search

SainsburyWellcomeCentre / aeon_experiments

Experiment workflows for Project Aeon
BSD 3-Clause "New" or "Revised" License
2 stars 0 forks source link

Double 0's still show up in BlockState data #497

Closed jkbhagatio closed 1 month ago

jkbhagatio commented 5 months ago

e.g.

image

jkbhagatio commented 5 months ago

A similar issue occurs for Patch State values during block transitions as well:

2c1d4ebe-1dab-44f1-82e4-1d78b8b9ee75

(for "Z:\aeon\data\raw\AEON3\social0.2\2024-02-05T15-43-07\Patch1\Patch1_State_2024-02-05T15-00-00.csv")

@glopesdev suggested:

"I’m thinking it might be wise to wait until the next heartbeat to do block transitions so everything has time to flush out So basically block transition would be “stop patches” > “wait heartbeat” > “resume”."

jkbhagatio commented 5 months ago

Ah, another issue in doing block analysis is that sometimes the patch state stream updates before the block state stream on block transition. So another reason to wait for heartbeats to do the transition

jkbhagatio commented 2 months ago

@glopesdev Duplicated in aeon_mecha here: https://github.com/SainsburyWellcomeCentre/aeon_mecha/issues/358

glopesdev commented 1 month ago

Fix is being deployed but still needs testing, so reopening for a little longer.