Run 76 Analysis Tasks and Status

1. Detector characterization. S104 is a new SiHV prototype and seems to behave differently than the previous one we used, S101. It would of interest to characterize:
   1. Phonon pulse differences with S101: saturation effects (pulse shape variation with amplitude), rise and fall times
   2. Phonon collection efficiency: c/w S101
   3. HV noise characteristics; noise width versus bias
2. (n,gamma) NR calibration
   1. look for signs of (n,gamma) signal coincident with NaI array
   2. look for R68-style signal in SiHV-triggered data

Run 76 data summary

• Tasks on direct path to producing final analysis
  • Data types
    • NaI-triggered coincidence, approx -80 V
      • High NaI trigger window
      • Low NaI trigger window
  • Analysis tasks
• Side tasks

Temporary analysis note

Access to files: I have been storing my attempts at converting the analysis to a Jupyter environment in a GitHub repository titled R76Analysis. Any C or ROOT files are likely also available on cdms2. The main example is Tutorial_Replication, which translates the creation of several figures shown in Matt's Tutorials. The code shown in these videos is in r76_analysis_hvcal.c.

Options for loading ROOT files:

• If you have ROOT installed, you likely already have PyROOT, which is the official implementation. If you are familiar with ROOT already, this is probably the way to go, as it provides Python access to existing ROOT functions.
• ★ uproot is a PyPI package for loading and writing ROOT files in python. It does not provide the other functions of root, but it does not depend on ROOT and is pure python. This is the implementation I use in my own notebooks.
• There are other python packages, such as root_numpy, as well, but I have not used them and therefore cannot provide further comment.

Example Conversions: These are locations of examples for converting certain C/ROOT functions that don't have direct parallels in Python. They can be found in the Tutorial_Replication notebook. The notebook attempts to replicate Matt's Tutorials. The code shown in these videos is in r76_analysis_hvcal.c.

• Chains and Friends: See cell 2. These can be handled by combining them into the same pandas dataframe.
• Aliases: See cell 4. Aliases can be handled a few ways, but the easiest seems to be to add a column to the dataframe with the alias' name. Unfortunately, this does mean that the “alias” is simply an array and will therefore need to be updated whenever its “inputs” are updated. I find this is not frequently an issue. For an alternate implementation, see cell 5.

Basic Steps: Cells cited are in Tutorial_Replication.

1. Import data [Cell 2] to one dataframe per series. If managing multiple series, store these dataframes in a list.
2. Run per-entry calculations [Cell 4] by adding columns to each dataframe for any values you want to calculate in advance.
3. Generate cuts [Cell 5] similarly to the calculations you just made – either by adding columns to each dataframe or by creating lists. These cuts will be boolean arrays for indexing, where True indicates
4. Plot [Cell 6] your data while slicing along the cuts to apply them.
5. Run analysis [Cell 8] on the sliced as you normally would – any fitting or other calculations.

Instructions for getting the singularity container running for work with raw pulses

Meetings on Wednesdays at 2:30 p.m. Central time on Kitty's Zoom.