Adafruit_CircuitPython_CircuitPlayground VS transformers

First, a bit about me. My first experience with programming and electronics was in 2017. I spent a few of weeks trying to learn Python, and didn't get very far with it. While purchasing accessories for my first Raspberry Pi, I grabbed a Circuit Playground Express, having no idea what it was (I thought I could connect it to the Pi like a sensor, but that was not the case). It turns out the CPX is a stand-alone microcontroller. When I finally looked into what to do with it, I struggled for a bit before finding CircuitPython. Within a short period of time, I made an LED blink. Nothing I had done in trying to learn programming up to that point hooked me as much as that moment. It was the difference between manipulating data and manipulating the physical world. Within a month, I made my first library PR, and within 2 months, began working with Adafruit by publishing my first guide. Since then, I have become a Creative Engineer with Adafruit where, among many other things, I am a community leader, CircuitPython library project maintainer, software developer, hardware designer, and technical guide and library author.

Fascinating work, very promising.

Can you summarise how the model in your paper differs from this one ?

https://github.com/huggingface/transformers/issues/27011

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