Algorithm¶

After having defined the problem, a suitable algorithm for optimizing it has to be found. This can be challenging and might require some literature research. pymoo offers quite a few standard implementations of well-known algorithms that can be quite useful in obtaining quick results or prototyping.

Each algorithm has different parameters to be initialized. For the functional interface, the algorithm object needs to be passed to the minimize method, starting the optimization run. For instance, for NSGA2 the object can be initialized as follows:

[1]:

from pymoo.algorithms.moo.nsga2 import NSGA2

algorithm = NSGA2()

For more details about algorithms, please have a look at this tutorial.