search

ossuaire / working-box

:fire: Artificially burn resources... for testing purpose.
GNU General Public License v3.0
0 stars 0 forks source link

Monitored data more meaningful #6

Open Chat-Wane opened 4 years ago

Chat-Wane commented 4 years ago

i.e, for now, each monitored data is stored using a double. It might be a list that allows processing mean, median, standard deviation etc.

Chat-Wane commented 4 years ago

Now, costs are lists [c1, c2, ..., c_k], k being the number of data being monitored before rewriting in the argfilter.

How to exploit these lists in the kernel density estimator that builds interval of costs ?