feat: add genetic algorithm library

[Cyntwikip]

Description

My implementation of genetic algorithm. This is a search heuristic for identifying the parameters in a system that will yield good results through the process of natural selection. This file is a modification of the library made by me as well. Here is the link of the original implementation: https://github.com/Cyntwikip/libraries/blob/master/genetic_algo/ga.py

Sample Usage:

from ga import GeneticAlgo as ga
import numpy as np

# Pool of genes
pool_scaler = ["MinMax (-1,1)", "MinMax (0,1)", "Std"]
pool_batch_size = [64, 72, 128, 256]
pool_lstm_layers = [[128, 128, 128],
                    [64, 64, 64],
                    [32, 32, 32],
                    [128, 64, 64],
                    [64, 32, 32],
                    [128, 64, 32]]
pool_dropout = [0., 0.1, 0.2]
pool_activation = ['tanh', 'sigmoid']
pool_loss = ['mean_squared_error', 'mae']
pool_epochs = [5]

genes_dict = dict( scaler = pool_scaler,
                batch_size = pool_batch_size,
                lstm_layers = pool_lstm_layers,
                dropout = pool_dropout,
                activation = pool_activation,
                loss = pool_loss,
                epochs = pool_epochs
            )

def survival(n):
    codes = ga.get_initial_population(n)
    np.random.seed(42)
    scores = np.random.randint(0,40,n)
    fitness = dict(zip(codes, scores))
    print(fitness)
    best_parents = ga.get_best_parents(fitness)
    print(best_parents)

ga = ga(genes_dict, debug=True)
survival(10)

More sample test cases will pushed in a follow-up commit.

Checklist:

• Code contain Problem statement, examples test cases.
• My code follows the style guidelines(Clean Code) of this project
• I have performed a self-review of my own code
• Explained Detailed approach of Solution
• I have commented on my code, particularly in hard-to-understand areas
• Performed Time complexity analysis of the algorithm used in code