# Doubts regarding test data set

**URL:** <https://community.drivendata.org/t/doubts-regarding-test-data-set/2782>\
**Category:** Box-Plots for Education\
**Created:** [October 9, 2018, 6:03pm UTC](https://community.drivendata.org/t/doubts-regarding-test-data-set/2782 "2018-10-09T18:03:15Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![AmoghaShettar](https://avatars.discourse-cdn.com/v4/letter/a/91b2a8/32.png) [@AmoghaShettar](https://community.drivendata.org/u/AmoghaShettar)\
**Post date:** [October 9, 2018, 6:03pm UTC](https://community.drivendata.org/t/doubts-regarding-test-data-set/2782/1 "2018-10-09T18:03:15Z")

</div>

1.Why text1,2,3,4 attributes don’t have any labels in test dataset?

2.Does it mean that we should consider them as NOT so important attributes?

3.Is it okay if we discard those attributes during pre-processing step?