Sharkbook - cannot reset spots

In which Wildbook did the issue occur? Sharkbook

What operating system were you using? (eg. MacOS 10.15.3) Windowns

What web browser were you using? (eg. Chrome 79) Firefox

What is your role on the site? (admin, researcher, etc) Admin

What happened? I found a shark in the system where the researcher had added incorrect spots. I went to remove the spots and reset, but the system does not appear to be correctly resetting the spots. In the matches it keeps reverting back to the first set of spots that were set incorrectly

What did you expect to happen? I thought I could remove the spots, reset the spot mapping, and review groth results based on new spots, but instead it is showing me matches based on the old crappy spots set.

What are some steps we could take to reproduce the issue? Review: Sharkbook: Wildbook for Sharks and remove spots, reset spots, and review groth results. You will see that the groth results do not appear to have reset.

If this is a bulk import report, send the spreadsheet to services@wildme.org with the email subject line matching your bug report

I’m seeing this, too. I can see on the Encounter that 12 spots were added, but the results still show the original placement of the 6 spots. I’ll write a ticket for this and I’ll post the link here when it’s ready.

Here’s the ticket link: Spot map results page displays cached spot maps · Issue #394 · WildMeOrg/Wildbook · GitHub

Any progress updates will be posted there.

Hi @Sarah_Aus_SpotaShark

In researching this more, it’s become apparent that it will take more time to try and fix Spot Mapping than it will to train MiewID on sand tiger sharks and eliminate the need to spot map altogether. I’m going to send you an email later today with the details so you can help us with that.

Wow - ok, well then let us try that! This sounds like a big task so it might be good to arrange a call with you next week to discuss this further?

1 Like

Hi Anastasia,
I have been going through the data to try and clean as much as possible and find a dataset for you to start working with to get the MiewID algorithms trained, and would like to start by offering this list.

If you need more sharks I can keep going and add more, but this hopefully gives you almost 300 shark individuals with 5 or more encounters each to work with.

Thanks
Sarah

Wonderful; thank you! I’ll pass it along to our machine learning team. :smile: