Quick Start

Get started with your MLOps workflow on the NimbleBox.ai platform in minutes. Use it.

Pre-requisites

To use NimbleBox.ai, you will need to create a NimbleBox.ai account​

With that, you will be able to create notebooks, start training models and collaborate with your team on projects.

Creating a new Project

On the NimbleBox Dashboard, click on 'New Project' or 'Create ' button to get started.

Form when you click on 'New Project' or 'Create'

You may start a new project using any of the following methods:

  • New Blank Project: Create your project from scratch.

  • Clone from GitHub: Clone a pre-existing project from a GitHub repository.

  • Shared Instance: Use one of the instances shared with you by your teammate.

  • Download from Google Drive: Load a folder from Google Drive.

  • Load from Template: Use one of the templates that have been shared with you as part of your team.

Starting Your Instance

When you click on 'Start Project', a screen will pop up with the following properties:

  • Hardware Configuration: The hardware you'd like to requisition for your instance.

  • Type of Instance: Your instance can be used with or without a GPU.

  • Auto Shutdown timer: The pre-set time you want the instance to automatically shut down.

The configuration screen when you start an instance

Development Interface

The development interface

NimbleBox.ai provides three development interfaces to write code. You may choose the interface you'd like and change it during runtime.

  • ​ Jupyter Notebook

  • ​ VS Code

  • ​ Jupyter Lab

Additionally, the platform has integrations to the following interfaces for a better experience:

  • ​ MLFlow

  • ​ Terminal

  • ​ File Manager

  • ​ Google Drive

​