check spp.key about trawldata HOT 4 OPEN

@JWMorley @bselden I made a video to show how I was making some of the corrections --- they still aren't perfect by any means. But I do show how I filter down to some of the entries that might merit a second glance.

https://www.youtube.com/watch?v=RZlUds2Ph_0

Feel free to go about this however you want, if you can find time. Any help is much appreciated.

from trawldata.

@JWMorley @bselden Note that I edited the link to the video; so in your email you will still only see the old link. The new link is here: https://www.youtube.com/watch?v=RZlUds2Ph_0

And that is the same link that will appear on the GitHub issues site.

from trawldata.

@JWMorley @bselden @mpinsky

So, I'm updating the data sets (the US ones for now), and I found 1131 new raw taxonomic ID's that aren't in spp.key already ... whoa. I'll put my auto-match code to work, but everything added in this way will be given the flag of "added_automatically".

Working on properly adding these to the spp.key. It's basically done, just need to integrate it well with make() and add checks.

from trawldata.

I've recently gone through most lines of the spp.key manually.

I've manually checked 2654 rows in the recent effort; another 548 are "ok", 53 "manual", 577 "fine", 586 "bad", 316 "becca_batch2", and a lot of other random flags that indicate it's been checked in some way. In theory, the "bad" rows might need to be fixed, but they generally aren't ID'd to species, and are tossed out in the trim row due to that flag; so they aren't a big worry.

There are 1009 rows that were "added_automatically", and 349 have an NA flag. None of these rows pertain to species that are in the current trawlDiversity analysis (due to subsampling years, day of year, and strata).

So this is very near completion, and is much less of a worry for my current analysis, but could still use some work. I also wouldn't be surprised is some of my "check" rows had errors/ typos (I found 1 or 2 already). So it ain't perfect.

from trawldata.