Shulou(Shulou.com)12/24 Report--

JFrog, a streaming software company and a provider of enterprise software supply chain platforms, launched ML model management, the industry's first set of features designed to simplify machine learning (ML) model management and security. The new ML model management capabilities in the JFrog platform align AI delivery with the enterprise's existing DevOps and DevSecOps practices, thereby accelerating, protecting, and managing the release of ML components.

Yoav Landman, co-founder and CTO of JFrog, said: "Today, data scientists, ML engineers and DevOps teams do not have a common process for delivering software. This often leads to friction between teams, difficulties of a certain size, and a lack of standards for management and compliance in the overall product portfolio. Without Python and the software packages it depends on, machine learning model artifacts are incomplete and are usually serviced by Docker containers. Our customers have regarded JFrog as the gold standard for product management and DevSecOps processes. Data scientists and software engineers have developed modern AI functions, and they are already native users of JFrog. As we introduce machine learning model management and model security and compliance into a unified software supply chain platform to help them deliver trusted software on a large scale in the AI era, this release is a logical next step. "

More and more users are using AI and ML. According to IDC research, the global AI / ML market, including software, hardware and services, is expected to grow by 19.6% in 2023, to more than $500 billion. However, as more and more ML models are put into production, end users are often faced with challenges such as cost, lack of automation, lack of expertise and scalability.

"it takes a lot of time and effort to deploy the ML model into production from beginning to end," said Jim Mercer, vice president of research for IDC DevOps and DevSecOps. However, even if it is put into production, users will face challenges such as model performance, model drift and deviation. As a result, having a single recording system helps automate the development, continuous management, and security of the ML model, and all other components are packaged into the application, providing a compelling alternative to optimizing the process. "

With JFrog's new ML model management features, enterprises can:

Hugging Face, a common ML repository commonly used by ● agents, caches the open source AI model that companies rely on to make it closer to development and production, preventing deletions or modifications.

● detects and blocks the use of malicious ML models.

● scans the ML model license to make sure it complies with company policy.

● stores self-developed or internally enhanced ML models and configures strong access control and version history to improve transparency.

● packages and distributes the ML model as part of any software release.

Yossi Shaul, Senior Vice President of products and Engineering at JFrog, said, "more and more companies are starting to incorporate ML models into their applications, and as some government regulations require software vendors to clearly list what is in their software, we believe that these guidelines will also cover ML and AI models in the near future. We are pleased to provide our customers with easy ways to agent, store, guarantee and manage models and other software components to help them accelerate their pace of innovation while preparing for future needs. "

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.