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

Artificial intelligence is the most important technological revolution and driving force in the next decade, and it is playing an increasingly important role in various industries. It complements the development of big data. It not only pushes human society into a more intelligent world, but also brings inestimable value to the application of data.

From November 28 to 30, Flink Forward Asia 2019 artificial Intelligence will focus on new technologies and applications of Flink in machine learning, including: what is the progress of Flink machine learning? How to combine Flink with frameworks such as TensorFlow? What are the practical applications of Flink in machine learning? To show you the specific application of Flink machine learning practice and the latest technology landing case. Deep Learning On Apache Flink Chen long | Tencent Engineer this lecture will introduce a new Flink project, dlonflink. It can make Flink have the ability to run distributed training and real-time update model services, users can not only use Scala development, but also based on Python, only need to modify a few lines of code to run TensorFlow on Flink, but also support PyToch and MXNet. We will introduce in detail the design of data exchange between Java and Python and the implementation of framework management. Architecture and practice of big data Analysis and Deep Learning Model reasoning using Analytics-Zoo on Flink Shi Dongjie | Intel Senior Software architect Intel Senior Software architect. He has been engaged in the research and development of enterprise computing, risk control, big data analysis, cloud computing container orchestration, data analysis and artificial intelligence for many years, and is one of the contributors to Intel open source framework BigDL and Analytics-Zoo. The main sharing content is as follows: introduction to Analytics-Zoo: big data Analysis + AI platform based on Apache Spark, Tensorflow, Keras and BigDL. Deploy the deep learning application on big data to the production environment, support for Flink stream processing scenarios and use OpenVINO acceleration. A Cluster Serving architecture based on message subscription and Flink stream processing. Apache Flink + TensorFlow, Pan Guoqing, Ctrip Real-time Intelligent anomaly Detection platform Pan Guoqing | with the development of Ctrip big data in recent years, the technology of real-time computing is becoming more and more mature, and there are more and more real-time computing scenarios. Flink has gradually become one of the first choices of real-time computing engines, from simple real-time ETL to complex CEP scenarios Flink can be well controlled. This sharing mainly introduces how to build Ctrip real-time intelligent anomaly detection platform based on Flink and Tensorflow, which is used to solve many problems of rule alarm system, such as low accuracy, low timeliness, complex rule configuration and labor-consuming, to achieve millisecond delay and intelligent detection of business indicators, and to achieve a strong fault-tolerant mechanism based on Flink. Practice and open source of Apache Flink-based machine learning algorithm platform Yang Xu | Alibaba senior algorithm expert Alibaba Computing platform Division is working with the Flink community to develop an open source machine learning algorithm library, based on which users can more easily build high-performance Flink machine learning assignments. We hope to promote the development of the Flink community in the field of machine learning through open source. At the same time, we also welcome more developers to join hands with us to build a more powerful and complete Flink algorithm library. This sharing focuses on the technical accumulation and harvest of the team in the process of developing a high-performance machine learning algorithm library based on Flink, as well as the progress of open source. Apache Flink AI Ecosystem Effort Chen Wichao | Alibaba technical expert Gao Yi | Alibaba technical expert Flink is a distributed computing engine that supports batch-stream data processing. In the actual production of artificial intelligence use scenarios, Flink has some unique advantages, including feature engineering, online learning, online prediction and so on. In order to better support the use of artificial intelligence scenarios, the community and various ecological manufacturers have done some work. This speech will introduce the recent work progress of Flink in the artificial intelligence ecosystem. Flink ML Pipline: defines the API that runs machine learning workflows on Flink, including feature engineering, model training, model prediction, etc. Alink: is a Flink-based machine learning algorithm library. Flink ML Workflow: based on Flink, combined with deep learning framework TensorFlow / PyTorch, it provides users with API of all stages in the whole machine learning link, and supports both batch and stream operation mode. Streaming-based mini-batch iteration architecture:Flink supports iterative new architecture design in streaming mode.

The original link to this article is the original content of Yunqi community and may not be reproduced without permission.

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.