fann_create_train_from_callback

(PECL fann >= 1.0.0)

fann_create_train_from_callback — 从用户提供的函数创建训练数据结构。

说明

fann_create_train_from_callback ( int $num_data , int $num_input , int $num_output , callable $user_function ) : resource

从用户提供的函数创建训练数据结构。当训练数据可数，用户需要构造一个用来接收训练数据集(包括输入，输出)数量和返回值为集合的函数。

参数

num_data

训练数据的数量。

num_input

每个训练数据的输入数。

num_output

每个训练数据的输出数。

user_function

用户提供的函数包含以下参数：

num - 训练数据集的数量。
num_input - 数量数据的输入数。
num_output - 数量数据的输出数。

函数应该返回一个包含input 和 output 键的数组，并且这两个键的值分别表示输入输出的值(皆为数组)。

返回值

成功时返回训练数据资源，发生错误返回 false。

范例

Example #1 fann_create_train_from_callback() example


<?php
function create_train_callback($num_data, $num_input, $num_output) {
    return array(
        "input" => array_fill(0, $num_input, 1),
        "output" => array_fill(0, $num_output, 1),
    );
}

$num_data = 3;
$num_input = 2;
$num_output = 1;
$train_data = fann_create_train_from_callback($num_data, $num_input, $num_output, "create_train_callback");
if ($train_data) {
    // Do something with $train_data
}
?>

参见

fann_read_train_from_file() - 读取存储训练数据的文件。
fann_train_on_data() - 在整个数据集上训练一段时间。
fann_destroy_train() - 销毁训练数据。
fann_save_train() - 将训练结构体保存至文件。

User Contributed Notes

geekgirljoy at gmail dot com 21-Oct-2016 05:34


This code can be used to read training data from MySQL rather than a text file.



<?php



// MySQL for This Example:

/*

CREATE TABLE `TrainingSets` (

  `ID` int(11) NOT NULL,

  `Name` varchar(150) COLLATE utf8mb4_unicode_ci NOT NULL,

  `TrainingData` text COLLATE utf8mb4_unicode_ci NOT NULL

) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci;



ALTER TABLE `TrainingSets` ADD PRIMARY KEY (`ID`);

  

INSERT INTO `TrainingSets` (`ID`, `Name`, `TrainingData`) VALUES(1, 'XOR', '-1 -1\n-1\n-1 1\n1\n1 -1\n1\n1 1\n-1');



ALTER TABLE `TrainingSets` MODIFY `ID` int(11) NOT NULL AUTO_INCREMENT, AUTO_INCREMENT=2;

*/



// This function calls pulls the TrainingData from MySQL

function get_training_data_from_db($id) {

    $table_name = "TrainingSets";

    $field = "TrainingData";

    $connection=mysqli_connect("host","username","password","database"); // change to your DB credentials

    $result=mysqli_query($connection,"SELECT $field FROM $table_name");

    $data=mysqli_fetch_assoc($result);

    mysqli_close($connection);



    return $data[$field];

}



// This function prepares the newline delimited data to be handed off to FANN

/*

Example of "newline delimited data" (like XOR in a Plain Text File) stored in MySQL:

-1 -1

-1

-1 1

1

1 1

-1

1 -1

1

*/

function prepare_data_from_db($training_data) {

    $training_data = explode( "\n", $training_data ); // convert training data rows to array

    $num_data = count($training_data);

    

    // Sift the data and split inputs and outputs 

    for($i=0;$i<$num_data;$i++) {

      if($i % 2) { // $training_data[$i] is Output

       $training_data['outputs'][] = explode( " ", $training_data[$i]);

      }else{ // $training_data[$i] is Input

       $training_data['inputs'][] = explode( " ", $training_data[$i]);

      }

    }

    // remove the unsifted data

    foreach ($training_data as $key => $value) {

        if (is_numeric($key)) {

            unset($training_data[$key]);

        }

    }

    return $training_data; // returned the prepaired associative array

}



// This function hands the prepared data over to FANN

function create_train_callback($num_data, $num_input, $num_output) {

    global $training_data;

    global $current_dataset;

   

    $dataset = array("input" => $training_data['inputs'][$current_dataset],

                    "output" => $training_data['outputs'][$current_dataset]);

    $current_dataset++;



    return $dataset;

}



// Initialize the program variables

$record_id = 1; // the 'ID' for the training data in MySQL

$current_dataset = 0;

$num_input = 2;

$num_output = 1;

$num_layers = 3;

$num_neurons = 3;

$desired_error = 0.001;

$max_epochs = 500000;

$epochs_between_reports = 1000;



$training_data = get_training_data_from_db($record_id); // Get the Training Data from MySQL

$training_data = prepare_data_from_db($training_data); // Prepare the data

$num_data = count($training_data["input"]); // How many sets are there?



// Hand the data over to FANN

$train_data = fann_create_train_from_callback($num_data, $num_input, $num_output, "create_train_callback");



// Test for $train_data 

if ($train_data) {

    

    // Create $ann

    $ann = fann_create_standard($num_layers, $num_input, $num_neurons, $num_output);



    // Test for $ann

    if ($ann) {

        fann_set_activation_function_hidden($ann, FANN_SIGMOID_SYMMETRIC);

        fann_set_activation_function_output($ann, FANN_SIGMOID_SYMMETRIC);



        // Train XOR ANN with training data obtainied from MySQL

        if (fann_train_on_data($ann, $train_data, $max_epochs, $epochs_between_reports, $desired_error)){

           print('XOR trained.<br>' . PHP_EOL);



           // Test $ann

           $input = array(-1, 1);

           $calc_out = fann_run($ann, $input);

           printf("xor test (%f,%f) -> %f\n", $input[0], $input[1], $calc_out[0]);

           

           // destore $ann 

           fann_destroy($ann);

        }

    }

}

?>