new draft for Mark

docs/cheatsheet2.md

@@ -0,0 +1,101 @@
+# SLpipeline cheat sheet (April 2024)
+
+We outline the steps to reduce data with LMTOY's SLpipeline on Unity.
+
+We assume LMTOY is present and up to date.   This command
+
+      lmtoy
+
+will inform you of some important parameters for your LMTOY.
+
+We are also assuming you are working on a particular project, which
+we will set as an environment variable, for example
+
+      PID=2024-S1-MX-2
+
+1. Go to the *script generator* directory for this project:
+
+         cd $WORK_LMT/lmtoy_run/lmtoy_$PID
+
+   This should have been set up for you. If not, contact PeterT.
+
+
+2. Inspect the **README.md**, **comments.txt** and **mk_runs.py** files for human and programmatic
+   information, and keep them up to date.
+
+   There should be more verbiage here what's in the *mk_runs.py** file....
+
+
+3. When **mk_runs.py** has been edited, make *run* files for Unity's **sbatch** system
+   using this command:
+
+         make runs
+
+   The most complex ones are two-IF SEQ, where it creates a run1a, run1b and run1c file.
+   These refer to a fresh start for both banks (run1a), then a specific run for bank=0 (run1b)
+   and finally a specific run for bank=1 (run1c).
+
+         sbatch_lmtoy.sh $PID.run1a
+	 
+         <patiently wait for all the runs to finish>
+      
+         sbatch_lmtoy.sh $PID.run1b
+
+         <patiently wait for all the runs to finish>
+      
+         sbatch_lmtoy.sh $PID.run1c
+
+   When only one bank is present, only run1a and run1b is needed. For RSR data only run1a is needed.
+
+4. When all these single obsnums are finished, make a summary
+
+         make summary
+
+   The results can be viewed on
+
+         http://taps.lmtgtm.org/lmtslr/$PID
+
+5. Now run the combination using the run2 script(s). Typically you only need
+
+         
+         sbatch_lmtoy.sh $PID.run2a
+
+   It will do both banks (if you have two IF's)
+
+6. Again do a summary and view the results
+
+         make summary
+
+## Advanced use:
+
+1. Mosaiced fields
+
+Example:  2024-S1-MX-2
+
+2. Combining data from different projects
+
+Example: 2024-S1-UM-1 and 2023-S1-UM-8  (notably their pointing center is different!!!)
+
+
+3. Removing all your jobs from the sbatch queue:
+
+         scancel $(squeue --me | tail +2 | awk '{print $1}')
+
+4. Find out if there are new obsnums for a specific project
+
+         source_obsnum.sh $PID
+
+    If this looks different from what you have in your mk_runs.py file,
+    for example new sources and/or new obsnums, add them to the file
+    and make new run files and sbatch them .
+
+    You can also manually search the raw data, for example
+
+         lmtinfo.py grep 2024-03-17 Science 
+         lmtinfo.py grep 2024-03-17 LineCheck Bs
+
+
+   pipeline parameters).  
+
+
+