Ability to import frequence in CSV file

gkunter / parsets

Automatically exported from code.google.com/p/parsets

Other

0 stars 0 forks source link

Ability to import frequence in CSV file #20

Open GoogleCodeExporter opened 8 years ago

GoogleCodeExporter commented 8 years ago

Love the application, but could not import frequency of records in csv
file, leading to a very long CSV file.
If I have missed how to do this (numerical record?) can more help be
provided on CSV import?

Original issue reported on code.google.com by robertmu...@gmail.com on 4 Aug 2009 at 10:51

GoogleCodeExporter commented 8 years ago

That's not currently possible. It's certainly possible to add that, though we'd 
need to think of a good way of 
incorporating that into the GUI. Do you have one numerical dimension (the 
count), or could there be more?

Original comment by rkosara on 4 Aug 2009 at 11:52

Changed state: Accepted

GoogleCodeExporter commented 8 years ago

I could think of an example where I might one to fit more than onenumerical 
dimesion.

Suppose that the initial analysis was on just the categorical dimensions but I 
had 
some continuous explanatory variables.  I could then uthe display to show how 
the 
predicted probabilities varied for different models.

In the simplest case, suppose I had a test taken by male and females with a 
pass 
fail grade.  This could generate a parallel sets plot.  However, if I also had 
measure of prior attainment then I could generate predicted probabilities at a 
given 
level of attainment and produce an equivalent parallel sets plots were the 
widths 
would be the probabilities of moving down the plot.

Original comment by bell...@gmail.com on 1 Apr 2010 at 1:11

GoogleCodeExporter commented 8 years ago

This is very difficult (there are no standard ways to specify rolled-up data) 
and seems to be of very limited use. It's also an enhancement, not a defect 
(fixing the classification).

Original comment by rkosara on 27 Dec 2010 at 3:30

Added labels: Priority-Low, Type-Enhancement
Removed labels: Priority-Medium, Type-Defect

GoogleCodeExporter commented 8 years ago

Really cool-application!

I would also like to be able to include a column that represents the frequency 
of that row, avoiding to repeat it many time. So a very simple way. 

An (fake) example is of trade flows among countries could be:
US EU 20
EU US 15
US CN 10
CN US 50
EU CN 20
CN EU 40

This feature would allow to represent not only survey-like data but also 
aggregated data (most commonly available).

Thanks before-hand!

Original comment by afamo...@gmail.com on 13 Oct 2014 at 7:08