Discover the essential Python libraries for modern data science in 2026—from NumPy and pandas to AutoML systems like mljar-supervised and AI-native environments like MLJAR Studio.
Explore the Navy SEALs approach to performance vs trust and discover how it relates to AI in software development. Learn why trust, not just performance, is crucial for building reliable AI systems and high-performing teams.
Explore practical AutoML projects using 10 popular datasets for binary classification, multiclass classification, and regression tasks. Learn how to effectively use MLJAR AutoML in Python notebook.
Let me introduce you an open-source Automated Machine Learning framework called mljar-supervised. It is available as Python package. It can be used with Python API or with graphical interface in the MLJAR Studio - our distribution of JupyterLab with user-friendly extensions.
When using machine learning models like LightGBM, it is important to keep the order of columns consistent in your data, even when using Pandas DataFrames. I was surprised to find that changing the column order after training can affect prediction results, a detail even ChatGPT missed.
Explore 8 open-source AutoML frameworks that automate machine learning tasks, from algorithm selection to hyperparameter tuning. Find the perfect tool for your next project with expert insights.
Explore how different machine learning algorithms stack up against each other. This article compares key methods, highlights their strengths and weaknesses, and helps you choose the best one for your needs.
The mljar-supervised Python package (version 1.0.0) now supports fairness-aware training, enabling the measurement and mitigation of bias in sensitive features for binary classification, multiclass classification, and regression.
Discover the evolution of Automated Machine Learning (AutoML) frameworks, exploring open-source solutions like AutoGluon, H2O, and MLJAR AutoML, and reflecting on the broader goals beyond performance optimization.
MLJAR now features an integration with Optuna for hyperparameter tuning. Optuna, a hyperparameter optimization framework, allows users to focus on maximizing model performance without strict time constraints.
Python Notebooks offer an interactive computing environment, widely used by Data Scientists for data analysis. MLJAR AutoML supports displaying interactive reports directly in Python Notebooks, enhancing transparency in machine learning pipelines.
Explore the inner workings of AutoML with a focus on the open-source MLJAR AutoML. Learn how it builds a machine learning pipeline automatically, gaining insights into its processes.
Automated Machine Learning (AutoML) effortlessly applies ML algorithms to PostgreSQL data for insights and predictions. Using mljar-supervised, an open-source Python package, this guide simplifies the process with code snippets.
MLJAR, an open-source AutoML library, transformed my machine learning journey. Its user-friendly interface automates tasks, making it a top performer in Hackathons. Experience efficiency without extensive coding.
Google has introduced AutoML Tables, a structured dataset AutoML service in Beta. This article compares it with an open-source solution, providing insights into their features and performance.
Automated Machine Learning (AutoML) streamlines the end-to-end machine learning process, encompassing data preprocessing, feature engineering, algorithm selection, hyperparameter optimization, and more. This image showcases various open-source AutoML tools.
In the battle of Random Forest versus AutoML, AutoML superior performance is clear. However, I will highlight other advantages, demonstrating how AutoML excels in handling real-world, messy data.
Automated Machine Learning (AutoML) involves building models without human intervention. Various AutoML packages include auto-sklearn, h2o autoML, TPOT, auto_ml (unmaintained), and mljar. Explore a comprehensive list of AutoML software.
See how machine learning, once limited to big companies, is now accessible for small start-ups. Explore a demo using MLJAR R API to predict customer churn for a telecom company.
Understanding hyper-parameters is vital in machine learning. Unlike parameters derived from data, hyper-parameters impact algorithm performance. Tuning them intelligently, using methods like grid search, is crucial for optimal models.
This post evaluates MLJAR performance on the Kaggle Give me some credit dataset, comparing results with predictive APIs from Amazon, Google, PredicSis, and BigML. Inspired by Louis Dorard article.
