0

I would like to calculate in Python 2.7 the mean in specific rows (see also Question in R) and have a list at the end. Here in my case I want to take the mean duration for each event. So something like this (preferably in one line and computational fast):

    mean_duration = [np.mean(r.duration, r[-1].duration) for r in catalog 
                     if r.eventid == r[-1].eventid]

Unfortunately, I cannot do it like this because my class (and therefor my catalog) said that it cannot use getitem.

Here is the part of my data set:

Eventid           Station      Duration
20120525_0010H    stat1        2.40   
20120525_0010H    stat2        2.80   
20120525_0010H    stat3        2.40    
20120525_0223E    stat1        2.80    
20120525_0223E    stat3        2.00     
20120525_0225L    stat2        3.20     
20120525_0230B    stat1        3.20     
20120526_0019C    stat3        3.20
Community
  • 1
  • 1
Seb
  • 1
  • 2

1 Answers1

0
data = {}
for r in catalog:
  data[r.eventid] = data.get(r.eventid, {})
  data[r.eventid][r.station] = data.get(r.station, {})
  data[r.eventid][r.station] = r.duration
mdur = [np.mean(data[eventid].values()) for eventid in eventids]
Seb
  • 1
  • 2