prediction price A non-numeric value encountered on LeastSquares

Question

I have an error appears when I execute my classe : PHP Warning: A non-numeric value encountered in /php-ai/php-ml/src/Regression/LeastSquares.php

What I tried to do it's to scrap some price for a product and use a regression approach to make a prediction. Currently it's a first approach to test and after I refactor this function

I do not find exactly what's happen with this error.

maybe I make a mistake somewhere

My function

use Phpml\Dataset\ArrayDataset;
use Phpml\Regression\LeastSquares;
use Phpml\CrossValidation\RandomSplit;
use Phpml\Metric\Accuracy;
use Phpml\Math\Average;
use Phpml\Math\StandardDeviation;
use Phpml\NeuralNetwork\MultilayerPerceptron;


function predict_price($data, $data_type) {
    try {
        // Validate input parameters
        if (!is_array($data)) {
            throw new InvalidArgumentException('Data parameter must be an array');
        }

        if (!in_array($data_type, ['csv', 'array', 'database'])) {
            throw new InvalidArgumentException('Invalid data type: ' . $data_type . '. Allowed data types: csv, array, database');
        }

        // create a new linear regression model
        $regression = new LeastSquares();

        // load the dataset
        if ($data_type === 'array') {
            if (!isset($data['samples']) || !is_array($data['samples'])) {
                throw new InvalidArgumentException('Data array parameter must contain a "samples" array');
            }

            if (!isset($data['labels']) || !is_array($data['labels'])) {
                throw new InvalidArgumentException('Data array parameter must contain a "labels" array');
            }

            $samples = array_map(function($sample) {
                if (!is_array($sample) || count($sample) != 2) {
                    throw new InvalidArgumentException('Each sample in the "samples" array must be an array with 2 values');
                }
                $timestamp = strtotime($sample[0]);
                if ($timestamp === false) {
                    throw new InvalidArgumentException('Invalid timestamp: ' . $sample[0] . '. Timestamp should be in the format yyyy-mm-dd');
                }
                return [$timestamp, $sample[1]];
            }, $data['samples']);

            $labels = $data['labels'];
            $dataset = new ArrayDataset($samples, $labels);
        } 

        // split the dataset into training and testing sets
        $randomSplit = new RandomSplit($dataset, 0.3);
        $trainingSamples = $randomSplit->getTrainSamples();
        $trainingLabels = $randomSplit->getTrainLabels();
        $testingSamples = $randomSplit->getTestSamples();
        $testingLabels = $randomSplit->getTestLabels();

        // train the model on the training set
        $regression->train($trainingSamples, $trainingLabels);

        // make predictions on the testing set
        $predictions = $regression->predict($testingSamples);

        // evaluate the performance of the model
        $metric = new Accuracy();
        $accuracy = $metric->score($testingLabels, $predictions);

        // calculate the confidence of the prediction
        $confidence = $regression->getCoefficients()[0];

        // return the accuracy of the model, the prediction for a new data point, and the confidence of the prediction
        $newSample = [date('yyyy-mm-dd'), 1];
        $newPrediction = $regression->predict([$newSample]);

        return array('accuracy' => $accuracy, 'prediction' => $newPrediction[0], 'confidence' => $confidence);
    } catch (InvalidArgumentException $e) {
        // handle input validation errors
        return array('error' => 'Invalid input: ' . $e->getMessage());
    } catch (Exception $e) {
        // handle other exceptions and return an error message
        return array('error' => 'Unexpected error: ' . $e->getMessage());
    }
}

and below my example to execute :

$historical_prices = array(
    array('2022-01-01', 10),
    array('2022-01-02', 11),
    array('2022-01-03', 12),
    array('2022-01-04', 13),
    array('2022-01-05', 14),
    array('2022-01-06', 15),
    array('2022-01-07', 16),
    array('2022-01-08', 17),
    array('2022-01-09', 18),
    array('2022-01-10', 19),
);

$data = array(
    'samples' => $historical_prices,
    'labels' => array(11, 12, 13, 14, 15, 16, 17, 18, 19, 20),
);

$result = predict_price($data, 'array');
echo 'Accuracy: ' . $result['accuracy'] .  "<br>";
echo 'Predicted price: ' . $result['prediction']  . "<br>";
echo 'Confidence: ' . $result['confidence']  . "<br>";
echo '-------<br>';

result :

Accuracy: 0 Predicted price: 2.9258048044576 Confidence: 1.1612428352237E-8