Flukebook identification problem

If this is a bulk import report, send the spreadsheet to services@wildme.org with the e-mail subject corresponding to your bug report.

In which Wildbook did the problem occur?
→ Flukebook

What operating system were you using (e.g. MacOS 10.15.3)?
→ I’m using MacOS 10.15.17

What web browser were you using? (e.g. Chrome 79)
→ I use google chrome

What is your role on the site (administrator, researcher, etc.)?
→ I’m Ioena, an intern with the Caribbean Cetacean Society (I use Louise’s account) in Martinique.

What happened?
–>This is the first time I’ve used Flukebook.
I imported my bulk imports for different species and dates. I managed to detect dolphin dorsal fins, but since I launched the “identification” part, it’s not loading, so I’m at 0 identifications out of 158 encounters… And that’s for all my bulks imported in the last 1 week. None of them load, so I can’t move on to the next steps.

What did you think would happen?
→ I thought that the identification of my encounters would work as the day/week, but they remain stuck at 0.

What steps should I take to reproduce the problem?

Hi @lsimon

Thanks for sending your spreadsheets!

Can you post links to the bulk imports that are stuck in ID so I can look into this further? I didn’t find any whose filename matched either of the spreadsheets you emailed me.

Edit: Nevermind, I just noticed that I’m confusing two different issues from different CCS users sharing your account. Can you also email your spreadsheets to services@wildme.org and link back to this post in your email?

Hi,
Thank you, I’ve e-mailed you my documents.

1 Like

Thanks for the spreadsheets! Make sure that the location ID is entered exactly as it appears in the Location ID drop-down menu of the report an encounter page. Any misspellings or differences in letter case means that each one is treated as its own, unique location and may not display the appropriate match candidates.

My recommendation is to update the spreadsheets with the correct location ID, delete the incorrect bulk imports, and re-import them again with the corrected spreadsheets.

Hello Anastasia, thank you for your reply,

I made such a careless spelling mistake ! I’m going to have to start importing again,

Thank you very much for finding the problem

Have a good day

1 Like

No worries! Eventually, we’ll have location ID validation in bulk import, so it will flag this kind of thing before you commit your import. If you still have problems with ID after updating the location ID, let me know.

Hello

I’ve changed the location on several bulks, I’ve imported them again, but they’re stuck at the detection stage (0 on … encounters).

I’ve looked at my files in excel to see if there’s anything wrong, but I can’t see any obvious problem (the GPS coordinates are either with commas or dots, but in both cases the detection doesn’t start so I don’t think it’s a problem).

So I don’t know whether it’s a Flukebook bug, a long queue or a mistake on my part?

Thanks again

Ioena

Hi @lsimon

The Flukebook queue looks good right now. Usually detections won’t start if the taxonomy was entered incorrectly, but that doesn’t seem to be the issue here. I’ve re-sent the imports to detection. I’ll check on them again later today to make sure they’ve completed.

It doesn’t look like these have made any progress since this morning. I’ll need more time to work on these.

There was an issue in our image analysis pipeline where it wasn’t assigning acmIDs to images during detection. Usually, this is the step in the process that allows the annotation to go through identification. We’re still working on this but here’s a status update on the imports you asked about:

I’ll post here again when I have an update.

All of the imports are ready for ID now; thanks for your patience!