Simple Open-Vocabulary Object Detection with Vision Transformers
abs: https://t.co/ytb2Tvliu1 pic.twitter.com/0xUjokjLcB

— AK (@ak92501) May 13, 2022

"Whether to go with a decoder-only or encoder-decoder transformer?"
It turned out that this question on the architecture of the model is not actually that important!
You just need the right objective function and a simple prompting to switch mode during pretraining/finetuning. pic.twitter.com/lanaYmHynW

— Mostafa Dehghani (@m__dehghani) May 12, 2022

So excited to share what I've been working on this past month.
Not enough RAM to load the model? No problem.
Not enough GPU RAM to host the model? No problem.
As long as it fits on your hard drive, we'll make it run. https://t.co/R4YojOqodl

— Sylvain Gugger (@GuggerSylvain) May 12, 2022

Scikit-learn 1.1 is out! https://t.co/lU8ZIa3g1n. Looking at the release highlights, one really nice things is the support for infrequent categories in the OneHot Encoder. Super useful! pic.twitter.com/dDCoNhLXLZ

— Sebastian Raschka (@rasbt) May 12, 2022

“Are there any software engineers that switched into a machine learning role and found it a lot more stressful due to deadlines combined with the uncertainty of research?”

Discussion: https://t.co/OHZdmj1ly2 pic.twitter.com/jV3v7QK8KB

— hardmaru (@hardmaru) May 11, 2022

Our 50+-year review of #forecast combinations is now out. @YanfeiKang @f3ngli @Xia0qianWang https://t.co/MTBhI2trNe

— Rob J Hyndman (@robjhyndman) May 10, 2022

I finally cleaned up the docs for https://t.co/kGGXu3OFx5
This basically merges @PyTorchLightnin @raydistributed and https://t.co/BQxWHUX4zm to boilerplate #nlproc experiments very quickly, with @weights_biases, custom S3 logs/plots, slurm sweeps etc. RL via Moolib coming soon!

— Victor Zhong (@hllo_wrld) May 10, 2022

.@Gradio Demo for CaptchaCracker, an open source Python library that provides functions to create and apply deep learning models for Captcha Image recognition on @huggingface Spaces
demo: https://t.co/jXbFUsSgqx
github: https://t.co/tEFCEB43uM pic.twitter.com/MsUisP4wBd

— AK (@ak92501) May 9, 2022

Hey friends 👋

So I came across this Python feature that is both beautiful and scary 😄

The __rshift__ and __rrshift__ dunder methods... combined!

Interesting, isn't it? 🙂 Here is how it works pic.twitter.com/XrIqaP4zze

— Radek Osmulski 🇺🇦 (@radekosmulski) May 9, 2022

The SQL translation of the dev version of dbplyr is impressive. Did you know now it can translate cut() to CASE WHEN, and there's an option to use WITH clause to avoid the hell of subqueries? #rstatshttps://t.co/6P3UjlRROP pic.twitter.com/r0Xk3YPihQ

— Hiroaki Yutani (@yutannihilat_en) May 6, 2022

The beginner feels like everything is difficult. The expert knows that nothing is easy.

— François Chollet (@fchollet) May 6, 2022

Presenting: The AI Economist
This is one of the most impactful lines of AI research I've ever worked on. Its implications span from immediately impactful to highly philosophical.
Blog: https://t.co/vmkYroUpUY
Paper: https://t.co/xjZXMAyQMn
A 🧵 with high level take-aways: pic.twitter.com/xjBl37gzUt

— Richard Socher (@RichardSocher) May 5, 2022