Shulou(Shulou.com)06/01 Report--

The main content of this article is to explain "what is Dagli". Interested friends may wish to take a look. The method introduced in this paper is simple, fast and practical. Let's let the editor take you to learn what Dagli is.

Although the development of machine learning is growing every day, a survey from Algorithmia shows that most enterprises spend 8 to 90 days developing ML models. Most people blame the inability to scale, followed by challenges in model repeatability, such as lack of official approval and lack of tools.

LinkedIn recently opened up the source code for Dagli, a machine learning library for Java and other JVM languages. The library allows you to easily draft model pipes that are error-resistant, understandable, modifiable, maintainable and deployable without incurring a technical burden.

A directed acyclic graph consists of vertices and edges, each edge pointing from one vertex to another. Using Dagli, model pipelines can be represented as directed acyclic graphs for training and reasoning. The Dagli environment prevents most possible logic errors by providing pipelined definitions such as static types, almost universal immutability, and other features.

Jeff Pasternack, a natural language processing scientist at LinkedIn, says machine learning models are usually part of an integrated pipeline. This makes the construction, training and deployment of production pipelines more challenging. In order to train and reason at the same time, repetitive or external work is usually needed to generate inelastic adhesive code that complicates the future development and maintenance of the model.

Dagli runs on the server, Hadoop, command line interface, IDE, and other familiar JVM settings. Many pipeline components can be used directly, including neural networks, logical regression, FastText, gradient enhanced decision trees, cross-validation, cross-training, feature selection, data readers, evaluation, and feature transformation.

Dagli provides access to superior features that are easy to maintain and a production-ready AI model. This provides data professionals with an extensible model that can take advantage of existing JVM technology stacks for a long time. For inexperienced software engineers, Dagli provides API that can be used with the JVM language and tools to avoid common logic errors.

The main goal is to create efficient and production-ready models that are easy to write, modify, and deploy. Efficient production avoids the usual technical debt and long-term maintenance challenges. Dagli uses modern highly multicore processors and powerful graphics cards to effectively train these real models on a single machine.

Dagli is released after LinkedIn makes the LinkedIn Fairness Toolkit LiFT available. It is an open source software library designed to measure fairness in AI and machine learning workflows. Earlier, LinkedIn also released DeText, an open source framework for ranking, classification, and language generation tasks related to natural language processes. It uses the semantic matching of applied depth neural network to learn the member intention in search and recommendation practice.

At this point, I believe you have a deeper understanding of "what Dagli is", might as well come to the actual operation of it! Here is the website, more related content can enter the relevant channels to inquire, follow us, continue to learn!

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.