Can you describe what the issue is you’re experiencing?
Firstly: After starting the identification of 86 bulk import images more than two weeks ago the status of all images is still “waiting to queue” (in column “Match Results by Class”).
Another bulk import was started about a week later (102). The second bulk import contained three images that needed to be annotated manually. Those three images were the only ones that were identified, while the others were stuck at “waiting to queue”. When I removed an automatical annotation from an image and added a manual one the image was processed. What is the reason for the images being stuck at “waiting to queue” for over two weeks?
Secondly: What does this error message mean?: " there was an error with task 982072e6-f796-4201-a77d-325a778f593e: No data to match against, please refine your matching parameters." (URL: Amphibian and Reptile Wildbook)
Can you provide steps on how to reproduce what you’re experiencing?
Firstly: Doing a bulk import and sending it to identification.
Secondly: clicking on the link “yellow_bellied_toad: waiting to queue” in the bulk import.
If this is a bulk import report, send the spreadsheet to services@wildme.org with the email subject line matching your bug report
Wildbook looks for matches based on the species, location ID, and viewpoint of the animal. It looks like something went wrong during detection where the viewpoint couldn’t be determined because all 86 encounters have a “null” viewpoint:
Additionally, you’ve used a Location ID that doesn’t exist in ARW (“HKB”). Your location ID has to be one of the options in the Location ID drop-down menu of the Report an encounter page. You can request new location IDs by creating a new Feature Request post here and filling out the template details.
If the automatic annotation was missing a viewpoint, deleting it and manually creating a new one allowed you to assign a viewpoint so it could proceed to identification.
The encounter was imported with the location ID set to “HKB” which doesn’t exist in ARW. Because it can’t find this location ID, there is nothing to match against.
My recommendation is to delete your bulk imports and update your spreadsheets with a different location ID that’s in ARW and try importing them again.
We’re not likely to approve new location IDs that are shortened as acronyms because we need them to be readable by other users of the Wildbook, so if you want to post a feature request for this location, its name will need to be spelled out.
I have another question
The situation is that I was already further in progress with this bulk import, the upload was finished and I was in the process of matching the images (I was also halfway through there). But then came the last update of ARW and all matching-actions in this project did not load anymore. So I thought it makes sense to delete it and upload it again from scratch. This didn’t work, which you already know.
I then noticed that in aother project (ff59db78-df63-437a-b1fe-88a41b821835) that I had already worked through 100% after the update, new errors also occurred in this project… The problem here seems to be an error in the identification status, but everything was fine before the update.
Could this error also have something to do with the viewpoint of the animal? Or is this a different error?
Seeing this message after importing typically indicates a problem with the spreadsheet itself. Can you email it to services@wildme.org so I can take a look?
I think this might be weirdness with the update. Let me discuss with the devs and get back to you.
That’s interesting, because when I enter “HKW” in the Encouter Search I find “Hils-Klinkerwerk”, so I thought that the ID was taken. So I better use the full name “Hils-Klinkerwerk” in my spreadsheet?
You know what? You’re absolutely right. I had forgotten that we haven’t yet standardized the location ID codes in ARW according to this ticket: Cleanup for mismatched name and id fields in locationID.json · Issue #711 · WildMeOrg/Wildbook · GitHub
I got pulled into another project for Wildbook upgrades and this location ID standardization task was left unfinished. For context, in the location ID code for the site, the id and name lines are supposed to match. The name is what appears in the drop-down menu of the site, but what the image analysis workflow actually uses to determine the location is what’s in the id part. Here’s what that bit of the code looks like for this location ID:
So in short; carry on! HKW is the correct way to enter this location in an import. I’m sorry I didn’t recall this sooner. When ARW does get this update made to the location ID file, we’ll be sure to post a notice on the site/Community.
I was able to get your spreadsheet committed and appearing correctly on my end. This has come up before as a result of Excel’s automatic formatting and is usually resolved by saving the file as a .csv and then saving that .csv file as .xlsx. That should make the error message disappear.
The devs confirmed your missing project matches are a bug. Because yellow-bellied toads are only identified by their bellies, ARW didn’t bother configuring the viewpoint on automatic detections since it can only work from one viewpoint. When the upgrade happened, it was looking for a viewpoint configuration (since most Wildbooks need one to work) and when it found that one wasn’t set, it couldn’t complete identification. We’re working on this and will let you know when it’s been fixed.
Thanks for your patience! Yellow-bellied toads for your Coppengrave projects should be displaying match results now. Can you let me know if you continue to see issues with detection or ID in ARW?
I have now also been able to comittetd the Excel table. It didn’t work for me to save the file as .csv and then save the .csv file as .xlsx. But I thought, because you said it was due to the spreadsheet, that I would just copy the contents of my spreadsheet and paste it into the “WildbookStandardFormat.xlsx” from the Bulk Import Instructions page and then upload it. Fortunately, this worked without error. The identification has now also worked well and matches are displayed.
Thanks for your help
Thank you for solving the problem so quickly. The results for the Coppengrave project are displayed again and there is no longer an identification status error in any encouter.
Just to clarify, because you have fixed the bug: If I now have to set a viwepoint when I manually annotate after the update, will any yellow-bellied toad with that set viewpoint still be matched with all other yellow-bellied toads with an automatic annotation where the viewpoint is automatically set to = null/zero?
@Mirjam I wanted to let you know that I updated the location ID file in ARW today. Now the location ID names in the drop-down menu of the Report an encounter page match how the location names should be entered in a bulk import spreadsheet.
So in the example from earlier, you now want to enter “Hils-Klinkerwerk” instead of “HKW” for future imports.
