Not seeing a way to run detection/identification on a bulk import (email question transfer on behalf of customer)

What Wildbook are you working in?

What is the entire URL out of the browser, exactly where the error occurred?
Not provided

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

I asked whether the images then need to be processed manually by annotation of each image and running the image matching
I understand that there is no automatic ID assignment, and that manual confirmation by the researcher is best practice. What I don’t understand is why the detection model did not add annotations for the images that I bulk imported. I estimate that manually adding the detection, running the ID assignment and then verifying the results would take around 5 minutes per encounter, which equates to around 126 hours for the 1512 encounters that I bulk imported. Unless I am misunderstanding something I did not see an ID option in the bulk import pipeline.

Can you provide steps on how to reproduce what you’re experiencing?
Not provided


Hey Justin,
The info you provided was a little vague, but I think I have an idea of what’s going on. Some background information:

When you upload a bulk import, it does not start processing automatically. An org admin or an admin needs to determine, based on server load, whether the bulk import should run through just detection (what generates the annotations) or through both detection and ID (what generates the potential match results).

If you are someone who is a primary point of contact for your organization with Wild Me, then you can be granted org admin privileges because you have awareness of your organization’s data needs. At present, you do not have that role. If this needs to be updated, I can do that, but will need confirmation.

The current state of your bulk import is that is has fully imported, and is in detection. However, the ml queue indicates that no jobs are running. This likely means that the site got out of sync during your detection job. This typically only happens when we have an outage or if a user goes over the recommended number of encounters in their import. As your import has over the recommended number of encounters (1000), I would assume that it is a combination of the two since this has been running for so long.

Recommended steps:

  1. Delete your current import.
  2. Split your import into half, staying well under the 1000 encounter limit.
  3. Re-upload the two imports.
  4. Either tag a staff member here (myself @tanyastere or @jason would be the best bet) to kick of the ID of the imports OR, if you are in a position to take on the role, we can upgrade you to an org admin and you can kick off the imports yourself.

Let me know if any of this needs clarifying. Otherwise, I will wait for confirmation/direction before we start working through this.