Change History

Release Date	What's New
2020-12-10	Optimized the team labeling function. Managing Team Labeling Tasks Updated the GUI screenshots during dataset creation. Creating a Dataset Added the function of specifying AZs for creating or expanding a dedicated inference resource pool. Creating a Dedicated Resource Pool
2020-11-10	Optimized the display of the development environment when a notebook is created. Creating a Notebook Instance Added the data processing function. Data Processing
2020-09-21	Added the application authentication function for real-time services. Accessing a Real-Time Service (Application Authentication)
2020-09-11	Added the model evaluation and diagnosis function. Model Evaluation Overview Creating a Model Evaluation Job Viewing Evaluation Results Evaluation Metrics Managing Evaluation Job Versions Sample Code for Model Evaluation Added the function of viewing the evaluation result when viewing a training job. Viewing the Evaluation Result Adding the Evaluation Code
2020-08-06	Updated the description of the notebook function, optimized the document structure, and added common JupyterLab operations. Introduction to Jupyter Notebook JupyterLab Overview and Common Operations Added the dataset of the table type, which supports the DWS data source input. Table Added guidance on how to integrate real-time service APIs. Integrating a Real-Time Service Removed information about the previous version of AI Market and Data Management.
202-07-06	Added support for video data on the Dataset Management page. Creating a Dataset Video Labeling Optimized auto search jobs and added one sample. Example: Using Backbone Replacement to Optimize Built-in Algorithms on Ascend 910 Without Modifying Code Added the support for the GitHub code library in the development environment. Creating a Notebook Instance with a Git Repository Using the Git Plug-in Added the support for the PyTorch 1.4.0 engine in development environments and model import. Creating a Notebook Instance Importing a Meta Model from OBS Added the sample code of the TensorFlow 2.X custom script. TensorFlow 2.1 Optimized the page for creating a model conversion task. Compressing and Converting Models
2020-06-08	Added table dataset and CSV file import and release. Creating a Dataset Specifications for Importing Data from an OBS Directory Added task startup and management to data feature analysis. Data Features Enhanced DevEnviron by introducing JupyterLab and TensorFlow 2.1, integrated multiple basic capabilities of ModelArts, and improved user experience. Creating a Notebook Instance Using Sample Notebook Instances of ModelArts Synchronizing Files with OBS JupyterLab Overview and Common Operations Uploading Large Files to a Notebook Instance Downloading Large Files in Notebook Instances to a Local PC Optimized auto search job: Added hyperparameter search with fix_norm, data augmentation with adv_aug, and multiple NAS methods such as BETANAS to achieve the optimal precision in mobile setting on ImageNet. Multisearch is available using at least five lines of code. Introduction to Auto Search Jobs Creating an Auto Search Job Code Compilation Specifications YAML Configuration File Description Example: Replacing the Original ResNet-50 with a Better Network Architecture Example: Searching for Hyperparameters Using Classic Hyperparameter Algorithms Example: Searching for Network Architectures Using the MBNAS Algorithm Example: Implementing Auto Data Augmentation Using a Preset Data Augmentation Policy Example: Using Multisearch Added TensorFlow 2.1. When creating a notebook instance or using a frequently-used framework to create a training job, TensorFlow 2.1 is ready to use. Using Frequently-used Frameworks to Train Models Introduction to Notebook Added the Python 3.7 runtime environment to Model Management. Importing a Meta Model from OBS
2020-04-07	Supported training based on Ascend 910. AI Market provides cloud algorithms that can be used for training based on Ascend 910 training. Added the inference capability based on Ascend 310. Deploying a Model as a Real-Time Service Deploying a Model as a Batch Service Deploying a Model as an Edge Service
2020-03-25	Added the description of the limited-time free flavors that can be used in ExeML, training jobs, and notebooks. Trying ExeML for Free Experiencing AI Development Lifecyle for Free Trying Notebook for Free Added cloud algorithms to AI Market to quickly start model training without coding.
2020-03-20	Added the algorithm management function. You can create algorithms and subscribe to algorithms from AI Gallery. Overview Creating an Algorithm Creating a Training Job Using an Algorithm Optimized the description of training jobs based on different algorithm sources. Using Existing Algorithms to Train Models Using Built-in Algorithms to Train Models Using Frequently-used Frameworks to Train Models Using Custom Images to Train Models Added support for team labeling for datasets of the text triplet, text classification, and named entity types, and updated the description of the data labeling function. Data Management Overview Updated all screenshots in this document because ModelArts UI is upgraded.
2020-02-21	Added two conversion templates. Conversion Templates
2020-01-07	Added the function of filtering hard examples for data management, real-time services, and batch services. Confirming Hard Examples Collecting Data Collecting Data Hid the old data management module.
2019-12-03	Added four sections for the model import function based on scenarios. Importing a Meta Model from a Training Job Importing a Meta Model from a Template Importing a Meta Model from a Container Image Importing a Meta Model from OBS Supported team labeling task management for datasets that support team labeling. Managing Team Labeling Tasks Added the text triplet dataset type. Text Triplet Optimized the sections about data management. Data Management Overview - Deleting a Dataset Added the function of managing auto search jobs. Introduction to Auto Search Jobs - Example: Replacing the Original ResNet-50 with a Better Network Architecture
2019-11-06	Added the Custom option for Instance Flavor, which is configured during edge service deployment. Deploying a Model as an Edge Service
2019-10-17	Optimized and added functions for data management based on software changes. Updated descriptions in all related sections and added the following content: Deploying a Model in One Click Data Features Team Labeling Overview Added sample code of custom scripts (including frequently-used engines). Examples of Custom Scripts Added monitoring description. ModelArts Metrics
2019-09-30	Enriched and optimized the description of model templates. Model Template Overview Optimized the description of model package specifications and provided more model package examples. Model Package Specifications Added the model conversion function. Model Compression and Conversion
2019-08-08	Added the auto stop function for real-time services and event information to the service details page. Deploying a Model as a Real-Time Service Viewing Service Details
2019-07-18	Modified the way of selecting the AI engine for notebook instances. Introduction to Notebook Creating a Notebook Instance
2019-06-30	Added the function of uploading large files in notebook instances. Uploading Large Files to a Notebook Instance
2019-06-20	Added the description of the input and output mode of the model templates. Built-in Object Detection Mode Built-in Predictive Analytics Mode Undefined Mode
2019-06-03	Added the function of importing a template from an existing template. Added Model Templates. Modified Importing a Model. Added the Multi-Engine(Recommend) engine to the development environment. Modified Introduction to Notebook and Creating a Notebook Instance.
2019-05-31	The original ModelArts User Guide is divided into three documents: ModelArts User Guide (ExeML), ModelArts User Guide (AI Beginners), and ModelArts User Guide (Senior AI Engineers). This document is ModelArts User Guide (Senior AI Engineers). In addition to ExeML, this document contains the operation guide to the ModelArts management console. In this update, the outline is modified and the description of each chapter is optimized.
2019-04-16	Modified the following content: Preparations DevEnviron
2019-04-12	Added the following content: Introduction to Image Classification Introduction to Object Detection Data Labeling - Text Labeling Data Labeling - Speech Labeling Modified the following content: Model Management Image Classification - Model Training Object Detection - Model Training DevEnviron Real-Time Services
2019-04-04	Added the following content: Data Labeling - Sound Classification
2019-04-01	Modified the following content: Before You Start Preparations ExeML Data Labeling-Beta Training Jobs Model Management Service Deployment AI Market
2019-03-25	Added the following content: Data Labeling-Beta
2019-03-18	Modified the following content: Image Classification, Object Detection, and Predictive Analytics Dedicated Resource Pools
2019-02-22	Added the following content: Added Sound Classification to ExeML to describe how to build a sound classification model and perform sound classification.
2019-01-21	Added the following content: Model Package Specifications Synchronizing Files with OBS and Using ModelArts SDKs
2018-12-21	Modified the following content: Importing a Model Real-Time Service, Batch Services, and Edge Services
2018-12-03	Modified the following content: Modified the structure and content of chapter 3.
2018-11-15	Modified the following content: Modified the path and procedure for downloading the sample data for creating models using a built-in algorithm.
2018-11-08	This is the first official release.

Last Article: Migrating the Data of MLS to ModelArts

Did this article solve your problem?

Thank you for your score！Your feedback would help us improve the website.

Products

Compute

Application

Dedicated Cloud

Storage

Management & Deployment

Migration

Network

Enterprise Intelligence

Video

Database

Edge Cloud Services

DevCloud

Security

Cloud Communications

Internet of Things

Solutions

Industry-Specific Solutions

General-Purpose Solutions

Security

DevOps

Enterprise Intelligence

Essential Platform

Big Data

Visual Cognition

Speech and Semantics

Support

Help Center

Customer Services

Developers

Console

语言 - Language

中国站 - 简体中文

中国站 - English

International - 简体中文

International - English

Help Center

Change History