Deleting 3 large import batches of ID'd individuals used to clean training data

Support
,
1

What Wildbook are you working in? ACW

What is the entire URL out of the browser, exactly?
Batch #1 (2150 encounters of 77 individuals): Wildbook for Carnivores | Login
Batch #2 (573 encounters of 60 individuals): Wildbook for Carnivores | Login
Batch #3 (144 encounters of 10 individuals): Wildbook for Carnivores | Login

In order to create a clean hyaena training dataset for PIE v2, we uploaded 3 large batches of ID’d individuals and ran matching to ensure all annotations were correctly assigned.

But each batch was only a partial upload from each organization’s full ID’d dataset. They now want to upload the full ID dataset. Which is great but I’d like to avoid having to figure out which images we kept and which we excluded from the upload.
So I’d like to delete the entire previous upload batch and re-upload the full dataset.

But my concern is that I think I’m supposed to remove all assigned encounters from each of the individuals before I can delete the import task - is that correct?

If so, pleeeeeease, is there any other way???

Any and all assistance and advice is VERY much appreciated!

thanks
Maureen

@MarkF - is there a quick answer here by any chance? Can I delete an import task that has encounters assigned to ID’d individuals? I’m worried that I’m going to lose access to these imports shortly as they’re about 6mths old now…

Thanks!
M

2

Hi, @ACWadmin1 !
Deletion of a bulk import task should remove those encounters from individuals.
If an individual ends up with 0 encounters after that, the individual will be removed, too.

Does this answer the question you’re asking?
Thanks,
Mark

3

@ACWadmin1 as a follow up thought: you might could do a “dummy” bulk import with like one encounter or something, attach it to an individual, and then delete the bulk import just to confirm that things are working as they should before doing a large deletion.

Either way should be fine, but I that’s what I might do. Paranoid? Absolutely. But not a bad idea.

-Mark

1 Like
4

@MarkF - thanks so much and yes, I’ll do a test - good thinking!

best
Maureen

1 Like