Shark is a fast, modular, feature-rich open-source C++ machine learning library designed for the development and real-world application of state-of-the-art machine learning and optimization algorithms. It provides methods for linear and nonlinear optimization, kernel-based learning algorithms, neural networks, support vector machines, and evolutionary algorithms. Shark offers a unique combination of computational intelligence techniques in a coherent, extensible framework suitable for both research and production applications. It is jointly maintained by researchers from the Institut für Neuroinformatik (Ruhr-Universität Bochum) and the Department of Computer Science (DIKU, University of Copenhagen), and is licensed under the GNU Lesser General Public License.
Visit shark's official website for product details and getting started.
Comprehensive API reference and user guides for getting started with Shark.
Access the source code, contribute to development, and report issues on Shark's GitHub page.
Join the community to ask questions, share experiences, and get help with Shark.