#United States #Santa Clara #DeGirum #PySDK #Edge AI

DeGirum Launches the First Hardware-Agnostic PySDK

In a significant stride towards simplifying AI development at the edge, DeGirum, a frontrunner in edge AI solutions, has announced the release of its transformative hardware-agnostic PySDK. This cutting-edge software development kit is set to reshape the landscape of AI application development across diverse industries such as industrial automation, automotive, and smart retail.

Understanding the PySDK

The key highlight of DeGirum’s PySDK is its ability to bridge the gap between complex AI algorithms and practical application integration. With the new SDK, developers can incorporate AI features with minimal effort—adding functionality with just two lines of code, regardless of the hardware being used.

Wide Hardware Support

The PySDK is designed to accommodate numerous hardware accelerators, including but not limited to:

• - NVIDIA® GPUs
• - Google® Edge TPU
• - Intel® and AMD® processors
• - Texas Instruments® and NXP® devices
• - Raspberry Pi®

This extensive compatibility allows developers to select the optimal hardware for their projects without being constrained by vendor-specific limitations.

Comprehensive Features

One of the significant advantages of the PySDK is its comprehensive inference API. This API manages key tasks related to AI processing, including:

• - Decoding of data inputs
• - Image resizing
• - Pre-processing and post-processing of data

By automating these essential tasks, the SDK ensures that developers can focus more on creating innovative features rather than managing complex processes.

Scalability and Flexibility

The PySDK not only simplifies the initial development of AI applications but also provides scalability options. Developers can enhance their applications by expanding hardware capabilities without having to alter pre-existing code. This feature addresses a common concern in software development where scaling often requires significant code refactoring.

Seamless Transition from Development to Deployment

DeGirum has further integrated its PySDK with the DeGirum AI Hub, facilitating a smooth transition from an AI development environment to deployment. This integration allows developers to prototype AI applications using hosted resources and then transition to real-time running environments effectively.

The DeGirum AI Hub

The DeGirum AI Hub serves as a support platform for developers, providing access to over 1,000 production-ready machine-learning models. This platform not only simplifies the prototyping phase but also assists in the conversion of training checkpoints into deployable models across various hardware. With features like model zoos, developers can access a wealth of models and update them on-demand, alleviating the need to bundle these models within final applications.

Introducing DeGirumJS

In addition to the PySDK, DeGirum is launching DeGirumJS, a JavaScript SDK. This new tool is engineered to facilitate seamless integration of AI capabilities into web and mobile applications, allowing developers to run real-time AI models directly in browsers. This brings features such as real-time object detection and image classification to web-based platforms, expanding the utility and accessibility of AI technologies.

Conclusion

Founded by veterans of the semiconductor industry in 2017, DeGirum continues its mission to make AI development more manageable and efficient. The launch of the hardware-agnostic PySDK, along with the capabilities of the AI Hub and DeGirumJS, equips developers with a holistic toolkit, empowering them to tackle AI challenges at the edge with agility and sophistication. As edge computing continues to be a focal point in AI development, DeGirum is poised at the forefront, driving advancements for a smarter, interconnected future.