poetry install
The experiments rely on Weights & Biases (wandb) for experiment tracking.
Launching goes in two steps. First, generate the bash scripts.
argmapper sweep exp/config/tabular_dp_experiments.yml --out exp/config/tabular_dp_experiments.runfile
argmapper sweep exp/config/cifar10_dp_experiments.yml --out exp/config/tabular_dp_experiments.runfile
Next, launch the commands in the generated bash scripts in any way you prefer. You can use:
argmapper launch exp/config/tabular_dp_experiments.runfile
argmapper launch exp/config/cifar10_dp_experiments.runfile
Keep in mind you need to have wandb set up and running so that the experiment metadata are saved.
Analyses of the experiments are done using Jupyter notebooks in the notebooks
directory.
They are saved as Python scripts in git. Use jupytext <notebook>.py --to ipynb
to regenerate the
notebook *.ipynb
. Other than the analysis of experiments, the notebooks directory contains
standalone experiments with synthetic data and theoretical expressions.