Lightning made the mistake of relying on inheritance instead of callbacks, not realizing that inheritance has fundamental well-known limitations for extensibility.
— Jeremy Howard (@jeremyphoward) February 5, 2021
They even boasted about this.
Later, they were forced to add callbacks. Oops. pic.twitter.com/w7g7XnRl6t
Who made this pic.twitter.com/212bG8vZf0
— Maarten van Smeden (@MaartenvSmeden) February 4, 2021
Check out new work on ML-driven design and exploration of custom accelerators, showing how #MachineLearning facilitates architecture exploration by rapidly identifying high-performing configurations across a range of applications. Learn more ↓ https://t.co/ZQaKSLnNxp
— Google AI (@GoogleAI) February 4, 2021
Just finished a complete tutorial on optimizing @huggingface models with @weights_biases
— Boris Dayma (@borisdayma) February 4, 2021
No extra line of code required 🥳https://t.co/aMyQ6EQyJS
Announcing Observable Inputs! 🎚🎛 Lightweight UI components to help you explore data and build interactive displays. https://t.co/odojmeV2tq
— Mike Bostock (@mbostock) February 4, 2021
So I wrote a 5400-word lecture note on the basics of data engineering for my students, covering:
— Chip Huyen (@chipro) February 4, 2021
* data formats (row- vs. column-based, text vs. binary)
* ETL
* batch processing vs. stream processing
* training datasets
WIP. Feedback much appreciated!https://t.co/oFdTDSIAin pic.twitter.com/CShOlt00hk
I've been asked to re-share this wonderful interactive 3D map shows land values in Japan from 1989 to 2019. Well worth checking this one out. Source: https://t.co/wQ9VHEeaVP pic.twitter.com/H7OcIkQrxm
— Simon Kuestenmacher (@simongerman600) February 4, 2021
Running ML models in less than 100KB on embedded devices. By Pete Warden. https://t.co/cOXXu6R2kG pic.twitter.com/0yf2Ltmfcd
— Reza Zadeh (@Reza_Zadeh) February 4, 2021
I think it's fair to say humanity took a bit of a beating in 2020, but one month into 2021 there are signs of a fightback.
— John Burn-Murdoch (@jburnmurdoch) February 3, 2021
It took a year to reach 100 million confirmed cases worldwide. It's taken six weeks to give out 100 million doses of the vaccine.https://t.co/usvjnUBgsa pic.twitter.com/rL5BaKIan4
Introducing the World Models Library, an open-source, platform-agnostic suite of tasks and tools for examination of world model design and performance in visual model-based reinforcement learning. Learn more and grab the code at https://t.co/l9jKoBWS2V pic.twitter.com/sT6W4m2FDe
— Google AI (@GoogleAI) February 3, 2021
Very cool new @fastdotai callback from @ignaciooguiza: `PredictionDynamics()`, which shows how predictions develop during training. Inspired by @karpathy.https://t.co/csMtCLKlbi pic.twitter.com/CeNfBIxnXC
— Jeremy Howard (@jeremyphoward) February 3, 2021
This is a really good example of really bad survey design. The two prior questions to "should Cruz resign" are clearly priming respondents with the information the survey writers want to convey in a biasing manner. It should not be taken at face value. pic.twitter.com/hxG9XeeK3o
— Natalie Jackson (@nataliemj10) February 3, 2021
