The paradigm used for starting hazard calculation jobs is to start from an input file. Make it easy to modify the input args for a calculation (e.g. change the logic trees, locations, vs30s, etc) and run multiple different hazard calculations.
@chrisdicaprio a few things I ran into getting running #30 locally...
[ ] it took a while getting local config file set up correctly. FYI in THS and other projects we use python-dotenv which support .env and standard envvars in a flexible way.
[ ] The docs had a THP_ prefix which need to be removed.
[ ] log messages are sometimes a bit terse/incomplete. If logging what's passed in make sure to include all the values. (e.g. my config wasn't working because I had a wrong IMT in config. It woujld have been helpful t see all the filter args logged in data.load_realisations().
The paradigm used for starting hazard calculation jobs is to start from an input file. Make it easy to modify the input args for a calculation (e.g. change the logic trees, locations, vs30s, etc) and run multiple different hazard calculations.