kn-ts-loop

Performs iterative inference on synthetic or real kilonova lightcurve data by progressively removing early detection points. Evaluates how the inferred timeshift evolves.

NOTE: The --resampling flag is not compatible with Ka2017.

Arguments

--idx: Index of the injection/lightcurve to analyze (default: 0).
--grid-dir: Path to the base grid directory containing your data.
--model: EM model to use (default: Bu2019lm).
--svd-path: Path to the SVD models directory.
--prior-file: Path to the prior file for lightcurve analysis.
--minus-pts: Number of early time points to progressively remove (default: 2).
--add-ul: Adds an upper limit instead of completely removing early points (default: False).
--true-merger-time: True merger time for timeshift calculation in ISOT format.
--nlive: Number of live points for nested sampling (default: 1024).
--resampling: Runs the resampling step via gwem-resampling after EM inferences.
--eos-posterior: Path to EOS posterior probability file for GW samples.
--eos-path: Path to EOS files used for resampling.
--GW-prior: Path to GW prior file for resampling.
--EM-prior: Path to EM prior file for resampling.
--restrict-dist-prior: Restrict the distance prior, will restrict the prior as true_val +/- value.

Example

Iterative inference removing the first 2 points, adding upper limits, and executing resampling:

kn-ts-loop --idx 0 \
           --grid-dir /path/to/KN_grid \
           --model Bu2019lm \
           --svd-path /path/to/svd \
           --prior-file /path/to/EM.prior \
           --minus-pts 2 \
           --add-ul \
           --resampling

The output will be systematically organized inside your grid directory under {grid_dir}/{idx}/minusX/.