didn’t see an error in the console, the last message there was:
RECORD EVENT: {"type":"Health Check Viewed","data":{"id":"_______","dataset":"SimpleSymbols2","url":"simplesymbols2","type":"object-detection","workspaceId":"_________"},"opts":{}}
Also FYI I have managed to run into the opposite situation
I think the way I triggered this is that I deleted a project and then recreated it from scratch and somehow the old images still count
When I generate a new version everything is still okay though. It’s annoying that the dataset indicators and train/valid/test indicators in the dataset aren’t accurate, but I suppose I can just look at health check to tell me how many images there are.
CAUGHT ERROR
Object { no: "metadata" }
FirebaseError: Firebase Storage: An unknown error occurred, please check the error payload for server response. (storage/unknown)
Client uploaded to the wrong offset (786432 instead of 1310720).
f errors.ts:82
S error.ts:37
C error.ts:115
l request.ts:184
p backoff.ts:56
d backoff.ts:72
t request.ts:153
^ I did finally get an error when this issue re-occurred.
Interesting; never seen that before! Which browser are you using? I found an obscure reference to this where someone mentioned it can happen in Safari.
Hi @vshesh are you still seeing this issue with Health Check generation? I want to follow up in case it still persists for you, as we haven’t received more of these reports from others in the last week.
From today - Seems like health check and the generation steps all show the same values. Hard reload doesn’t change it. This is not a blocker for me using roboflow (since image generation works correctly), just a potentially confusing experience for the user.