If your paper is indexed on @huggingface paper pages: https://
huggingface.co/papers you can now link the page directly from github or project pages using badges: https://
huggingface.co/datasets/huggi
ngface/badges#paper-page
…
OPEN SOURCE
-
Hugging Face Paper Pages Now Support Direct GitHub Badges Linking
By
–
-
Informal Group of Hackers and AI Developers Building Cool Projects
By
–
Btw for those that keep asking who this "we" is, it's just me, @thetanmay
, and a group of curious hackers, devs, ML and applied AI folks that want to build cool stuff. This informal group is growing larger — we have no big plans except to keep shipping, learning, and helping! -
Text-Generation-Inference: HuggingFace’s Recommended LM Generation Tool
By
–
You should use text-generation-inference, that’s what we’re using at HuggingFace. Don’t use the transformers library for generation with large LM. https://
github.com/huggingface/te
xt-generation-inference
… -
Text-Generation-Inference: HuggingFace’s LLM Inference Solution
By
–
You should try text-generation-inference, that’s what we’re using at HuggingFace. Don’t use the transformers library for generation with large LM. https://
github.com/huggingface/te
xt-generation-inference
… -
Free Audio Machine Learning Course with Hugging Face Transformers
By
–
Are you ready to upskill your Machine Learning skills with Audio? Tune in at 18:00 CEST today and participate in our 100% free and fully-open source @huggingface Transformers Audio course! https://
youtube.com/live/wqkKResXW
B8?feature=share
… -
GPT Engineer and AI Tools for Junior Developer Scaffolding
By
–
Like SmolAI and @AutoCodePro
, a new repo that basically solves the same use-case: https://
github.com/AntonOsika/gpt
-engineer
… Junior engineer level scaffolding is going to get MUCH easier! -
Behind-the-scenes training details of Flan-T5 model revealed
By
–
Insightful thread by @hwchung27 that shares some behind the scenes of training flan-t5. These details are often overlooked and sometimes mentioned in footnotes in papers (or not mentioned at all, i.e., "not paper worthy"). Sometimes they are stack dependent, model dependent or
-
SlimPajama Reduces AI Training Compute Costs by 50 Percent
By
–
Why we built SlimPajama – it's all about training efficiency. Without de-duplication, a model would have to go through 1.2T tokens before seeing ~600B unique tokens. SlimPajama sees 600B tokens in half the time. That saves you 50% on compute costs! https://
cerebras.net/blog/slimpajam
a-a-627b-token-cleaned-and-deduplicated-version-of-redpajama
… -
OpenAI Functions Release: LangChain Support for Chat Models
By
–
Emergency OpenAI Functions Release `pip install langchain==0.0.199` Support for functions in chat model wrapper Convert @langchain tools to functions Docs: https://
github.com/hwchase17/lang
chain/blob/master/docs/modules/agents/tools/tools_as_openai_functions.ipynb
… Next up… I think its time for a new type of agent -
Open-Sourcing Research Work for CVPR 2023 Conference
By
–
In support of open science and the research community, we're open-sourcing this work — and we'll be presenting it at #CVPR2023 next week. Training code and model checkpoints