This topic provides an overview of Data Science Workshop (DSW) 1.0. DSW of Machine Learning Platform for AI is an Integrated Development Environment (IDE) on the cloud. It provides interactive development environments for developers of different levels.

DSW integrates with open-source JupyterLab and provides plug-ins for customized development. You can directly launch Notebook to write, debug, and run Python code without O&M configurations. DSW also provides a variety of computing resources and supports heterogeneous data sources. DSW offers all-in-one machine learning services. You can use EASCMD to deploy models trained in DSW as RESTful APIs to provide online services.

Features

  • Supports resource monitoring in real time. DSW can monitor CPU/GPU usage during algorithm development.
  • Supports more than one source data, such as MaxCompute, Object Storage Service (OSS), and Network Attacked System (NAS) file systems.
  • Allows you to write and run SQL statements.
  • Supports multiple resource types, including a variety of CPUs and GPU models.
  • Allows you to switch among different types of resources, which reduces resource usage costs.
  • Provides built-in big data development packages and algorithm libraries, and allows you to install third-party libraries.

Development environments

Development environment
Section Description
Toolbar
Top navigation pane and shortcut tools
File list
Coding area
Resource usage graphs

Resource specifications

Resource type Resource specification Region
pai.medium.1xv100 GPU V100
  • China (Beijing)
  • China (Shanghai)
  • China (Hangzhou)
  • China (Shenzhen)
pai.medium.1xt4 GPU T4
  • China (Hangzhou)
  • China (Shanghai)
  • China (Beijing)
  • China (Shenzhen)
  • China (Hong Kong)
  • Singapore (Singapore)
  • Malaysia (Kuala Lumpur)
  • Indonesia (Jakarta)
  • Germany (Frankfurt)
  • India (Mumbai)
pai.medium.1xp100 GPU P100
  • China (Hangzhou)
  • China (Shanghai)
  • China (Beijing)
  • China (Shenzhen)
  • China (Hong Kong)
  • Singapore (Singapore)
  • Malaysia (Kuala Lumpur)
  • Indonesia (Jakarta)
  • Germany (Frankfurt)
  • India (Mumbai)
Note A DSW instance deployed in the China (Beijing) region supports up to eight GPUs.
pai.medium.1xm40 GPU M40 China (Shanghai)
pai.large.2core4g 2 CPU Core+4 GB
  • China (Hangzhou)
  • China (Shanghai)
  • China (Beijing)
  • China (Shenzhen)
  • China (Hong Kong)
  • Singapore (Singapore)
  • Malaysia (Kuala Lumpur)
  • Indonesia (Jakarta)
  • Germany (Frankfurt)
  • India (Mumbai)
pai.xlarge.4core8g 4 CPU Core+8 GB
pai.2xlarge.8core16g 8 CPU Core+16 GB
pai.4xlarge.16core32g 16 CPU Core+32 GB
pai.6xlarge.24core48g 24 CPU Core+48 GB

Computing frameworks

Billing method Framework Version
Subscription TensorFlow 2.0 and 1.12.3
PyTorch 1.3.1
Pay-as-you-go TensorFlow 2.0 and 1.14.0
PyTorch 1.1.0