# Data label errors

**URL:** <https://community.drivendata.org/t/data-label-errors/4020>\
**Category:** Hakuna Ma-data\
**Created:** [December 21, 2019, 6:20am UTC](https://community.drivendata.org/t/data-label-errors/4020 "2019-12-21T06:20:50Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![ocords](https://avatars.discourse-cdn.com/v4/letter/o/8c91f0/32.png) [@ocords](https://community.drivendata.org/u/ocords)\
**Post date:** [December 21, 2019, 6:20am UTC](https://community.drivendata.org/t/data-label-errors/4020/1 "2019-12-21T06:20:50Z")

</div>

Hi,

I noticed some errors in train\_labels.csv. There are many instances of gazelles/empty being labeled as lionfemale, for example:

SER\_S1#B07#1#162  
SER\_S1#D03#7#4  
SER\_S1#D04#6#143  
SER\_S1#D05#5#312  
SER\_S1#D06#2#295

For lionfemale, cursory sampling of the labels suggests most of the labels are incorrect.

Did anyone else run into this?

Thanks,  
O&S

---

<div class="post-metadata">

**Author:** ![brismith](https://yyz2.discourse-cdn.com/flex028/user_avatar/community.drivendata.org/brismith/32/884_2.png) [@brismith](https://community.drivendata.org/u/brismith)\
**Post date:** [December 21, 2019, 8:37pm UTC](https://community.drivendata.org/t/data-label-errors/4020/2 "2019-12-21T20:37:35Z")

</div>

SER\_S1#B07#1#162 is images S1\_B07\_R1\_PICT0483, 484 and 485 are a lioness in the dark.  
SER\_S1#D03#7#4 is images S1\_D03\_R7\_PICT0010, S1\_D03\_R7\_PICT0011 and S1\_D03\_R7\_PICT0012 and a lioness walks in from the right.  
SER\_S1#D04#6#143 - tricky there might be something in the middle of the frame…  
SER\_S1#D05#5#312 in isolation for me could be a lion - but a couple of minutes later is certainly a gazelle so probably unlikely.  
SER\_S1#D06#2#295 looks empty.  
I guess the quality of the labelling is part of the challenge.

---

<div class="post-metadata">

**Author:** ![xsainnz](https://avatars.discourse-cdn.com/v4/letter/x/da6949/32.png) [@xsainnz](https://community.drivendata.org/u/xsainnz)\
**Post date:** [December 25, 2019, 9:41am UTC](https://community.drivendata.org/t/data-label-errors/4020/3 "2019-12-25T09:41:43Z")

</div>

Yes, there are many, here are a few for S1 species with \<20 sequences:  
seq\_id category correction  
SER\_S1#P07#1#282 rhinoceros buffalo  
SER\_S1#G02#1#53 genet hare  
SER\_S1#H11#1#2052 genet serval  
SER\_S1#N10#1#40 genet serval  
SER\_S1#H06#3#246 civet jackal  
SER\_S1#D05#5#430 waterbuck unsure (wb have round ears, see SER\_S1#G03#3#136)  
SER\_S1#E09#2#386 waterbuck reedbuck  
SER\_S1#D04#6#201 bushbuck hippo  
SER\_S1#E06#4#4 honeybadger mongoose  
SER\_S1#K07#1#541 honeybadger guineafowl  
SER\_S1#N04#1#1026 honeybadger mongoose  
SER\_S1#N04#1#1096 honeybadger mongoose  
SER\_S1#N04#1#750 honeybadger mongoose  
SER\_S1#Q12#1#155 honeybadger mongoose  
SER\_S1#T13#1#1287 honeybadger mongoose  
SER\_S1#T13#1#769 honeybadger mongoose  
SER\_S1#U13#1#93 honeybadger mongoose  
SER\_S1#I02#1#595 caracal wildcat