Missleading information Q2 Tabular data exploration

Module 1. The Predictive Modeling Pipeline Tabular data exploration
#1

The “(filter and transform)” on the second question was misleading for me.
When I read it, I thought about something like:
df = adult_census[adult_census[“education-num”]>10]

#2

So you mean you did not tick the box for this answer: “a) pandas to manipulate (filter and transform) data”?

Can you explain what you found confusing the “filter and transform” or the “filter” part.

:thinking: designing quizzes is hard … maybe let’s wait if someone else got tripped up by the same answer.

I would propose either:

  1. pandas to manipulate data (staying vague maybe too vague)
  2. pandas to manipulate data, for example apply some transformation or filtering to the data

If you have a better suggestion, please do!

#3

yes, I did not tick the box.

I think you should wait other reviews, it may be a bad understanding on my part.

I think it may be a bad definition on my side:
What I understand by “filter” is “cut some row based on column values”,
and by “transform” is stuff like normalization.

With your first proposal “pandas to manipulate data”, I will have ticked the box.

#4

I agree with @DTJ01.
I did not check the box as I don’t think we filtered nor transformed the data. We just visualized it as a table or histograms with panda, or as plots with seaborn.

#5

Thanks for your input!

Would you have ticked the box with the more vague wording “pandas to manipulate data” ?

#6

Probably yes

#7

I changed the wording to be more vague due to popular demand and “pandas to manipulate data”. The quizz and answer (the answer was a bit weird too, probably it was for an older version of the question) need to be updated in FUN though. The relevant commit for @lfarhi or @MarieCollin is:

#8

It’s fixed in FUN

1 Like
#9