Surface Albedo Mapping Using RotAtional Inversion (SAMURAI
)
A1. Set the simulation parameters in: fitlc_params.py
B1. Run the mcmc code:
python main_fitlc_mcmc_EPOXI_NEW.py
This will create a directory for this unique run (labeled by timestamp). The output HDF5 file containing the mcmc chains and data can be found there.
A2. Set the simulation parameters in: map_EPOXI_params.py
B2. Run the mcmc code:
python main_map_EPOXI_mcmc.py
Essentially the same output as the lightcurve fitting method.
C. Analyze mcmc output:
python mcmc_analysis.py -d <run_directory> trace
This creates 'trace' plots for the trajectories of all mcmc chains through each parameter's space (with trace
keyword). It is helpful to inspect the trace plots to determine the 'burn-in' by eye.
D. Perform physical analysis:
python mcmc_physical.py -d <run_directory> -b <burn-in index> posterior area-alb model-data
This will generate histograms of the posterior distrubution for each parameter (with posterior
keyword), a plot showing the retrieved area and albedos (with area-alb
keyword), and a plot showing the model fit to the data (with model-data
keyword).