Difference between revisions of "20190923-Analyzer-Mtg"
From PREX Wiki
Jump to navigationJump to search| (2 intermediate revisions by the same user not shown) | |||
| Line 4: | Line 4: | ||
== Logistic information == | == Logistic information == | ||
| − | Meeting room at JLab: ''' | + | Meeting room at JLab: '''CC F226''' |
BlueJeans calling instructions: | BlueJeans calling instructions: | ||
| Line 17: | Line 17: | ||
# Pull requests | # Pull requests | ||
#* #239---Feature-Aggregator Add Regressed | #* #239---Feature-Aggregator Add Regressed | ||
| + | #** Needs some "general cleanup" before merging | ||
#* #247---Feature alarm updates | #* #247---Feature alarm updates | ||
| + | #** Merged | ||
# New feature plans | # New feature plans | ||
#* Continuing development of minirun functionality | #* Continuing development of minirun functionality | ||
#* Data-handlers which can process results from other data-handlers | #* Data-handlers which can process results from other data-handlers | ||
#** I've made some progress, and expect to be committing it this week | #** I've made some progress, and expect to be committing it this week | ||
| + | #** How do we want to check for residual correlation after correction? Done within a second pass in JAPAN, or as a post-processing script? | ||
#* BMW first pass extractor | #* BMW first pass extractor | ||
| − | + | #** Cameron has been in contact with Wouter to get started, will follow up this week | |
=== Prompt analysis toolchain === | === Prompt analysis toolchain === | ||
# Status on the farm | # Status on the farm | ||
| + | #* The analysis is basically ready, but we need to set up directories | ||
| + | #* Also need to set up a procedure to stage outputs to tape. Cameron has checked the volatile location: 1TB reserved, 3TB available. We should ask for an increase | ||
# Path to migration to japan | # Path to migration to japan | ||
| + | #* Needs: Correlation/correction done on mini-run basis, BMW extractor | ||
| + | #* Do we want to do all of the prompt plots again? Certainly for CREX; let's think about it for PREX | ||
| + | #** If we redo them for PREX, we probably should make a new set of directories. Do we have enough space in the web directories? | ||
| + | #* Aggregation | ||
# Dithering analysis status & updates? | # Dithering analysis status & updates? | ||
| + | #* Cameron has done some cleanup of the dithering analysis scripts, and it could be added as a postprocessing step | ||
| + | #* We still would need to decide which supercycles to use to calculate the slopes for a particular run-range | ||
# Lagrange analysis | # Lagrange analysis | ||
| + | #* Tao thinks this probably should be merged with the dithering processing step | ||
| + | #* It calculates sensitivities from dithering data, just as the dithering analysis does. Perhaps have one tool to get all the sensitivities, and then use them in different ways? | ||
| + | #** Wouter makes the comments: Someone should think hard about the assumptions behind this Lagrange multiplier method: 1) assumes an identical Gaussian probability model for both dithering and natural beam motion data, 2) is sensitive to outliers, 3) assumes that all uncertainty is in the dependent variables (think: fits y values against x axis, ignoring uncertainty in all x points)<br />[See] https://arxiv.org/pdf/1008.4686.pdf for a fun read | ||
=== Other issues === | === Other issues === | ||
| + | #* Let's aim at getting everything ready for a respin by October 23. | ||
| + | #* Cameron has skimmed through the dataset. Including the AT and SAMs in the regression, and thus cutting events when those channels have an issue is generally a small loss in events (~1e-4 lost event fraction). Maybe do three sets of regression: main detectors only, with ATs, and with ATs and SAMs. | ||
== Present == | == Present == | ||
| + | Tao, Weibin | ||
| − | + | Phone: Paul K., Cameron, Devi, Robert R., Sakib, Devaki, Wouter | |
| − | Phone: | ||
===Excused=== | ===Excused=== | ||
Latest revision as of 10:55, 23 September 2019
Back to Main Page >> Analyzer_Meeting
previous meeting << >> following meeting
Logistic information
Meeting room at JLab: CC F226
BlueJeans calling instructions: Toll-Free Number (U.S.& Canada): 888-240-2560 International toll number: 408-740-7256 Bluejeans CODE: 385 600 867 Bluejeans link: https://bluejeans.com/385600867
Agenda
Analyzer code updates
- General discussion
- Pull requests
- #239---Feature-Aggregator Add Regressed
- Needs some "general cleanup" before merging
- #247---Feature alarm updates
- Merged
- #239---Feature-Aggregator Add Regressed
- New feature plans
- Continuing development of minirun functionality
- Data-handlers which can process results from other data-handlers
- I've made some progress, and expect to be committing it this week
- How do we want to check for residual correlation after correction? Done within a second pass in JAPAN, or as a post-processing script?
- BMW first pass extractor
- Cameron has been in contact with Wouter to get started, will follow up this week
Prompt analysis toolchain
- Status on the farm
- The analysis is basically ready, but we need to set up directories
- Also need to set up a procedure to stage outputs to tape. Cameron has checked the volatile location: 1TB reserved, 3TB available. We should ask for an increase
- Path to migration to japan
- Needs: Correlation/correction done on mini-run basis, BMW extractor
- Do we want to do all of the prompt plots again? Certainly for CREX; let's think about it for PREX
- If we redo them for PREX, we probably should make a new set of directories. Do we have enough space in the web directories?
- Aggregation
- Dithering analysis status & updates?
- Cameron has done some cleanup of the dithering analysis scripts, and it could be added as a postprocessing step
- We still would need to decide which supercycles to use to calculate the slopes for a particular run-range
- Lagrange analysis
- Tao thinks this probably should be merged with the dithering processing step
- It calculates sensitivities from dithering data, just as the dithering analysis does. Perhaps have one tool to get all the sensitivities, and then use them in different ways?
- Wouter makes the comments: Someone should think hard about the assumptions behind this Lagrange multiplier method: 1) assumes an identical Gaussian probability model for both dithering and natural beam motion data, 2) is sensitive to outliers, 3) assumes that all uncertainty is in the dependent variables (think: fits y values against x axis, ignoring uncertainty in all x points)
[See] https://arxiv.org/pdf/1008.4686.pdf for a fun read
- Wouter makes the comments: Someone should think hard about the assumptions behind this Lagrange multiplier method: 1) assumes an identical Gaussian probability model for both dithering and natural beam motion data, 2) is sensitive to outliers, 3) assumes that all uncertainty is in the dependent variables (think: fits y values against x axis, ignoring uncertainty in all x points)
Other issues
- Let's aim at getting everything ready for a respin by October 23.
- Cameron has skimmed through the dataset. Including the AT and SAMs in the regression, and thus cutting events when those channels have an issue is generally a small loss in events (~1e-4 lost event fraction). Maybe do three sets of regression: main detectors only, with ATs, and with ATs and SAMs.
Present
Tao, Weibin
Phone: Paul K., Cameron, Devi, Robert R., Sakib, Devaki, Wouter
