This repository contains a number of scripts enabling the end-to-end reduction of NIRISS/WFSS data.
A minimal installation environment can be found in environment.yml, to be used with any conda-like package manager (tested with micromamba). As of writing, this contained all of the necessary requirements to run through Stage 1 (reduction, extraction, and redshift fitting). After initialising the environment, it is also necessary to run setup_vars.sh, which adds the necessary environment variables to the bash shell, and downloads the latest trace/wavelength configuration files from several Zenodo repositories.
By default, the grizli and CRDS configuration files will be stored under the user's home directory. If an alternative location is preferred, modify line 3 of setup_vars.sh, specifically CONF_DIRS="${HOME}/conf_dirs" to point to the preferred directory.
This is split into three separate parts, with an additional script, offline_steps.py, that may be necessary on some clusters.
If (as is common on a number of HPC setups), the compute nodes themselves do not have external network access, two additional steps are necessary.
The first is to change a keyword in the TOML config, under [general], setting process_offline = true. This will check if the required files are present, and end the processes if not. Much better to find out at the start than after getting halfway through reducing a field.
The second is to run the offline_steps.py script, passing the same configuration file as for the other steps. This queries CRDS for the necessary reference files and downloads them, as well as constructing a reference astrometric catalogue in order to align the direct imaging to an external reference.
This takes the *uncal.fits files from MAST, and returns the *GrismFLT.fits used by grizli.
This runs the beam extraction and redshift fitting. Recommended to run this using mpirun to take advantage of parallel processing.
TBD.