• by gbickford on 4/24/2024, 2:35:27 AM

    > Relationship with CVNets

    > CoreNet evolved from CVNets, to encompass a broader range of applications beyond computer vision. Its expansion facilitated the training of foundational models, including LLMs.

    We can expect it to have grown from here: https://apple.github.io/ml-cvnets/index.html

    It looks like a mid-level implementations of training and inference. You can see in their "default_trainer.py"[1] that the engine uses Tensors from torch but implements its own training method. They implement their own LR scheduler and optimizer; the caller can optionally use Adam from torch.

    It's an interesting (maybe very Apple) choice to build from the ground up instead of partnering with existing frameworks to provide first class support in them.

    The MLX examples seem to be inference only at this point. It does look like this might be a landing ground for more MLX specific implementations: e.g. https://github.com/apple/corenet/blob/5b50eca42bc97f6146b812...

    It will be interesting to see how it tracks over the next year; especially with their recent acquisitions:

    Datakalab https://news.ycombinator.com/item?id=40114350

    DarwinAI https://news.ycombinator.com/item?id=39709835

    1: https://github.com/apple/corenet/blob/main/corenet/engine/de...

  • by ipsum2 on 4/24/2024, 3:47:00 AM

    It's interesting that Apple also actively develops https://github.com/apple/axlearn, which is a library on top of Jax. Seems like half the ML teams at Apple use PyTorch, and the other half uses Jax. Maybe its split between Google Cloud and AWS?

  • by coder543 on 4/24/2024, 2:50:04 AM

    They also mention in the README:

    > CatLIP: CLIP-level Visual Recognition Accuracy with 2.7x Faster Pre-training on Web-scale Image-Text Data

    This is the first I’m hearing of that, and the link seems broken.

  • by mxwsn on 4/24/2024, 2:11:52 AM

    Built on top of pytorch.

  • by leodriesch on 4/24/2024, 3:56:26 AM

    How does this compare to MLX? As far as I understand MLX is equivalent to PyTorch but optimized for Apple Silicon.

    Is this meant for training MLX models in a distributed manner? Or what is its purpose?

  • by miki123211 on 4/24/2024, 2:36:04 AM

    What's the advantage of using this over something like Huggingface Transformers, possibly with the MPS backend?

  • by jn2clark on 4/24/2024, 8:48:00 AM

    I would love an LLM agent that could generate small api examples (reliably) from a repo like this for the various different models and ways to use them.

  • by buildbot on 4/24/2024, 2:21:45 AM

    Does this support training on Apple silicon? It’s not very clear unless I missed something in the README.

  • by big-chungus4 on 4/26/2024, 7:57:32 PM

    I went through their folders, they have have a lot of classes that just inherit from pytorch and torchvision classes and seemingly do nothing new. All optimizers, schedulers and most layers do that. They do however have a reasonable amount of blocks, i.e. specific combinations of layers from various papers, similar to monai.networks.blocks. Out of "building pieces" they also have a few newly implemented losses, metrics.

  • by RivieraKid on 4/24/2024, 3:23:41 PM

    What library would you recommend for neural net training and inference on Apple M1? I want to use it from C++ or maybe Rust. The neural net will have 5M params at most.

  • by benob on 4/24/2024, 8:51:11 AM

    > OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework https://arxiv.org/abs/2404.14619

    Apple is pushing for open information on LLM training? World is changing...

  • by andreygrehov on 4/24/2024, 2:09:30 AM

    What hardware would one need to have for the CoreNet to train efficiently?

  • by orena on 4/24/2024, 2:27:38 PM

    The style is not very different than NeMo(nvidia)/fairseq(Facebook)/espent(oss) etc..

  • by m3kw9 on 4/24/2024, 2:57:12 PM

    Ok, why would anyone use this when you have industry standard methods already?

  • by gnabgib on 4/24/2024, 1:28:52 AM

    h1: CoreNet: A library for training deep neural networks

  • by symlinkk on 4/24/2024, 2:04:14 AM

    Pretty funny that Apple engineers use Homebrew too.

  • by irakeshpurohit on 4/29/2024, 6:34:56 AM

    anyone have this hosted so anyone can try this out?

  • by javcasas on 4/24/2024, 9:53:43 AM

    Looks at Apple: CoreNet Looks at Microsoft: Net Core

    My inner trademark troll demands a bucket of popcorn.