Added the submission #10
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
name: CI | |
on: | |
push: | |
# Sequence of patterns matched against refs/heads | |
branches: | |
# Push events on main branch | |
- paper-sisap24-indexing-challenge | |
# Sequence of patterns matched against refs/tags | |
tags: "*" | |
jobs: | |
test: | |
name: ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }} - ${{ github.event_name }} | |
runs-on: ${{ matrix.os }} | |
strategy: | |
fail-fast: false | |
matrix: | |
version: | |
- "1.10" | |
os: | |
- ubuntu-latest | |
arch: | |
- x64 | |
exclude: | |
- os: macOS-latest | |
arch: x86 | |
python-version: ["3.11"] | |
steps: | |
- uses: actions/checkout@v4.1.7 | |
with: | |
submodules: "true" | |
- name: Set up Python ${{ matrix.python-version }} | |
uses: actions/setup-python@v5 | |
with: | |
python-version: ${{ matrix.python-version }} | |
- name: Install dependencies for downloading the dataset | |
run: | | |
sudo apt-get install curl libcurl4-openssl-dev | |
- name: Download database and queries | |
if: steps.cache-data2024.outputs.cache-hit != 'true' | |
env: | |
DBSIZE: 300K | |
run: | | |
mkdir data2024 | |
cd data2024 | |
curl -O https://sisap-23-challenge.s3.amazonaws.com/SISAP23-Challenge/laion2B-en-clip768v2-n=$DBSIZE.h5 | |
curl -O http://ingeotec.mx/~sadit/sisap2024-data/public-queries-2024-laion2B-en-clip768v2-n=10k.h5 | |
curl -O http://ingeotec.mx/~sadit/sisap2024-data/gold-standard-dbsize=$DBSIZE--public-queries-2024-laion2B-en-clip768v2-n=10k.h5 | |
cd .. | |
- uses: conda-incubator/setup-miniconda@v3 | |
with: | |
auto-update-conda: true | |
python-version: ${{ matrix.python-version }} | |
- name: Install LMI dependencies | |
shell: bash -el {0} | |
run: | | |
conda create -n lmi -y python=3.11 | |
conda activate lmi | |
conda install -c pytorch -y faiss-cpu=1.8.0 | |
conda install h5py=3.11.0 | |
pip install --no-cache-dir numpy==1.26.4 tqdm==4.66.4 loguru==0.7.2 scikit-learn==1.5.1 | |
pip install --no-cache-dir torch==2.3.1 --index-url https://download.pytorch.org/whl/cpu | |
- name: Benchmark LMI | |
shell: bash -el {0} | |
run: | | |
pwd | |
ls -l | |
ls -l data2024 | |
conda activate lmi | |
# Parameters for 300K: | |
python3 task1.py --dataset-size ${DBSIZE} --sample-size 100000 --chunk-size 100000 &>task1.log | |
python3 task2.py --dataset-size ${DBSIZE} --sample-size 100000 --chunk-size 100000 &>task2.log | |
python3 task3.py --dataset-size ${DBSIZE} --sample-size 100000 --chunk-size 100000 &>task3.log | |
# Parameters for 100M: | |
# python3 task1.py &>task1.log | |
# python3 task2.py &>task2.log | |
# python3 task3.py &>task3.log | |
python3 eval.py --results result res.csv | |
- uses: actions/upload-artifact@v4.3.4 | |
with: | |
name: Results | |
path: | | |
res.csv | |
task1.log | |
task2.log | |
task3.log |