shumai VS transformers

https://github.com/tensorflow/tfjs/tree/master/tfjs-backend-...

([...], tflite-support, tflite-micro)

From facebookresearch/shumai (a JS tensor library) https://github.com/facebookresearch/shumai/issues/122 :

> It doesn't make sense to support anything besides WebGPU at this point. WASM + SIMD is around 15-20x slower on my machine[1]. Although WebGL is more widely supported today, it doesn't have the compute features needed for efficient modern ML (transformers etc) and will likely be a deprecated backend for other frameworks when WebGPU comes online.

tensorflow rust has a struct.Tensor:

JS for ML actually does exist https://github.com/facebookresearch/shumai

> It's really a shame that data science, ML, and notebooks are so wrapped up in it. Otherwise we could jettison the whole thing into space

Although I personally feel Python has its place, I contribute to a project that hopes to diversify the ML/scientific computing space with a TypeScript tensor lib called Shumai: https://github.com/facebookresearch/shumai

Doesn’t really matter for large batch/large model training on GPUs that don’t need much coordination.

But Python speed is one of the main motivations for a JS/TS based ML lib I’m working on: https://github.com/facebookresearch/shumai

As a hedge against CPython never becoming fast, we're creating a project called Shumai that attempts to deeply integrate with a new JavaScript runtime (Bun[3]).

You can expose objects. Here's how it is done in Bun: https://github.com/facebookresearch/shumai/blob/main/shumai/...

We've been using this feature heavily in Shumai.

I think you are vastly overestimating the complexity associated with this (user exposed ref-counting/garbage collection) and may not be totally up to date on what's implemented.

This thought is pretty much the exact motivation behind a recent effort I’m helping out with https://github.com/facebookresearch/shumai

transformers is a package by Hugging Face, that helps you interact with models on HF Hub (GitHub)

Is t5x an encoder/decoder architecture?

Some more general options.

The Flax ecosystem

https://github.com/google/flax?tab=readme-ov-file

or dm-haiku

https://github.com/google-deepmind/dm-haiku

were some of the best developed communities in the Jax AI field

Perhaps the “trax” repo? https://github.com/google/trax

Some HF examples https://github.com/huggingface/transformers/tree/main/exampl...

Sadly it seems much of the work is proprietary these days, but one example could be Grok-1, if you customize the details. https://github.com/xai-org/grok-1/blob/main/run.py

The HuggingFace transformers library already has support for a similar method called prompt lookup decoding that uses the existing context to generate an ngram model: https://github.com/huggingface/transformers/issues/27722

I don't think it would be that hard to switch it out for a pretrained ngram model.

Most of this tutorial is based on Hugging Face course about Transformers and on Niels Rogge's Transformers tutorials: make sure to check their work and give them a star on GitHub, if you please ❤️

* Superconvergence + LR range finder + Fast AI's Ranger21 optimizer was the goto optimizer for CNNs, and worked fabulously well, but on transformers, the learning rate range finder sadi 1e-3 was the best, whilst 1e-5 was better. However, the 1 cycle learning rate stuck. https://github.com/huggingface/transformers/issues/16013

Thanks! :) I'm pushing them into transformers, pytorch-gemma and collabing with the Gemma team to resolve all the issues :)

The RoPE fix should already be in transformers 4.38.2: https://github.com/huggingface/transformers/pull/29285

My main PR for transformers which fixes most of the issues (some still left): https://github.com/huggingface/transformers/pull/29402

If you want to tinker with the architecture Hugging Face has a FOSS implementation in transformers: https://github.com/huggingface/transformers/blob/main/src/tr...

If you want to reproduce the training pipeline, you couldn't do that even if you wanted to because you don't have access to thousands of A100s.