python-scipyHow do I use Python Scipy's Differential Evolution to optimize a function?

Differential Evolution (DE) is a powerful evolutionary algorithm used to optimize a function. It is included in the SciPy library and can be used as follows:

from scipy.optimize import differential_evolution

def func(x):
    return x[0]**2 + x[1]**2

bounds = [(0,2), (0,2)]
result = differential_evolution(func, bounds)

print(result.x)
# Output: array([1.99999998, 1.99999998])

The differential_evolution function takes two arguments: the function to be optimized, and a tuple of bounds for each parameter. The function should return a single value. In this example, we are optimizing a simple function of two variables. The output of the function is an OptimizeResult object, which contains the optimal parameters, and other useful information.

Parts of the code:

• from scipy.optimize import differential_evolution: imports the Differential Evolution function from the SciPy library.
• def func(x):: defines the function to be optimized.
• bounds = [(0,2), (0,2)]: defines the bounds for each parameter of the function.
• result = differential_evolution(func, bounds): runs the Differential Evolution algorithm, passing the function and bounds as arguments.
• print(result.x): prints the optimal parameters.

Helpful links

• SciPy Differential Evolution Documentation
• Differential Evolution Algorithm

Edit this code on GitHub