Vector dimension

Vector dimension validation (predict)

Below is a minimal example: the predict function expects an array of 10 numbers and throws an exception if the dimension does not match.

Example of use


                
<?php

function vectorPredict(array $features): float {
    if (count($features) !== 10) {
        throw new InvalidArgumentException('Expecting a vector of dimension 10');
    }

    // Further calculations
    return 0.75;
}

try {
    $features = [0.12, 0.85, 0.33, 0.67, 0.91, 0.44, 0.58, 0.76, 0.29, 0.50];

    $result = vectorPredict($features);
    echo 'Model score: ' . round($result, 3) . PHP_EOL;

    // Result interpretation
    if ($result > 0.7) {
        echo 'High probability of a positive outcome';
    } elseif ($result > 0.4) {
        echo 'Medium probability';
    } else {
        echo 'Low probability';
    }
} catch (Exception $e) {
    echo 'Error: ' . $e->getMessage();
}

Result: Memory: 0.001 Mb Time running: < 0.001 sec.


                Model score: 0.75
High probability of a positive outcome

The idea is straightforward: dimension is a contract between the data pipeline and the model. In production, such checks help you detect feature-generation bugs early.