Datasets

Overview

Datasets in AI Studio are dataset resources for ML workloads. They can reference model repositories and Hugging Face and use a chosen storage type.

Prerequisites

• A JupyterHub cluster exists — created by the Tenant Admin using the JupyterHub with KAI Scheduler cluster template
• You have a tenant user account — created by a Tenant Admin
• A storage class

Create Dataset

Step 1: Open Datasets

1. Log in to Armada Bridge as a tenant user.
2. In the left sidebar, click AI Studio. In the new tab that opens, click Datasets in the sidebar.

Step 2: Create New Dataset

1. Click Create New Dataset.
2. Enter Dataset Name, select Storage Type, add Description, enter Model Repo Name and Huggingface Token, then click Create Dataset.

Step 3: Verify Creation

1. Confirm the dataset appears in the Datasets tab.
2. Wait for status to change from Processing to Created (may take a few minutes).

Next Steps

• Create and Access Workspace — Create a JupyterHub server and attach your datasets.