Ekkono is an embedded edge machine-learning library, purpose-built for IoT. The unique design makes it resource efficient with a small footprint. Still, Ekkono does not compromise on functionality:
• Prediction and Forecasting
• Batch and Incremental Learning
• Change and Anomaly Detection
• Predictions with Confidence
• Attribution/Sensitivity Analysis
• Simulation of Alternative Scenarios
Ekkono’s product is 100% software and totally platform- agnostic. The core is a C++ library. The product is designed for programmers and the API offers bindings to several programming languages (e.g. C# and Python), and the option to generate C code. The SDK (software development kit) is a comprehensive toolbox to support implementation and integration. Tools that help with algorithm selection and optimization enable a programmer with limited data science experience to deploy advanced edge machine-learning.
Ekkono does learning on streaming (sensor) data. The product supports execution of pre-trained models, as well as incremental learning at the edge, where the data pipeline is automated and built into the process. It supports a number of machine-learning techniques, including decision trees, random forest, and neural networks. Besides incremental learning, Ekkono also offers unique features such as conformal predictions, efficient techniques for anomaly and change detection, and more.
This comprehensive product lets companies harmonize on one solution. Machine-learning is a lifelong journey where every feature improves its capabilities. Ekkono empowers companies with a programmable tool to rapidly implement edge intelligence – yourself.