Overview

Platform for AI of Alibaba Cloud is a machine learning or deep learning engineering platform intended for enterprises and developers. It provides easy-to-use, cost-effective, high-performance, and easy-to-scale plug-ins that can be applied to various industry scenarios. With over 140 built-in optimization algorithms, Platform for AI provides whole-process AI engineering capabilities including data labeling (PAI-iTAG), model building (PAI-Designer and PAI-DSW), model training (PAI-DLC), compilation optimization, and inference deployment (PAI-EAS).

AI Engineering Platform

Preparing Data

In the data preparation phase, PAI-iTAG provides intelligent data labeling services. PAI-iTAG supports different types of data, such as image, text, video, and audio, supports multimodal data labeling, and provides various labeling content components and topic components. You can use predefined labeling templates provided by PAI-iTAG or customize your own labeling templates. It also provides fully managed data labeling services that are outsourced.

Model Development

In the model development phase, you can use PAI-Designer and PAI-DSW to develop models. PAI-Designer Supports different types of hardware resources, including CPUs and GPUs, and features high throughput and low latency. It allows you to deploy large-scale complex models with a few clicks and perform elastic scale-ins and scale-outs in real time. It also provides a comprehensive O&M and monitoring system. PAI-DSW PAI-Blade provides universal optimization technologies that can be applied to different business scenarios. You can run your models at optimal inference performance and greatly improve the model QPS by using PAI-Blade to implement joint optimization.

Model training

In the model training phase, you can use PAI-DLC, which is a one-stop platform for cloud-native deep learning and training, to train models. It is compatible with predefined algorithm frameworks, allows you to customize algorithm frameworks, and supports mega-scale task execution for distributed deep learning. PAI-DLC features high flexibility, high stability, high performance, and ease of use.

Model Deployment

In the model deployment phase, you can use PAI-EAS to perform online inference and use PAI-Blade to implement inference optimization. PAI-EAS Elastic Inference Service Platform Supports different types of hardware resources, including CPUs and GPUs, and features high throughput and low latency. It allows you to deploy large-scale complex models with a few clicks and perform elastic scale-ins and scale-outs in real time. It also provides a comprehensive O&M and monitoring system. PAI-Blade Universal Inference Accelerator PAI-Blade provides universal optimization technologies that can be applied to different business scenarios. You can run your models at optimal inference performance and greatly improve the model QPS by using PAI-Blade to implement joint optimization.

AI Asset Management

Machine Learning Platform for AI provides full lifecycle management for major data assets used for AI-based production, such as models, datasets, and images, as well as output data. In addition, it allows you to share data assets with other users, analyze the training results of different models, and backtrack issues and exceptions. Machine Learning Platform for AI helps you improve performance and save costs throughout the process of AI development and application.