aim / README.md
osanseviero's picture
Remove pinning
f3b03c7
|
raw
history blame
1.83 kB
metadata
title: Aim
emoji: πŸ”₯
colorFrom: purple
colorTo: blue
sdk: docker
license: other
fullWidth: true

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

How to setup your own Aim space?

Setting up an Aim space is incredibly easy. There are two primary methods to do so:

  • Duplicating the space. Simply click on the three-dot icon and select Duplicate this Space from the drop-down menu that appears on the right.
  • Alternatively, you can create a new space using the Aim space, select the Docker option under Select the Space SDK, and choose AimStack.

Both options will generate an empty Aim space along with a newly initialized repository.

After successfully creating an Aim space, you can push your logs to be displayed. πŸš€ To do so, we need to replace the initial aim_repo.tar.gz file with your own Aim repo by running the following command:

tar -czvf aim_repo.tar.gz /PATH/TO/MY/AIM/REPO

And that's it! The space will be automatically updated, and you can start enjoying Aim's supercharged and beautiful UI. 😊

Learn more about Aim

Aim is an open-source, self-hosted AI Metadata tracking tool designed to handle 100,000s of tracked metadata sequences. Two most famous AI metadata applications are: experiment tracking and prompt engineering.

Aim provides a performant and beautiful UI for exploring and comparing training runs, prompt sessions. Additionally, its SDK enables programmatic access to tracked metadata β€” perfect for automations and Jupyter Notebook analysis.

For more information about Aim, please visit the GitHub page and documentation. You can also explore HuggingFace Spaces provided by Aim.