Artificial Intelligence

Thanks - but not happy to download to my PC.
Quite content now that I have tested these - remember iNat has many iffy RG identifications - I’d prefer to leave them alone.

It seems that, at last, the PlantNet Contributions are sending out Changes Flags - about time and thanks
@miked @Chris_Valentine
My first was today

2 Likes

Yes, it’s been happening for a few days now.

I have just noticed them appearing in the tracker, so we now know when another seaweed blunder has appeared without even opening the observation.

To be fair, most of the birds have been correct - with one or two glaring exceptions!

The birds and mammals AI is currently broke, I’m afraid. It’ll have to wait until Jan for me to fix it.

Help needed, please - I thought that I should begin the new year by contributing to the WORLD FLORA of Pl@ntnet.
Added a Cape Point species that I was certain of, but just one picture as a test. Managed to get it posted

but now wondering if I could add more views- but can’t see how to do that.

Also. I thought it might be good to add a link to my iSpot post

Any suggestions/comments?
Good Idea/bad idea?
@miked @Chris_Valentine

I am not sure but I (or indeed you) could ask the plantnet team. If you can’t find their email then I have their individual ones and could pass on your message.

Thanks Mike. I tried some others today, with no success. They have masses of WORLD observations posted, some might be red herrings and in my mind it may just become a muddle of inaccurate identifications, like RG.
Think I might just put things on hold and rather continue adding to, and refining my iSpot collection.
iSpot has so many strengths.
Please pass on to their team if you wish.

The FASTCAT AI (UK birds and mammals) scripts are finally working again and those users whose observations get a comment should now be notified via their Activity Tracker. I can certainly see plenty of notifications in the Changes Tracker.

Thanks but two whole yellow pages in my tracker this morning. I should stop adding agreements to birds!
Good luck with the upgrade tomorrow Chris

There had been a backlog of observations going back to 12th December - you won’t usually get so many together.

Hi Marlandza,

we are glad that you would like to share your data with us. To contribute multiple views of the same plant, you must do so when creating the observation. If you have already created the observation with only one view, I’m afraid you’ll have to delete it and re-create it with all views (sorry about that…). If you have a very large number of observations, we might be able to import them automatically for you. I don’t know if it is possible to export them from your iSPot account for example?

Best,
Alexis Joly

3.5 pages for me!! But I like to acknowledge the efforts that people have made to submit records.

I checked about 20 bird IDs by Fastcat this morning and all were correct - which is impressive. Quite tempting to post some of my rather more ‘cryptic’ bird photos to see how it copes!!

The birds identifier does seem to work extremely well. Its more difficult with the mammals because the primary image is often that of traces of the animal, including remains!

Thanks for your kind response. I do have a fair amount of Cape Plants and would love to share them - however I am reticent of sharing any when I am not really sure that the name is correct.
I am beginning to get the hang of what I need to do to submit - struggled with the location at first.
It’s an enormous task that you are taking on - best wishes with it’s success.
M

Hi again all especially @miked and @Chris_Valentine and @ajoly

I am wanting to know why no Plantnet panel appeared on this obs.

As you’ll see, to add interest to a more pedestrian obs of Sycamore in winter, I ran the first image through Seek/iNat and Pl@ntnet phone app. I wanted to see what the Plantnet AI on this site would say as well. But no Plantnet comment has appeared.

Thanks for the advice Alexis - have deleted and added pics and location - hope this works

Marland

That’s a good question. The Pl@net script has to date been the most reliable, but I have had to do quite a bit on all of the AI scripts recently so I can only guess that particular observation didn’t get processed. Sorry.