jtet/感知器

感知器的简单PHP实现。

维护者

详细信息

github.com/jtet/Perceptron

主页

源代码

问题

安装: 95

依赖: 0

建议者: 0

安全: 0

星星: 14

关注者: 2

分支: 7

公开问题: 0

dev-master 2021-02-04 14:00 UTC

Requires

  • php: >=5.3.0

OSL-3.0 76ea8d355a8424a4af6ad0c339f401318fcbdc36

  • jtet <jtet.github.woop@gmail.com>

perceptronartificial intelligence

This package is not auto-updated.

Last update: 2024-09-28 14:44:08 UTC

README

Build Status

什么是感知器？

"the perceptron is an algorithm for supervised classification of an input into one of two possible outputs.
It is a type of linear classifier, i.e. a classification algorithm that makes its predictions based on a
linear predictor function combining a set of weights with the feature vector describing a given input."

更多信息请访问 http://en.wikipedia.org/wiki/Perceptron

训练

while($p->getIterationError() > $x)
{
    for ($i = 0; $i < count($inputVectors); $i++){
        $p->train($inputVectors[$i], $outcomes[$i);
    }
}

测试输入

echo $p->test($inputVector)? "True": "False";

示例

以下代码将感知器训练为NAND门

$p = new \JTet\Perceptron\Perceptron(2);

$i = 0;
while($i < 100000)
{
    $input = array(0, 0);
    $output = 1;
    $p->train($input, $output);

    $input = array(0, 1);
    $output = 1;
    $p->train($input, $output);

    $input = array(1,0);
    $output = 1;
    $p->train($input, $output);

    $input = array(1,1);
    $output = 0;
    $p->train($input, $output);

    $i++;
}

echo $p->test(array(1,1))? "Incorrect\n": "Correct\n";
echo $p->test(array(0,1))? "Correct\n": "Incorrect\n";
echo $p->test(array(0,0))? "Correct\n": "Incorrect\n";
echo $p->test(array(1,0))? "Correct\n": "Incorrect\n";

获取感知器

将以下内容添加到您的composer.json文件中，并运行composer update。

"require": {
        "jtet/perceptron": "dev-master"
    }

许可证

感知器在OSL-3.0许可证下可供您使用。