allow input FITRES file option to select list of CIDs to include in training or predict mode
replaced hard-coded scone dictionary of INTEGER:TYPE with parsing data file header to get this information; may require minor update in SNANA sim to write more parsable information.
for the create_heatmap script names, use 5 digit numbers with padded zeros; e.g.,
create_heatmaps__00154.sh
output separate log per create_heatmap script (easier to trace problems and understand what's happening)
top of each log file should include full scone command (written by scone) so that anyone can cut-and-paste the command to rerun the job
end of each log file should indicate a clear SUCCESS or FAIL that pippin can read and report number of failures.
if all heatmap jobs are successful, pippin should backup the .sh and .log files into a single BACKUP_create_heatmap.tar (to reduce file counts); this would be a pippin task
autogenerated input files should include comment line(s) indicating the provenance of the file; e.g., # autogenerated input from scone [or pippin] to do XYZ task
env setup should put scone scripts in path to avoid explicit need for "cd /codeDir" to run scone codes.
each scone script should start with
!/usr/bin/env python
so that there is no need to specify "python [codeName]" and instead typing [codeName] is sufficient (if it's in the path)
helenqu
commented
11 months ago
!/usr/bin/env python
so that there is no need to specify "python [codeName]" and instead typing [codeName] is sufficient (if it's in the path)