Quick Start

Prerequisites

To begin using Gradient, follow these preliminary steps:

  1. Optional: Create a team to invite collaborators

Now you can create Notebooks, Workflows, Models, Deployments, and more! Note: if you are self-hosting Gradient, please visit the Gradient Private Cloud section for more info.

In order to unlock Workflows, Models, and Deployments, please contact our solutions team.

Logging in for the first time

When you first log into the Paperspace Console, you can select Gradient from the product dropdown:

Create a Notebook

Notebooks can be created on the Notebooks tab. Just select a template, choose your instance type, and then click create.

Check out the FREE GPU option when launching Notebooks!

Select Notebooks > Create a Notebook to enter the notebook create flow

Check out the ML Showcase for a list of projects you can fork into your own account

You can stop, start, fork, and swap out the instance type anytime. Choose from a wide selection of pre-configured templates or bring your own. View the overview video to learn more:

Learn how to create a Jupyter Notebook. 1m48s.

Advanced MLOps

In order to unlock Workflows, Models, and Deployments, please contact our solutions team.

Create a Project

Projects organize your work. To create a Project, navigate to Projects and click Create Project, provide a name, and click create.

Running your first Workflow

Using Workflows requires creating a cluster.

You can run Workflows from the web interface or CLI. The first step is to create a new Workflow.

Steps

  1. Download or copy the sample Workflow YAML file to your computer

  2. Run the Workflow from the CLI

gradient workflows run \
--id <your-workflow-id> \
--clusterId <your-cluster-id> \
--path ./workflow.yaml

Note: We recommend stashing your API key with gradient apiKey XXXXXXXXXXXXX or you can add your API key as an option on each command. See Connecting Your Account.

The following command will create and start a Workflow that will run a sample project. Be sure to replace <your-workflow-id> with your Workflow ID and <your-cluster-id> with your Cluster ID.

Behind the scenes, your Workflow will be executed on your cluster. Congratulations! You ran your first Workflow on Gradient 🚀

Explore the rest of the platform

From Models to Deployments, there's a lot more to the Gradient platform. We recommend using the Web UI to explore the primary components and also be sure to install the CLI and check out the SDK.