Flag for Taxon: Green Russula (Russula virescens)
| Flagger | Content Author | Content | Reason | Flag Created | Resolved by | Resolution |
|---|---|---|---|---|---|---|
| fungee | Green Russula (Russula virescens) |
two thirds of the observations of this are in E. N. A. where this doesn't exist. Those should just be moved to Russula unless you are able to separate R. parvovirescens from crustosa, modesta, aeruginea, etc., etc. |
Feb. 18, 2021 16:44:23 +0000 | radbackedsalamander |
can't be resolved without manual labor |
Comments
@nschwab Not sure how to deal with this
@radbackedsalamander I believe the only solution is to re-identify all these records. I know that there are a lot of them but it's the way I saw it dealt with in the past. I recently implemented infrageneric classification so it might be useful to know the following:
Russula virescens complex corresponds to Russula subsect. Virescentinae
Russula modesta corresponds to subsect. Modestinae
Russula aeruginea corresponds to subsect. Griseinae
Russula variata corresponds to subsect. Cyanoxanthinae
All of these are placed in subgenus Heterophyllidia which can be used if it can't be identified further.
It has now been fixed. I re-identified all the observations for the area @fungee showd.
👍
Now there are many false identification in North America to correct...
@nschwab how did you fix all the IDs so fast? Is there a way to "blanket ID" a taxon or was it just manually?
@radbackedsalamander You can press Ctrl + Left Click to open the observation in a new tab when clicking on the name on the identifications page. Then you can open multiple of them and go quickly through them.
ah ty.
I have been keeping up with it fairly often, but I am super thankful for anyone's help. I still am not sure why they cannot be all be stopped from being suggested in the first place since they aren't here. Atlas'ed out or what have you.
@fungee I'm currently working on Russula aeruginea already removed over half of MA's observations and almost every north american RG observation
it's tedious but possible
Awesome!
@fungee If I understand well, atlases can work that way.
At least they make for a good way to find incorrect observations:
https://www.inaturalist.org/observations?geoprivacy=open,obscured¬_in_place=8264,8857,8434,6718,6774,8433,7122,8267,8859,7236,6857,8196,8050,8228,7399,8147,7578,8860,7599,7147,8858,8241,7800,7016,7506,8266,7207,7008,7020,7161,7183,8505,8057,8268,8051,8265,6753,7094,6973,10282,10298,10302,10303&taxon_id=125720&verifiable=true
Clarification; is E. N. A = Eastern North America?
I am asking since if that is the case, then quite a few records have to be re-identified
Yeah, it's really super hard to keep on top of this as the AI continues to suggest this based on who knows what.
@fungee It would be nice to have a feature to "block CV suggestions" when they're poisoned with bad IDs until it's fixed.
That would work well, at least block the to species IDs.
Here is the link I'm going to use for this one and the simple statement I have so far. https://www.inaturalist.org/observations/identify?taxon_id=125720&place_id=19,37,18,36,12,44,13,7590,6883,13336,7587,6853,9116,17,41,2,47,49,8,48,51,42,4,97454,39,33,7,31,29,20,32,35,24,38,28,25,36,19,26,45,30,43,23,21,27&per_page=200 "Two-thirds of the observations of this mushroom are in E. N. A. where this doesn't exist. Those should probably just be moved to Russula unless you are able to separate R. parvovirescens from crustosa, modesta, aeruginea, variata, etc. This is part of the push described here https://www.inaturalist.org/journal/fungee/46596-new-ai-computer-vision-first-for-android to fix the Computer Vision Machine Learning that only suggest things that look like species that are named on here. If we eliminate the species named from geographically segregate species we might get more local suggestions on here at least."