West-call/051722.md

@@ -16,6 +16,7 @@ Char / Minute Taker / Focus Session [Rota](https://git.ligo.org/groups/gstlal/-/
 
 ## Action Items for next week
 (**Leave action items here for next week.**)
+  - set up focus session for HM, OSG
 
 ## Agenda / Minutes
 
@@ -50,3 +51,59 @@ Char / Minute Taker / Focus Session [Rota](https://git.ligo.org/groups/gstlal/-/
 * AOB
 
 ## Chat log
+* Announcements (5 minutes)
+  - Next week is F2F so we won't have this call. The week afterwards, the focus session is low latency
+* Last week's call  
+  - none
+* Last week's East call - [agenda](https://wiki.ligo.org/CBC/EastAgenda20220510)
+  - MR for moving things into `gstlal-ugly` wasn't merged. 
+* Quick updates (45 minutes)
+  - Operations (5 minutes)
+    - LL CBC operations
+      - Plans for MDC05 (06/21):
+        - p(astro)
+          - hear later in the afternoon call
+        - new template bank
+          - no update
+        - idq online
+          - Rachael: focus session
+    - Offline CBC operations
+      - Offline analysis finished. VT not as good as O3. Inj missed/found table is weird.
+    - LL IDQ operations
+       - Rachael: update during focus session
+  - O4 Dev (30 minutes)
+    - Low latency integrated testing and Monitoring
+      - Becca: working on updates to test suite for state vector jobs for each ifos. Implement by the end of the week. 
+    - Template bank
+      - No update
+    - Likelihood ratio, background and foreground sampling
+      - Leo: set up new class and running sampling job to get histogram to compare new class and old class
+      - Leo: chisq study by Andre shows a good fit of snr time series and chisq. https://dcc.ligo.org/DocDB/0182/L2200039/001/Presentation%205-13-22.pdf 
+      - Leo: dtdphi Jacobean: Ryan tried reranking early warning tests. Didn’t do good job. Asked Shomik to work on it to check and do more tests
+    - Optimization and throughput benchmarking
+      - Alex: no update
+    - Data format wrangling
+      - Chad isn't here
+    - Enabling running on OSG/IGWN grid
+      - Cort: reading about osg/igwn grid
+    - DQ dev
+      - Rachael: no update
+    - HM search
+      - Stefano: almost there for precessing templates. Do some runs. Decide which parameter space we’re interested in and what parameters. Focus session for Stefano after chisq? 
+  - Exploratory development (5 minutes)
+    - no update
+  - Misc projects
+    - no update
+* Focus session
+  - idq - Rachael, Patrick [slides](https://docs.google.com/presentation/d/1nxPVfxevWP_t0emtACTVEwRQDnG_RCv3ucn1_06y4fw/edit?usp=sharing)
+    - cycling issues were seen and is now improved: cycling issues are where iDQ repeats itself and seeing edge effects. Happening because calibration wasn’t given enough data. Improved now after giving more data to calibration. 
+    - slide 4: the cause for the dip around ln4 isn't known, but isn't concerning. Isn't a general feature, it's seen in this particular chunk (chunk34). The window size is 10. 
+    - slide 5: the changes here have decreased the run time by 10~15% and memory usage was decreased. Waiting for Virgo. 
+    -  slide 8: Difficult to do the percentage stuff for online. Instead, use integrating over temp duration for iDQ time series. We get info over the temp duration when integrating over temp duration. Launch a test analysis running this next week or so.
+    - Questions:
+      - Leo: is the calculation gonna be part of LR calculation or a separate job running in parallel?
+        - Rachael: iDQ won’t need it’s own job. Integrated into inspiral_lr and gstlal_inspiral to ensure iDQ is there. 
+        - Patrick: it’ll work like the horizon distance
+      - Ryan: are there plots like this for the new iDQ? 
+        - Patrick: there’s 3 different plots, one for FAP 1e-3, 3e-3, 1e-2 -> https://git.ligo.org/idq/review/-/blob/master/O3/veto.md. 
+* AOB
\ No newline at end of file