[Neurips23] NLE submission for the sparse track #297

	# Contributed by @GuilhemN in https://github.com/erikbern/ann-benchmarks/pull/233
	name: Billion-Scale ANN Benchmarks, NeurIPS 2023

	on: [push, pull_request]

	jobs:
	build:
	runs-on: ubuntu-latest
	strategy:
	matrix:
	include:
	- algorithm: faiss
	dataset: random-filter-s
	track: filter
	- algorithm: linscan
	dataset: sparse-small
	track: sparse
	- algorithm: nle
	dataset: sparse-small
	track: sparse
	- algorithm: diskann
	dataset: random-xs
	track: streaming
	- algorithm: diskann
	dataset: random-xs
	track: ood
	fail-fast: false

	steps:
	- uses: actions/checkout@v2 # Pull the repository

	- name: Set up Python 3.10
	uses: actions/setup-python@v2
	with:
	python-version: '3.10'

	- name: Install dependencies
	run: \|
	pip install -r requirements_py3.10.txt
	python install.py --neurips23track $TRACK --algorithm $ALGO
	env:
	ALGO: ${{ matrix.algorithm }}
	TRACK: ${{ matrix.track }}

	- name: Run the benchmark
	run: \|
	python create_dataset.py --dataset $DATASET
	python run.py --algorithm $ALGO --max-n-algorithms 2 --neurips23track $TRACK --dataset $DATASET --timeout 600
	sudo chmod -R 777 results/
	if [ $TRACK != streaming ]
	then
	python plot.py --dataset $DATASET --neurips23track $TRACK --output plot.png
	fi
	python data_export.py --output test.csv

	env:
	ALGO: ${{ matrix.algorithm}}
	DATASET: ${{ matrix.dataset }}
	TRACK: ${{ matrix.track}}

Provide feedback