Library which finds distribution parameters which are best for your data. Several methods (10+) and plenty of distributions (around 40).
Internally fitting is done by calculating (more or less) parameters which are used then as a starting point for optimization.
Based on fitdistrplus
from R world.