IBM Watson Studio is IBM’s latest offering for developers and data scientists. It empowers teams to build AI models, optimize decisions on IBM Cloud Pak for Data, and unify their work in an open architecture environment. Watson Studio includes Jupyter notebooks which allow you to run code-based or visual data science tools like PyTorch, TensorFlow, scikit-learn with IBM’s ecosystem tools. It also comes with a graphical interface called JupyterLab that will be familiar to anyone who has used the popular IDE Eclipse before.
Watson Studio is designed to make it easier for data scientists to build models and work with large amounts of data. The platform provides a wide variety of services and tools that can be used for tasks such as data preparation, model training, deployment, and monitoring. The platform also includes a web-based IBM Watson Studio Navigator which provides access to IBM products and services, including Data Science Experience (DSX), Cloud Machine Learning, Big data Platform (IBDP), and Watson.
If you are looking for an open-source platform to conduct your data science workflows, Watson Studio might be the best option for you. Because, it empowers teams to build AI models, optimize decisions on IBM Cloud Pak for Data, and unify their work in an open architecture environment.
The platform is available for both private and public clouds.
History of IBM Watson Studio
IBM Watson Studio was first released in February of 2018 as part of IBM Cloud Pak for Data. It is built on top of the IBM Cloud Platform which gives users access to a wide variety of IBM services and products.
Watson Studio is a part of IBM’s effort to provide an open architecture for AI. IBM has made a number of acquisitions in recent years to bolster its AI capabilities, including the purchase of The Weather Company, Truven Health Analytics, and Clearleap. The platform is also built on top of other IBM products such as IBM Big data Platform, IBM Data Science Experience, and IBM Cloud Pak for Analytics.
Creating an Account in IBM Watson Studio
You need an IBM Cloud account to enjoy its products and services.
To create a My IBM account inside IBM, you first need to go to the IBM cloud Registration and create your account.
Then head over to Watson studio by clicking here.
You can try IBM Watson at no cost. Click on the try on cloud at no cost.
By clicking so will take you to follow the page, where you are going to choose your region. Here choose the closest data center to you.
I chose Dallas. When you choose your region it will open up a few more options shown below.
If you didn’t create an IBM cloud account at the first step, now you can. Or simply log in with your IBM Cloud account credentials by clicking on “Log in with your IBMid“.
After Watson Studio Apps are ready to use, click on the “Go to IBM Watson Studio“.
Finally, you are looking at the IBM Watson Studio user interface.
Understanding IBM Watson Studio User Interface
The Home page is the first page that you see when you log in to the platform. It provides an overview of your account. And it contains tiles that provide links to various aspects of the platform such as Recent Projects, Recently Edited Files, and Recently Viewed Services.
The home page provides a variety of options for users. The left-hand navigation bar contains links to all of the major components. These components include Home, Workspaces, Projects, Notebooks, DataSources, Services, and more.
The Workspaces tab provides a view of all of the workspaces that you have access to in Watson Studio. A workspace is a container for files, projects, and notebooks. The platform supports multiple workspaces which can be helpful for organizing your work.
The Projects tab provides a view of all of the projects that you have access to in IBM Watson Studio. A project is a collection of files, data sources, services, and notebooks that are used together to solve a problem or create a model. The platform supports multiple projects which can be helpful for organizing your work and sharing files with other users.
How to create a new project in Watson Studio?
IBM Watson Studio supports multiple projects which can be helpful for organizing your work and sharing files with other users.
If you don’t see the new project button, just click on the left pane, click on the projects. And then click view all projects.
There you have it under all projects. Now you can start your new Watson Studio project.
Notebook Creation in IBM Watson Studio
A notebook is an interactive document where data scientists, developers, and analysts can explore ideas, create models, or train existing models using data. Watson Studio supports multiple notebooks which can be helpful for organizing your work.
We’ll discuss more about notebooks later in this article.
The data sources tab
The Data Sources tab provides a view of all of the data sources that you have access to in Watson. A data source is a collection of files or folders that contain data for use. The platform supports multiple data sources which can be helpful for organizing your data.
IBM Watson Studio Services
The Services tab provides a view of IBM Watson Studio services that you have access to. The platform supports multiple services which can be helpful for sharing resources across projects and workspaces, including IBM DSX, IBM Big Data Platform (IBDP), IBM Cloud Machine Learning (ICML), and IBM Watson. Watson Studio also supports IBM Cloud Object Storage (ICOS) as a service, IBM Big Query as a service, and IBM Cognos Analytics Platform as a service which can be helpful for sharing resources across workspaces.
Jupyter Notebooks in Watson Studio
Creating Watson Studio Notebooks Vs Importing Jupyter Notebook
Notebooks can be created using the Watson Studio UI or via a Jupyter Notebook server. The first option allows users to create new notebooks right from the Watson Studio platform without downloading or using any additional software.
The second approach is useful if you already have an existing notebook that you want to use in the IBM platform or require specific configuration options that are not available in the Watson Studio UI itself.
The IBM Watson Studio UI makes it easy to create new notebooks, while the Jupyter Notebook server provides more flexibility for working with existing notebooks or configurations. Try out both options and see which one works best for you!
Creating a Notebook within Watson Studio
In order to create notebooks, first, you have to start a project.
Click on the new project button. You will be given two options. I’m starting with an empty project in the sake of this article.
Defining details for your new Watson Studio Project
Define details for your project and click create.
Adding assets to the Watson Studio new Project
Finally, you can add assets to your new project. You can add your dataset and the notebook as assets.
Adding a notebook asset
Click add to project and choose notebook option, shown below.
Enter the name, the description for your notebook. Choose the runtime as well.
As I have already mentioned earlier in this post, here we have the option to import notebooks outside of the Watson Studio. You may import the notebook from your local drive or from a jupyter notebook server URL.
Start writing in your notebook
Finally, you can write your code.
Interested in Edge Ai Applications? You may train your edge ai application in IBM Watson Studio and deploy the trained model in your own device.
Linking GitHub to Watson Studio
IBM Watson Studio is designed to be an open, extensible environment where users can easily create and deploy various IBM services. Developers use the IBM Cloud tooling in GitHub to store their IBM-specific source code files from a GitHub repository directly into Watson Studio, without having to launch any other tools or set up IBM APIs. They also have options for deploying IBM Watson services directly from IBM Cloud.
IBM’s Git Virtual File System (GVFS) replaces large files with virtual pointers within a git repository. So, the developers can work efficiently without worrying about taking up too much disk space or clogging networks with huge data transfers during collaboration.
IBM Watson Studio pricing
Watson Studio is available as a free IBM Cloud trial. The pricing starts at $59/mo with an annual commitment for commercial use or $29/mo without the contract requirement.
The subscriptions are also available on a pay-as-you-go basis. Hence IBM Watson Studio offers flexibility for customers to use the software as needed and only pay for what they use. For more information on Watson Studio pricing, visit the official website.
IBM Watson Studio is a powerful tool that can be used to create and share data science projects. It gives you the ability to store, manage, and analyze your datasets with ease – all without much writing code! If you are a coding expert you can write codes from scratch and create any solution you wish! We’ve covered everything from what it takes to set up an account in t1e platform and how to get started using this platform for yourself. You should now have enough information about Watson platform pricing as well as its features so you can decide if it would be right for your needs. If not, we recommend checking out some of our other blog posts on different types of analytics software like Einstein Discovery in Tableau, Azure machine learning, or Google Data Studio.
Our next blog post is going to be about a complete tutorial on a machine learning solution deployment in IBM Watson Studio. So be sure to subscribe to neuralnetlab newsletter and stay in touch!