Other Development Boards

You can use any sensor or development board as a rapid prototyping platform for machine learning applications. DEEPCRAFT™ Studio delivers an end-to-end workflow from real-time data collection and code generation to on-device deployment—across a wide range of hardware, enabling rapid iteration and a streamlined path from concept to embedded inference.

• Data collection: Stream real‑time data from any sensor or board using the Streaming Protocol for Real‑Time Data Collection. Using the Protocol, implement the firmware for your development board that stream data adhering to the required specifications, eliminating the need for data formatting. Refer to Registering sensors and boards for data streaming using Protocol version 2 for detailed instructions on firmware implementation.

  Refer to Data collection using other development Boards for collecting real-time data and your specific board documentation, for flashing boards with the implemented firmware.

• Code generation: After training and evaluating your model, generate the code using the ANSI C99 suitable for any platform. To know how to generate the code, refer to Code Generation for Other Development Boards and Code Generation using Graph UX.

• Deployment: Deploy the model to your development board using the APIs. To know how to deploy the model onto any development board, refer to Model Deployment on Other Development Boards and Edge API.