Not matching against all encounters in database?

What Wildbook are you working in?


What is the entire URL out of the browser, exactly where the error occurred?

Can you describe what the issue is you’re experiencing?

When running matches on our newer photos from 2020 on, I’m concerned that they are only being matched against photos from that time period and not against any of our photos that have been matched and identified before 2020. The system says that matches are being run against 2327 candidates which is the number of new photos that we had recently uploaded. This would be a major issue since we have hundreds of photos since 2012 and I can’t tell if any of them are being matched against when running new matches, since none of our previously identified individuals from before 2020 are coming up when I run a match. Can you tell what photos are being used to run matches against? If photos from before 2020 are not being run through the system, I could be marking individuals with a new ID when they were actually seen before and the system just isn’t picking them up.

Hopefully this makes sense, please let me know if there is anything I need to clarify here. In summary, I’m concerned that when I run a match, it isn’t matching against ALL of the photos in our database. Thank you!

Can you provide steps on how to reproduce what you’re experiencing?

Running a match on any of the photos from 2020 to the present will reproduce the issue that I’m concerned about.

If this is a bulk import report, send the spreadsheet to with the email subject line matching your bug report

Hi @UR-Stelle

I replicated this and also see numbers lower than expected. I am investigating.


Hi @UR-Stelle

I found a large batch of your images that were not registered with machine learning and have pushed them through the detector. This likely makes up for the gap your were seeing in what we were matching against. I have kicked off a new matching job that should prime the finFindR and CurvRank algorithms and have them extract the new image trailing edges. The Flukebook machine learning queue has about 60 jobs in it, and after that this new job should kick off and then take a few hours. However, we should be ready for expanded matching on your catalog by tomorrow.

Thank you for reporting this and noting this dicrepancy in data volume! Good catch!


1 Like

Hi @UR-Stelle

Matching for Dana Point is now against 3250 other fin annotations. Here is an example matching job:

Thank you for bringing this to my attention. I believe we have unblocked all of your data now, but please let me know if you see other examples from any other accounts associated with your lab.

Thank you,

1 Like

Awesome, thank you so much for all of your help! I’m so glad you were able to clear this up. I really appreciate it!