You can use the initial_population parameter to set your custom initial population.

ga_instance = pygad.GA(...,
 initial_population=
 ...)

Can you further clarify the other question. Do you mean 70 generations or solutions?

@ahmedfgad
I use this genetic algorithm code. And the "initial_population" parameter is not defined in this code :

import numpy as np
from geneticalgorithm import geneticalgorithm as ga

def f(X):
return np.sum(X)

varbound=np.array([[0,10]]*3)

algorithm_param = {'max_num_iteration': 3000,
'population_size':100,
'mutation_probability':0.1,
'elit_ratio': 0.01,
'crossover_probability': 0.5,
'parents_portion': 0.3,
'crossover_type':'uniform',
'max_iteration_without_improv':None}

model=ga(function=f,
dimension=3,
variable_type='real',
variable_boundaries=varbound,
algorithm_parameters=algorithm_param)

model.run()

I mean 70 solutions.

I am going to minimize a function. Its sample code is as follows; I don't know how to run this code using PyGAD

def LEVYFUNCTION(x):
term1 = (math.sin(3math.pix[0]))**2;
term2 = (x[0]-1)**2 * (1+(math.sin(3math.pix[1]))**2);
term3 = (x[1]-1)**2 * (1+(math.sin(2math.pix[1]))**2);
return (term1 + term2 + term3)

varbound=np.array([[-10,10]]*2)
algorithm_param = {'max_num_iteration': 2000,
'population_size':50,
'mutation_probability':0.1,
'elit_ratio': 0.01,
'crossover_probability': 0.5,
'parents_portion': 0.3,
'crossover_type':'uniform',
'max_iteration_without_improv':None}
model=ga(function=LEVYFUNCTION,dimension=2,variable_type='real',variable_boundaries=varbound,algorithm_parameters=algorithm_param)
model.run()

y = (sin(3pix1))^2 + (x1-1)^2 * (1+(sin(3pix2))^2) + (x2-1)^2 * (1+(sin(2pix2))^2)
x= [-10, 10]

Global Minimum:
f(x)=0, x=(1,1)

If possible please guide how can I run this code using PyGAD?

And if the minimum and maximum range of x is not the same, how should it be defined?