Aim: Record, search and compare ML training runs
1–10 of 27 posts
Re: Aim: Record, search and compare ML training runs
#2Honestly, it's really hard for me to tell from the documentations if this is something that is self-hosted, is run locally or is a SaaS product. The phrase "open source" shows up nowhere I can see for example.
As best I can tell it's an open source version of comet or wandb that you run locally (or I guess host somewhere). Is that right?
Re: Aim: Record, search and compare ML training runs
#3I've seen a lot these projects recently. I have used MLFlow for a while and enjoyed it. How does this compare?
Re: Aim: Record, search and compare ML training runs
#4Honestly, it's really hard for me to tell from the documentations if this is something that is self-hosted, is run locally or is a SaaS product. The phrase "open source" shows up nowhere I can see for example. As best I can tell it's an open source version of comet or wandb that you run locally (or I guess host somewhere). Is that right?
From what I can glean from the docs, you run the UI locally which reads from experiments from `.aim/`
The experiments results could be populated by running the training job remotely and committing the `.aim/` directory afterwards.
It's kinda like a UI for DVC
Re: Aim: Record, search and compare ML training runs
#5I've seen a lot these projects recently. I have used MLFlow for a while and enjoyed it. How does this compare?
or how either compares with Tensorboard and its HPARAMS tab? (PSA: PyTorch Lightning makes it even easier to use).
Re: Aim: Record, search and compare ML training runs
#6Honestly, it's really hard for me to tell from the documentations if this is something that is self-hosted, is run locally or is a SaaS product. The phrase "open source" shows up nowhere I can see for example. As best I can tell it's an open source version of comet or wandb that you run locally (or I guess host somewhere). Is that right?
From what I can glean from the docs, you run the UI locally which reads from experiments from `.aim/` The experiments results could be populated by running the training job remotely and committing the `.aim/` directory afterwards. It's kinda like a UI for DVC
yes, it looks like it's meant to be run in a similar way to Tensorboard
Re: Aim: Record, search and compare ML training runs
#7I've seen a lot these projects recently. I have used MLFlow for a while and enjoyed it. How does this compare?
Yes, I was looking for the comparison to local MLFlow and Tensorboard (and Losswise, and Weights & Biases, and all the remote options in this ecosystem). It looks nice, but without that, it's difficult to know whether it's worth spending more time looking into it.
Re: Aim: Record, search and compare ML training runs
#8Oh nice, it's like open-source version of W&B, and seems much better than Tensorboard. How many experiments I can compare at the same time without crashing chrome? :)
Re: Aim: Record, search and compare ML training runs
#9Hi all I am one of the co-authors to this project, and will try to answer all the Qs here.
Was just forwarded this link - one of the community members must have posted it.
Re: Aim: Record, search and compare ML training runs
#10This is great stuff :))