Machine learning with Laravel and Rubix

Welcome to our ultimate guide for machine learning tutorials for beginners. In this article, we’ll provide you with step-by-step instructions on how to get started with machine learning using Laravel, a popular PHP framework, and Rubix ML, a powerful machine learning library. By the end of this tutorial, you’ll have a solid understanding of the fundamentals of machine learning and the confidence to start building your projects.

What is Machine Learning?

Before we dive into the tutorials, let’s first define machine learning. Machine learning is a subset of artificial intelligence that focuses on creating algorithms to learn from and make predictions based on data. This is achieved by training a model using historical data, which then allows the model to make predictions or decisions without being explicitly programmed to perform the task.

Why Use Laravel and Rubix ML for Machine Learning?

Laravel is a widely used PHP framework known for its elegant syntax, rapid development cycle, and powerful features. It’s perfect for building web applications and APIs, making it a great choice for integrating machine learning into your projects.

On the other hand, Rubix ML is a high-level machine learning library designed for PHP. It has many machine-learning algorithms and tools for data preprocessing and model evaluation. Laravel and Rubix ML provide a robust and efficient solution for developing machine learning applications.

Now that you have a basic understanding of machine learning and the tools we’ll use, let’s jump into our step-by-step tutorials.

Tutorial 1: Setting Up Your Environment

Step 1: Install Laravel

First, you need to install Laravel. You can do this by following the official Laravel installation guide found here.

Step 2: Configure Your Environment

Now that you have both Laravel and Rubix ML installed, it’s time to configure your environment. Create a new Laravel project using the command:

laravel new machine_learning_tutorials
# Note you could also use the composer commmand to create a project.
composer create-project laravel/laravel machine_learning_tutorials

This will create a new Laravel project in a directory named machine_learning_tutorials. Navigate to this directory and open the .env file. Ensure that your database credentials are correctly set up.

Step 3: Install Rubix ML

Next, install Rubix ML in your Laravel project using the following command:

composer require rubix/ml

This command will add Rubix ML as a dependency in your Laravel project, allowing you to use its features easily.

Tutorial 2: Building a Simple Classifier

In this tutorial, we’ll create a simple classifier that can predict if a person is diabetic based on their age, sex, BMI, and blood pressure.

Step 1: Create a Database Table and Model

First, create a new database migration using the command:

php artisan make:migration create_diabetes_data_table

Edit the newly created migration file to include the following columns: age, sex, bmi, blood_pressure, and diabetic. Run the migration with:

php artisan migrate

Next, create a new Eloquent model named DiabetesData Using the command:

php artisan make:model DiabetesData

Step 2: Import Training Data

For this tutorial, we’ll use the Pima Indians Diabetes dataset, which you can download here. Import the dataset into the diabetes_data the table you created earlier.

Step 3: Train the Classifier

Now that our dataset is imported, it’s time to train our classifier. In this tutorial, we’ll use the K-Nearest Neighbors (KNN) algorithm from Rubix ML. First, create a new controller using the following command:

php artisan make:controller DiabetesClassifierController

Open the DiabetesClassifierController.php File and add the following code:

phpCopy codeuse Rubix\ML\Datasets\Labeled;
use Rubix\ML\Kernels\Distance\Manhattan;
use Rubix\ML\Classifiers\KNearestNeighbors;
use App\Models\DiabetesData;

public function trainClassifier()
{
    # Caution: For simplicity of the tutorial and because we know the amount of data we imported, we are using all(); But in real world scenarios YOU NEVER WANT to do so. 
    $data = DiabetesData::all();
    $samples = [];
    $labels = [];

    foreach ($data as $row) {
        $samples[] = [
            $row->age,
            $row->sex,
            $row->bmi,
            $row->blood_pressure,
        ];

        $labels[] = $row->diabetic;
    }

    $dataset = new Labeled($samples, $labels);
    $classifier = new KNearestNeighbors(3, new Manhattan());

    $classifier->train($dataset);

    // Save the trained classifier to a file
    $classifier->save('knn_classifier.rbx');

    return response()->json(['message' => 'Classifier trained successfully']);
}

This code trains the KNN classifier using the imported dataset and saves the trained classifier to a file for later use.

Step 4: Create an API Endpoint for Predictions

Now that we have a trained classifier, let’s create an API endpoint to make predictions. Add the following code to your DiabetesClassifierController:

use Rubix\ML\Persisters\Filesystem;

public function predictDiabetes(Request $request)
{
    $request->validate([
        'age' => 'required|numeric',
        'sex' => 'required|numeric',
        'bmi' => 'required|numeric',
        'blood_pressure' => 'required|numeric',
    ]);

    $persister = new Filesystem('knn_classifier.rbx');
    $classifier = $persister->load();

    $prediction = $classifier->predict([
        $request->age,
        $request->sex,
        $request->bmi,
        $request->blood_pressure,
    ]);

    return response()->json(['diabetic' => $prediction]);
}

This code loads the trained classifier from the file and uses it to make a prediction based on the user’s input.

Finally, add the following routes to your routes/api.php file:

use App\Http\Controllers\DiabetesClassifierController;

Route::post('/train', [DiabetesClassifierController::class, 'trainClassifier']);
Route::post('/predict', [DiabetesClassifierController::class, 'predictDiabetes']);

Now you can train your classifier by making a POST request to /train and make predictions by sending a POST request to /predict with the required parameters.

Wrapping Up

Congratulations! You’ve successfully built a simple machine learning classifier using Laravel and Rubix ML. While this tutorial focused on a specific use case, you can quickly adapt these concepts to other machine-learning problems.

Remember, practice is the key to mastering machine learning. With this guide and your newfound knowledge, you’re on your way to becoming a machine learning expert. Keep experimenting with different algorithms and datasets to unlock your full potential in machine learning.

By Louis M.

About the author – My LinkedIn profile

Related Links:

• How to Evaluate, Validate, and Optimize Your Machine Learning Models
• ChatGPT: How to Use AI to Make Your Customers Happier
• ChatGPT: The Future of AI-Powered Customer Support
• Top 10 Best Machine Learning Algorithms You Should Know
• ChatGPT: How to Use AI to Save Time and Money