Workspaces on Onepanel
Meet Workspaces 👋 — a new cloud IDE for deep learning, powered by Onepanel
Workspaces is an interactive environment for developing and training deep learning models on Onepanel. You can run Jupyter notebooks, Python scripts, access the terminal, and much more. Even better, you can seamlessly toggle between CPU and GPU machines while you’re working, right when you’re ready for that extra computing power. All the files and data in a Workspace are preserved across sessions, just like your local computer.
In fact, we like to think of Workspaces as your persistent, on-demand deep learning computer in the cloud.
We’re huge fans of Project Jupyter, and we’re going all-in on JupyterLab — their next evolution of the Jupyter Notebook. JupyterLab is a fully extensible interactive computing environment, with all sorts of powerful features:
- Multiple notebook kernels in a single customizable UI
- File browser with rich outputs for images, CSVs, TSVs, and more
- Full terminal access
We think you’ll love the new look and feel of the interactive Jupyter experience on Onepanel— we’re already in ❤️.
On-demand GPU machine, configured with your framework of choice
It can be hard for an individual or company to keep up with the ongoing and relentless march forward by the popular deep learning frameworks — it seems like there’s a new TensorFlow version every few weeks!
Onepanel already provides dozens of pre-configured environments with the latest versions of most popular frameworks like TensorFlow and PyTorch, so that you don’t have to worry about keeping your configuration and drivers updated. With Workspaces, it’s gotten easier — you can now easily toggle between our deep learning environments in a single click.
Toggle between GPU and CPU machines
Even better, we wanted to make it as easy as possible to use GPU-compute in your projects. Previously, you could run a workspace on Onepanel with either a CPU machine or a GPU machine — but this was something you had to decide up front.
Now, you can start your development process using one of our cheaper CPU machines, and then seamlessly toggle your Workspace to a GPU machine — without losing your work, directly in your active Workspace, saving you both money and time.
Better insights into your machine and training metrics
You’re not alone — deep learning can still feel like a black box. But we’re working to fix that with Workspaces.
TensorBoard
Workspaces come with TensorBoard pre-enabled for all environments. Just click the TensorBoard button in the metrics bar to start visualizing your training process.
System Metrics
We’re also highlighting your machine’s current vitals through our System Metrics. At a glance, you’ll be able to see the latest and historical stats for your CPU or GPU utilization, disk space, RAM usage, and more.
Attaching datasets to active environments
Deep learning is nothing without data, and Onepanel is no stranger to managing your company’s datasets for training jobs. We’re also hosting thousands of popular public datasets, readily available for your next project.
Just select the dataset , download it using the CLI :
onepanel download onepanel-demo/datasets/dataset_nameHow to get started with Workspaces today
Click the “Create Workspace” button on any of your existing or new Onepanel projects. You’ll be able to either create a blank Workspace or bootstrap your Workspace with code from any public git URL — like GitHub, GitLab, or Bitbucket.
