git clone [email protected]:scikit-learn-inria-fondation/pdist_aggregation.git
cd pdist_aggregation
makeSimply:
git clone [email protected]:scikit-learn-inria-fondation/pdist_aggregation.git
cd pdist_aggregation
conda env create -f environment.yml
conda activate pdist_aggregationSee:
make helpYou can adapt the benchmarks' configuration editing
benchmarks/config.yml.
And then simply run the benchmark script:
make benchmark-parallelPDF reports will be written in a subfolder in results.
The implementation can be capped to the sequential execution using:
make benchmark-sequential⚠ Currently this make target has been written for GNU/Linux as it makes uses of
taskset(1)but you can adapt it easily using environment variables.