This review provides an overview of deep learning techniques in deterministic computational mechanics, focusing on five main categories: simulation substitution, simulation enhancement, discretizations as neural networks, generative approaches, and deep reinforcement learning. The aim is to help researchers identify key concepts and promising methodologies in this field. The review is structured to focus on general methods rather than specific applications, making it accessible to researchers new to the field. Key topics include data-driven modeling, physics-informed learning, pre-processing, physical modeling, numerical methods, and post-processing. The review also discusses advanced neural network architectures such as convolutional, recurrent, and graph neural networks, as well as generative models like autoencoders and generative adversarial networks. Finally, it explores the application of deep reinforcement learning in computational mechanics.This review provides an overview of deep learning techniques in deterministic computational mechanics, focusing on five main categories: simulation substitution, simulation enhancement, discretizations as neural networks, generative approaches, and deep reinforcement learning. The aim is to help researchers identify key concepts and promising methodologies in this field. The review is structured to focus on general methods rather than specific applications, making it accessible to researchers new to the field. Key topics include data-driven modeling, physics-informed learning, pre-processing, physical modeling, numerical methods, and post-processing. The review also discusses advanced neural network architectures such as convolutional, recurrent, and graph neural networks, as well as generative models like autoencoders and generative adversarial networks. Finally, it explores the application of deep reinforcement learning in computational mechanics.