The Intel® Geti™ SDK: A Game-Changer for Rapid AI Model Development and Deployment in your production system

By Dr. Paula Ramos – AI Evangelist at Intel (LinkedIn profile)

In today’s fast-paced world, businesses require quick and accurate decision-making abilities. This is where Artificial Intelligence (AI) models come into play. AI models can analyze vast amounts of data, extract insights, and provide businesses with the intelligence required to make quick, data-driven decisions.

Organizations have a lot of challenges to implementing AI in their pipeline, which can span from data collection to long cycles involved in annotation to siloed workflows of model development pipelines. They need to invest in bringing the right data into the hands of the right people involving AI developers and domain experts together to work collaboratively on the model development.

Organizations have multiple teams who need to collaborate in the workflow, from software engineers developing artificial intelligence algorithms to data scientists managing and implementing computer vision models in real production scenarios and domain experts sharing their knowledge of the specific problem. By building a bridge between these different groups, organizations can create a more cohesive team that leverages the strengths and expertise of everyone. For example, data scientists can provide technical expertise, while domain experts bring business knowledge and insights. This collaborative approach can lead to more accurate and effective computer vision models that address the organization’s needs better.

With the Intel® Geti™ platform, Intel’s platform for building computer vision models, we aim to solve this exact problem. By easing the complexities of model development and enabling greater collaboration between teams, the platform reduces the time needed to build models. It unlocks faster time-to-value for digitization initiatives with AI.

There are two critical stages in the entire value chain – developing an AI model to solve a specific problem and deploying the model in production to generate real-time analysis. Both stages involve complex processes, and the inference stage can be particularly challenging. Inference is the process of using a trained model to predict on new data. Inference requires a robust infrastructure capable of handling high-throughput, low-latency, and low-energy consumption.

The OpenVINO toolkit by Intel helps customers solve their challenges setting up inference pipelines. With OpenVINO, businesses can optimize their AI models for the inference stage, allowing them to deploy their models efficiently and effectively.

So, with the Intel Geti platform, we are helping customers solve the model development stage challenges. Combined with the OpenVINO toolkit, we are addressing the challenges customers face in the deployment stages. The Intel Geti platform also offers a software development kit (SDK) that helps users to take advantage of easy-to-use functionalities. The SDK utilizes OpenVINO to build deployment pipelines and accelerate inference on various Intel hardware platforms, including CPUs, and GPUs, without needing to be an expert in computer vision. This SDK is going to be the focus of this blog.

The Intel® Geti™ SDK

The Intel Geti platform SDK is a Python package that contains tools to interact with an Intel Geti server via the REST APIs (Representational State Transfer Application Programming Interface). This SDK provides functionality for project creation from annotated datasets on disk, project downloading (images, videos, configuration, annotations, predictions, and models), project creation and uploads from a previous download, deploying a project for local inference with OpenVINO, getting and setting project and model configuration, launching and monitoring training jobs, and media upload and prediction.

The Intel Geti SDK is a game-changer for rapid AI model development and can help you leverage the power of OpenVINO Runtime or OpenVINO Model Server to improve your inference pipeline. This blog will explore the benefits of using the Intel Geti SDK for inference, including low latency, high throughput, optimized memory and energy consumption, and high accuracy. So, let’s dive in and explore the world of AI inference with the OpenVINO toolkit and the Intel Geti SDK.

How the Intel® Geti™ SDK works?

As mentioned earlier, the Intel Geti SDK interacts with the Intel Geti server over the REST APIs. This REST API is a set of guidelines and standards for creating web services that allow different software systems to communicate with each other. It defines how to format requests and responses, what data can be sent and received, and how to handle errors. A REST API is typically accessed through HTTP requests and can be used by developers to integrate different software systems or to build new applications. The Intel Geti platform uses the same REST APIs to connect to the server through HTTPS protocols.

Intel Geti SDK provides developers with tools and resources to build applications that can interact with a specific system or service, such as a REST API. Through pre-built functions and resources, SDK makes it easier for developers to build such integrations.

Advantages to using Intel® Geti™ SDK:

• Deploy your projects without worrying about pre and post-processing steps. Intel Geti SDK manages it for you in the backend of the library. The Intel Geti SDK is agnostic to the model architecture or the computer vision task. It doesn’t matter if it’s a detection model or a segmentation model, the deployment.infer() doesn’t change!
• Run the inference in different ways: 1) using the same Intel Geti Instance, 2) your local machine, or 3) other remote servers. The Intel Geti SDK uses OpenVINO Runtime in the background, wraps the models and gives you the options to run it locally or as a microservice.
• The Intel Geti SDK provides some demo examples using Jupyter notebooks. It is a great way to learn and explore the capabilities of integrating the SDK into your system. Thus, familiarity with the SDK can help you easily use the REST APIs to integrate interacting with the Intel Geti platform functionalities in your production system.

One of the best things about the Intel Geti SDK is the Jupyter notebook examples that come with it. These notebooks are designed to teach users how to use the SDK in a tutorial style, making it easy to get started. Through these notebooks, users can learn about different use cases for the package and see how the platform can be used to develop AI models quickly and efficiently.

Whether you’re a seasoned data scientist or just getting started with AI, the Intel Geti platform has everything you need to succeed in developing AI models.

Conclusion:

Our customers are already taking advantage of the platform and SDK capabilities in many ways. You can read about it in case studies and white papers.

If you want access to the Intel Geti platform, please connect with us by filling out this form. And suppose you are already using the Intel Geti SDK and have feedback to improve its functionalities further. In that case, we invite you to add your questions to our repository’s discussion section https://github.com/openvinotoolkit/geti-sdk. Enjoy the blog and the Intel Geti SDK notebooks!

#iamintel #intelgetiSDK #intelgeti #computervision #openvino #aiforall

About me:

Hi, all! My name is Paula Ramos. I have been an AI enthusiast and have worked with Computer Vision since the early 2000s. Developing novel integrated engineering technologies is my passion. I love to deploy solutions that real people can use to solve their equally real problems. If you’d like to share your ideas on how we could improve our community content, drop me a line! I will be happy to hear your feedback. Here is my LinkedIn profile: https://www.linkedin.com/in/paula-ramos-41097319/

Notices & Disclaimers:

Intel technologies may require enabled hardware, software, or service activation. No product or component can be absolutely secure. Your costs and results may vary. Intel does not control or audit third-party data. You should consult other sources to evaluate accuracy. Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade. No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document. © Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. Other names and brands may be claimed as the property of others.