Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit 19d6bb0

Browse files
Merge pull request #328 from ahmedfgad/github-actions
PyGAD 3.5.0 Documentation
2 parents 16b5a85 + a2231eb commit 19d6bb0

23 files changed

+12323
-2572
lines changed

‎docs/md/HEADER.md

Lines changed: 164 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,164 @@
1+
[PyGAD](https://github.com/ahmedfgad/GeneticAlgorithmPython) is an open-source Python library for building the genetic algorithm and optimizing machine learning algorithms. It works with [Keras](https://keras.io) and [PyTorch](https://pytorch.org).
2+
3+
> Try the [Optimization Gadget](https://optimgadget.com), a free cloud-based tool powered by PyGAD. It simplifies optimization by reducing or eliminating the need for coding while providing insightful visualizations.
4+
5+
[PyGAD](https://github.com/ahmedfgad/GeneticAlgorithmPython) supports different types of crossover, mutation, and parent selection operators. [PyGAD](https://github.com/ahmedfgad/GeneticAlgorithmPython) allows different types of problems to be optimized using the genetic algorithm by customizing the fitness function. It works with both single-objective and multi-objective optimization problems.
6+
7+
![PYGAD-LOGO](https://user-images.githubusercontent.com/16560492/101267295-c74c0180-375f-11eb-9ad0-f8e37bd796ce.png)
8+
9+
*Logo designed by [Asmaa Kabil](https://www.linkedin.com/in/asmaa-kabil-9901b7b6)*
10+
11+
Besides building the genetic algorithm, it builds and optimizes machine learning algorithms. Currently, [PyGAD](https://pypi.org/project/pygad) supports building and training (using genetic algorithm) artificial neural networks for classification problems.
12+
13+
The library is under active development and more features added regularly. Please contact us if you want a feature to be supported.
14+
15+
# Donation & Support
16+
17+
You can donate to PyGAD via:
18+
19+
- [Credit/Debit Card](https://donate.stripe.com/eVa5kO866elKgM0144): https://donate.stripe.com/eVa5kO866elKgM0144
20+
- [Open Collective](https://opencollective.com/pygad): [opencollective.com/pygad](https://opencollective.com/pygad)
21+
- PayPal: Use either this link: [paypal.me/ahmedfgad](https://paypal.me/ahmedfgad) or the e-mail address ahmed.f.gad@gmail.com
22+
- Interac e-Transfer: Use e-mail address ahmed.f.gad@gmail.com
23+
- Buy a product at [Teespring](https://pygad.creator-spring.com/): [pygad.creator-spring.com](https://pygad.creator-spring.com)
24+
25+
# Installation
26+
27+
To install [PyGAD](https://pypi.org/project/pygad), simply use pip to download and install the library from [PyPI](https://pypi.org/project/pygad) (Python Package Index). The library lives a PyPI at this page https://pypi.org/project/pygad.
28+
29+
Install PyGAD with the following command:
30+
31+
```python
32+
pip3 install pygad
33+
```
34+
35+
# Quick Start
36+
37+
To get started with [PyGAD](https://pypi.org/project/pygad), simply import it.
38+
39+
```python
40+
import pygad
41+
```
42+
43+
Using [PyGAD](https://pypi.org/project/pygad), a wide range of problems can be optimized. A quick and simple problem to be optimized using the [PyGAD](https://pypi.org/project/pygad) is finding the best set of weights that satisfy the following function:
44+
45+
```
46+
y = f(w1:w6) = w1x1 + w2x2 + w3x3 + w4x4 + w5x5 + w6x6
47+
where (x1,x2,x3,x4,x5,x6)=(4,-2,3.5,5,-11,-4.7) and y=44
48+
```
49+
50+
The first step is to prepare the inputs and the outputs of this equation.
51+
52+
```python
53+
function_inputs = [4,-2,3.5,5,-11,-4.7]
54+
desired_output = 44
55+
```
56+
57+
A very important step is to implement the fitness function that will be used for calculating the fitness value for each solution. Here is one.
58+
59+
If the fitness function returns a number, then the problem is single-objective. If a `list`, `tuple`, or `numpy.ndarray` is returned, then it is a multi-objective problem (applicable even if a single element exists).
60+
61+
```python
62+
def fitness_func(ga_instance, solution, solution_idx):
63+
output = numpy.sum(solution*function_inputs)
64+
fitness = 1.0 / numpy.abs(output - desired_output)
65+
return fitness
66+
```
67+
68+
Next is to prepare the parameters of [PyGAD](https://pypi.org/project/pygad). Here is an example for a set of parameters.
69+
70+
```python
71+
fitness_function = fitness_func
72+
73+
num_generations = 50
74+
num_parents_mating = 4
75+
76+
sol_per_pop = 8
77+
num_genes = len(function_inputs)
78+
79+
init_range_low = -2
80+
init_range_high = 5
81+
82+
parent_selection_type = "sss"
83+
keep_parents = 1
84+
85+
crossover_type = "single_point"
86+
87+
mutation_type = "random"
88+
mutation_percent_genes = 10
89+
```
90+
91+
After the parameters are prepared, an instance of the **pygad.GA** class is created.
92+
93+
```python
94+
ga_instance = pygad.GA(num_generations=num_generations,
95+
num_parents_mating=num_parents_mating,
96+
fitness_func=fitness_function,
97+
sol_per_pop=sol_per_pop,
98+
num_genes=num_genes,
99+
init_range_low=init_range_low,
100+
init_range_high=init_range_high,
101+
parent_selection_type=parent_selection_type,
102+
keep_parents=keep_parents,
103+
crossover_type=crossover_type,
104+
mutation_type=mutation_type,
105+
mutation_percent_genes=mutation_percent_genes)
106+
```
107+
108+
After creating the instance, the `run()` method is called to start the optimization.
109+
110+
```python
111+
ga_instance.run()
112+
```
113+
114+
After the `run()` method completes, information about the best solution found by PyGAD can be accessed.
115+
116+
```python
117+
solution, solution_fitness, solution_idx = ga_instance.best_solution()
118+
print("Parameters of the best solution : {solution}".format(solution=solution))
119+
print("Fitness value of the best solution = {solution_fitness}".format(solution_fitness=solution_fitness))
120+
121+
prediction = numpy.sum(numpy.array(function_inputs)*solution)
122+
print("Predicted output based on the best solution : {prediction}".format(prediction=prediction))
123+
```
124+
125+
```
126+
Parameters of the best solution : [3.92692328 -0.11554946 2.39873381 3.29579039 -0.74091476 1.05468517]
127+
Fitness value of the best solution = 157.37320042925006
128+
Predicted output based on the best solution : 44.00635432206546
129+
```
130+
131+
There is more to do using PyGAD. Read its documentation to explore the features of PyGAD.
132+
133+
# PyGAD's Modules
134+
135+
[PyGAD](https://pypi.org/project/pygad) has the following modules:
136+
137+
1. The main module has the same name as the library `pygad` which is the main interface to build the genetic algorithm.
138+
2. The `nn` module builds artificial neural networks.
139+
3. The `gann` module optimizes neural networks (for classification and regression) using the genetic algorithm.
140+
4. The `cnn` module builds convolutional neural networks.
141+
5. The `gacnn` module optimizes convolutional neural networks using the genetic algorithm.
142+
6. The `kerasga` module to train [Keras](https://keras.io) models using the genetic algorithm.
143+
7. The `torchga` module to train [PyTorch](https://pytorch.org) models using the genetic algorithm.
144+
8. The `visualize` module to visualize the results.
145+
9. The `utils` module contains the operators (crossover, mutation, and parent selection) and the NSGA-II code.
146+
10. The `helper` module has some helper functions.
147+
148+
The documentation discusses these modules.
149+
150+
# PyGAD Citation - Bibtex Formatted
151+
152+
If you used PyGAD, please consider citing its paper with the following details:
153+
154+
```
155+
@article{gad2023pygad,
156+
title={Pygad: An intuitive genetic algorithm python library},
157+
author={Gad, Ahmed Fawzy},
158+
journal={Multimedia Tools and Applications},
159+
pages={1--14},
160+
year={2023},
161+
publisher={Springer}
162+
}
163+
```
164+

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /