AimStack
Published in

AimStack

Aim 3.10 — Visualize terminal logs, M1 support & better query autocomplete

Hey team, Aim 3.10 is now available!

We are on a mission to democratize AI dev tools. Thanks to the awesome Aim community for the help and contributions.

Here is what’s new in Aim 3.10:

  • Visualize terminal logs
  • M1 support
  • Better autocomplete experience
  • Aim citation available
  • LightGBM integration
  • CatBoost integration

Check out the 3.10 release milestone and the GitHub Release Notes.

Special thanks to Daniel Wessel, arnauddhaene, uduse, lukoucky, Armen Aghajanyan and others for contributions and feedback. We really appreciate it.

Visualize terminal logs

When it comes to automating training of multiple runs with job schedulers or workload managers on a cluster, it becomes hard to track the terminal logs of the runs.

Now Aim automatically streams the terminal logs to the UI. Near-real-time.

The terminal logs can be turned off if the run instance is created with the following flag in place:

aim_run = Run(capture_terminal_logs=False)

Check out more about this feature in Aim docs.

M1 support

With the awesome work by the PyTorch team on enabling support for GPU-accelerated PyTorch training on Mac, there has been a huge demand to enable aim on M1 as well.

Aim now supports M1 too. Now you can use Aim with PyTorch on Mac to track and deeply compare your experiments 😊.

Better autocomplete experience

We have integrated a rich code autocomplete system as the Aim community loves to deeply query their training runs. This has been a highly requested improvement and a huge productivity booster. Expect more improvements here ❤️

Aim citation available

Now you can cite Aim from your paper if you are using Aim to compare your experiments. The citation file.

Thanks to the researchers from Meta AI Research for the incentive and initiative. 🙌

CatBoost integration

CatBoost is one of the fastest Gradient Boosting on Decision Trees library. Aim now supports CatBoost out of the box. Check out the Aim CatBoost docs here.

Here is a short example that shows how to use AimLogger for CatBoost.

LightGBM integration

LightGBM is one of the most widely-used and battle-tested gradient boosting frameworks. Due to high-demand we have also added an out-of-the-box Aim LightGBM integration.

Here is a short code example and the docs on how to use the integration.

Learn More

Aim is on a mission to democratize AI dev tools.

We have been incredibly lucky to get help and contributions from the amazing Aim community. It’s humbling and inspiring.

Try out Aim, join the Aim community, share your feedback, open issues for new features, bugs.

This article was originally published on Aim Blog. Find more in depth guides and details of the newest releases there.

--

--

--

Aim logs your training runs, enables a beautiful UI to compare them and an API to query them programmatically.

Recommended from Medium

Interview: Gavin Lew and I Talk About the Key to AI Success and User Experience (Part 2)

Creating a perfect economic strategy with information and Intelligence

When Creativity meets A.I.

Element.AI CEO Jean-François Gagné: “Big companies could be hurt by unwise AI implementation’’.

Ethical Analysis of Netflix

Google and AI

I spent a week at Google AI - here are the lessons I learned

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Gev Sogomonian

Gev Sogomonian

Aim co-creator. Co-founder and CEO AimHub. Prior Altocloud (acqu. Genesys). Runner, Swimmer.

More from Medium

Hugging Face Collaborates with Microsoft to Launch Hugging Face Endpoints on Azure

How YerevaNN researchers use Aim to compare 100s of models

Building Production Docker Containers from AI / ML Model Files with Chassis

Three stages of the MLOps pipeline: Data science and model training, software engineering and model training, and production deployment for those models

Apple Silicon Experiment 4 — PyTorch mps backend