Skip to main content

In some cases I like to create cumulative statistics using the StatisticsCalculator. I'm glad it's kept as a Advanced option in the StatisticsCalculator.

 

However, it feels a bit odd that the StatisticsCalculator seems to require/expect attributes for 'Summary statistics'.

I.e. if I only want to create cumulative statistics, the StatisticsCalculator shows an error (Red Cogwheel) indicating; 'Parameter 'Statistics to Calculate: ' must have a value';

image 

Although it's possible to ignore the error and run the translation, or provide a dummy attribute to calculate regular 'summary statistics', it doesn't feel fully correct.

It would be nice to also have the option to only create Cumulative Statistics. I guess that may be possible by (also) embedding the regular 'summary statistics' in a Checked ParameterGroup (and be default checked).

well it does seem a bit weird. I guess they could add a little checkbox to disable the main UI. But I guess that might add it's own headaches.

 

I see also you get a warning if there's nothing in there. You could create an idea to have the transformer improved. But in the end there is no real change to the outcome here.


well it does seem a bit weird. I guess they could add a little checkbox to disable the main UI. But I guess that might add it's own headaches.

 

I see also you get a warning if there's nothing in there. You could create an idea to have the transformer improved. But in the end there is no real change to the outcome here.

Check, thanks for the feedback. I created the idea; https://community.safe.com/s/bridea/a0rDm000000UljcIAC

 

I don't really know what you mean with 'there is no real change to the outcome here'. Do you relate to the current situation?

 

Also, for people that also came across this topic. There is an alternative to calculating Cumulative Statistics; by using an AttributeCreator or AttributeManager with the 'Adjacent Feature' option enabled (with 1 prior feature). But that requires a bit more manual fiddling. Personally I prefer the option to have a transformer that does this for me.


Check, thanks for the feedback. I created the idea; https://community.safe.com/s/bridea/a0rDm000000UljcIAC

 

I don't really know what you mean with 'there is no real change to the outcome here'. Do you relate to the current situation?

 

Also, for people that also came across this topic. There is an alternative to calculating Cumulative Statistics; by using an AttributeCreator or AttributeManager with the 'Adjacent Feature' option enabled (with 1 prior feature). But that requires a bit more manual fiddling. Personally I prefer the option to have a transformer that does this for me.

Nice, Yeah I just mean if you put in the dummy attribute in there to remove the red gear but don't pick any thing for it to calculate the the output in the cumulative port if still what you're expecting.

This change isn't really effecting your ability to do what you need to do more like it just doesn't feel quite right.

Also the adjacent feature handling I thiiiink is probably slower in the AttributeCreator but not sure on that.


Check, thanks for the feedback. I created the idea; https://community.safe.com/s/bridea/a0rDm000000UljcIAC

 

I don't really know what you mean with 'there is no real change to the outcome here'. Do you relate to the current situation?

 

Also, for people that also came across this topic. There is an alternative to calculating Cumulative Statistics; by using an AttributeCreator or AttributeManager with the 'Adjacent Feature' option enabled (with 1 prior feature). But that requires a bit more manual fiddling. Personally I prefer the option to have a transformer that does this for me.

Ah, I did not yet consider just specifying a dummy attribute for Summary Statistics, but not specifying any statistic measure to calculate for that dummy attribute...

 

Nevertheless, I think we're on the same page. Not really a big issue, just doesn't feel quite right.


Reply