Getting started with ML Platform on Alibaba Cloud

I’ve been working with and advocating with Cloud Services for a long time now. Starting with Google Cloud platform way back in late 2017 and Amazon Web Services in the mid 2018’s, the late 2018 saw me take on a new Cloud Service that finally opened it’s data center just a few month back in India. While I’ve been talking about Alibaba Cloud or Aliyun, in particular their Machine Learning Platform, the feedback I’ve received was that for a few, the entire setup was a bit complicated, so I decided to write this article to make everything streamlined. So, let’s get started.

Once you actually reach the landing page for the Machine Learning API, you have an option for either of buying an instance, which is billed on a Pay-As-You-Go basis. The total service fee is equal to the billing fee of the component you use multiplied by the number of computing hours. But for now, we’d be concentrating on the Console.

Machine Learning Landing Page

When you click on Console, the next page you go to is the the Main Console, where in if you haven’t already created a project in DataWorks, you’ll see a message similar to this.

Machine Learning Main Console

So now the next step is navigating to the DataWorks Console and creating a project, but before that, there is another thing to be done, which is creating an Access Key.

WorkSpace Console

Creating an AccessKey is as simple as clicking the button and the rest it done for you.

Creating Access Key using single click

And now we start going reverse in the order through which we came here. After the AccessKey is created, we go back and resume creating a workspace. Once here since we intend to utilize their ML Platform for AI, we need to go and buy it.

NOTE – Since they offer 300 credits for the Individual Tier, we’d be utilizing those here. So after going forward with the payment, we come back to this console and activate the services we need and proceed.

Notice all the radio buttons are grayed out. This means that an instance for the same is to be bought, and so we move forward with buying one each for Machine Learning Platform For AI (PAI) and Data Integration . As mentioned above, since we’re utilizing the credits, we get these instances for free.

Machine Learning Billing and Algorithms

Once done with this, we finally create the workspace using a unique name and then click on the workspace name to reach this page.

From here on it’s pretty straightforward,i.e. create a table, unless if you forgot to create a RAM user, in which case you’d receive an error similar to this.

To create a RAM User refer to this link

Leave a Reply

Your email address will not be published. Required fields are marked *