy (k = 10)

fashion-mnist-784-euclidean-hard (k = 10)

fashion-mnist-784-euclidean-middle (k = 10)

mnist-784-euclidean-diverse (k = 10)

mnist-784-euclidean-easy (k = 10)

mnist-784-euclidean-hard (k = 10)

mnist-784-euclidean-middle (k = 10)

sift-128-euclidean-diverse (k = 10)

sift-128-euclidean-easy (k = 10)

sift-128-euclidean-hard (k = 10)

sift-128-euclidean-middle (k = 10)

Results by Algorithm

annoy

faiss-ivf

hnsw(faiss)

bruteforce-blas

NGT-onng

Contact

ANN-Benchmarks has been developed by Martin Aumueller ([email protected]), Erik Bernhardsson ([email protected]), and Alec Faitfull ([email protected]). Please use Github to submit your implementation or improvements.