This article provides an introduction to Machine Learning (ML) for physicists, aiming to bridge the gap between physics and ML by leveraging the knowledge of statistical physics. The review covers fundamental concepts such as the bias-variance tradeoff, overfitting, regularization, generalization, and gradient descent, and progresses to more advanced topics like supervised and unsupervised learning, ensemble models, deep learning, clustering, energy-based models, and variational methods. The authors emphasize the use of physics-inspired datasets, such as the Ising model and Monte-Carlo simulations, to introduce ML concepts and tools. The review also discusses the challenges and overlaps between ML and statistical physics, and highlights the potential for physicists to contribute to ML research. The article concludes with an outlook on the future applications of ML in physics and open problems in ML where physicists can make significant contributions. The review is structured to provide both theoretical foundations and practical insights, supported by Jupyter notebooks that demonstrate the implementation of ML techniques using Python. The authors aim to equip physicists with the necessary skills to apply ML in their research and to understand the theoretical underpinnings of ML algorithms. The article also emphasizes the importance of understanding the difference between fitting and predicting, and the role of model complexity in achieving good generalization. Overall, the review serves as a comprehensive resource for physicists interested in learning and applying ML techniques in their work.This article provides an introduction to Machine Learning (ML) for physicists, aiming to bridge the gap between physics and ML by leveraging the knowledge of statistical physics. The review covers fundamental concepts such as the bias-variance tradeoff, overfitting, regularization, generalization, and gradient descent, and progresses to more advanced topics like supervised and unsupervised learning, ensemble models, deep learning, clustering, energy-based models, and variational methods. The authors emphasize the use of physics-inspired datasets, such as the Ising model and Monte-Carlo simulations, to introduce ML concepts and tools. The review also discusses the challenges and overlaps between ML and statistical physics, and highlights the potential for physicists to contribute to ML research. The article concludes with an outlook on the future applications of ML in physics and open problems in ML where physicists can make significant contributions. The review is structured to provide both theoretical foundations and practical insights, supported by Jupyter notebooks that demonstrate the implementation of ML techniques using Python. The authors aim to equip physicists with the necessary skills to apply ML in their research and to understand the theoretical underpinnings of ML algorithms. The article also emphasizes the importance of understanding the difference between fitting and predicting, and the role of model complexity in achieving good generalization. Overall, the review serves as a comprehensive resource for physicists interested in learning and applying ML techniques in their work.