Rubin ToO Workshop Tutorial

This folder contains a guided Jupyter tutorial for simulating a single Rubin target-of-opportunity kilonova follow-up case.

Start Here

Open rubin_too_workshop.ipynb and run the notebook from top to bottom. The notebook is structured as a workshop tutorial with setup notes, editable parameters, and two example runs:

• 100 Mpc: baseline event distance
• 200 Mpc: same event moved farther away

The notebook writes new products to outputs/ so participants can rerun the tutorial without overwriting the provided example files.

Files

• rubin_too_workshop.ipynb: main tutorial notebook
• S251112cm_obs_vanilla.h5: Rubin visit table used as the simulation input

Create the Conda Environment

From this folder, run:

conda env create -f environment.yml
conda activate rubin-too-workshop
python -m ipykernel install --user --name rubin-too-workshop --display-name "Rubin ToO Workshop"

Then open rubin_too_workshop.ipynb and choose the Rubin ToO Workshop kernel.

The notebook sets NUMBA_DISABLE_JIT=1 before importing redback to avoid Numba cache issues on workshop machines. It uses the Bulla BNS kilonova model through redback-surrogates, so the environment pins numpy=1.26.4 for Torch/kilonovanet compatibility.

Expected Python Packages

The tutorial expects a Python/Jupyter environment with:

• numpy
• pandas
• matplotlib
• astropy
• lightcurvelynx
• redback
• tables or another pandas-compatible HDF5 backend