Machine learning single-machine algorithm arrangement system and method

Abstract

The invention provides a machine learning single-machine algorithm arrangement system and method. The system is characterized by comprising a machine learning algorithm dependency module used for pre-defining a machine learning algorithm container mirror image and storing the created machine learning algorithm container mirror image; an application container platform which is used for creating an application container based on the machine learning algorithm container mirror image; a user-defined business algorithm module which is used for storing the business algorithm codes in a code warehouse or storing the business algorithm codes in a distributed file system as the static files; and an algorithm environment framework which is used for embedding the application container, and particularly relates to a programming language Python implementation method and an interface. According to the present invention, the second-level creation of the machine learning algorithm environment is realized, the completion of the algorithm execution is ensured, and the effect of supporting the multi-algorithm task load is achieved.

Description

technical field [0001] The invention relates to computer technology, in particular to a machine learning stand-alone algorithm arrangement system, method and algorithm operation method. Background technique [0002] Docker is an LXC-based application container technology. As a general container solution, it can control the resource allocation of container applications in the host machine, including controlling the number of CPU cores, memory, network, and disk volume size. A complete Docker generally consists of the following parts: dockerClient client, Docker Daemon daemon process, Docker Image image, DockerContainer container. Docker uses the C / S architecture Docker daemon as the server to accept requests from customers and process these requests (creating, running, and distributing containers). The client and server can run on the same machine, or communicate through socket or RESTful API. Docker daemon generally runs in the background of the host host, waiting to recei...

AI Technical Summary

Problems solved by technology

[0011] 1. The operating environment that cannot support multi-tenant and multi-task (single-computer algorithm calculation) can be quickly built, and the resource overhead is large;

[0012] 2. Does not support flexible configuration of environmental parameters, ready-to-use, timely recovery, poor programmable management

Images