Difference between revisions of "Analysis software wishlist from May 2020 CM"
From GlueXWiki
(→Amplitude Analysis) |
(→Event Generation) |
||
Line 2: | Line 2: | ||
=== Event Generation === | === Event Generation === | ||
− | * have flat mass generation be actually flat (currently increases at high mass in many channels) - fixed, | + | * have flat mass generation be actually flat (currently increases at high mass in many channels) - fixed, pull request from Alex A.: [https://github.com/JeffersonLab/halld_sim/pull/131] |
* correctly handle unstable particles at the lower vertex | * correctly handle unstable particles at the lower vertex | ||
* correctly throw events flat in angles, like in omega pi system | * correctly throw events flat in angles, like in omega pi system | ||
* consistent plan for handling different models of t-dependency | * consistent plan for handling different models of t-dependency | ||
− | |||
=== Amplitude Analysis === | === Amplitude Analysis === |
Latest revision as of 10:58, 18 May 2020
Event Generation
- have flat mass generation be actually flat (currently increases at high mass in many channels) - fixed, pull request from Alex A.: [1]
- correctly handle unstable particles at the lower vertex
- correctly throw events flat in angles, like in omega pi system
- consistent plan for handling different models of t-dependency
Amplitude Analysis
- Standard way of generating flat trees from DSelector output
- ROOT scripts?
- Update and generalize tree_to_amptools?
- Benchmark against analysis launch?
- New AmpTools data reader?
- DSelector flat tree output?
- Look at FlattenFSRoot and adapt data reader?
- Slice up data and perform different fits
- bins of mass
- bins of t
- different wave sets
- different starting parameters
- resampling for bootstrap error estimation
- Output format
- text files?
- SQLite DB?
- Plotting
- Generic GUI program?
- command line program?
- easy interface to results based on different fit variations
- visualize both amplitudes and comparison between data and weighted MC
- Common routines/procedures for error propagation