Draw a histogram in Wave Analytics

Question

What is the best way to display a distribution of some values like below:

namelookup,connect,appconnect,pretransfer,redirect,starttransfer,total,datetime
3.239255,3.501542,4.051384,4.051477,0.000000,4.339824,4.564202,Tue 4 Apr 2017 01:20:32 UTC
0.229922,0.457110,0.989821,0.989896,0.000000,1.233979,1.683616,Tue 4 Apr 2017 01:21:04 UTC
0.232324,0.456637,0.990802,0.990862,0.000000,1.233964,1.682308,Tue 4 Apr 2017 01:21:36 UTC

I was thinking something like histogram or a scatter plot to show frequency of each. Unfortunately Wave doesn't seem to support time data type so I cannot create grouping by it. Are the some alternatives?

Answer 1

You can use the Date format to load datetime values. While your specific datetime examples aren't supported by the platform, if you can convert them to something like 04/04/2017 01:20:32 AM which is supported.

I don't have any details for your data pipeline, so here's a generic example of datetime pre-processing in Python:

import datetime
oldDate = "Tue 4 Apr 2017 01:20:32 UTC"
datetime.datetime.strptime(oldDate, '%a %d %b %Y %I:%M:%S UTC').strftime('%m/%d/%Y %I:%M:%S %p')
'04/04/2017 01:20:32 AM'

Hope that helps!