From transformers is all we need to transformers is all we care about π pic.twitter.com/sgHDDI3YFi
β Sebastian Raschka (@rasbt) March 23, 2022
From transformers is all we need to transformers is all we care about π pic.twitter.com/sgHDDI3YFi
β Sebastian Raschka (@rasbt) March 23, 2022
Whoa. 96% of the winning solutions used Python. This is the way.
β Sebastian Raschka (@rasbt) March 15, 2022
Interesting tidbit: all winning NLP solutions used transformers. However, most winning computer vision solutions were still convolutional nets (mostly EffficientNet). https://t.co/VQHbqz84Pi
New blog post!β¬οΈ Deep Neural Nets: 33 years ago and 33 years from now https://t.co/pbZvYh3Mck we reproduce what I think may be the earliest real-world application of a neural net trained end-to-end with backprop (LeCun et al. 1989), try improve it with time travel, and reflect. pic.twitter.com/MKZ7S3GUdv
β Andrej Karpathy (@karpathy) March 14, 2022
An important but elusive quality to learn in a PhD is research style. It is valuable to be aware of this before you start a PhD. Among other updates, I added an extensive discussion on research style to my "choosing a grad school" blog post. Enjoy! https://t.co/HrMkPbGZcv
β Tim Dettmers (@Tim_Dettmers) March 14, 2022
The latest contribution to the Sharpened Cosine Similarity project is a PyTorch Lightning demo from @PSodmann. It shows off the streamlined coding patterns @PyTorchLightnin is famous for.https://t.co/KZUuaOoAF1
β Brandon Rohrer (@_brohrer_) March 12, 2022
I am going through the RecSys 2020 tutorial by @NVIDIAAI and I found this interesting bit on Decision Trees π‘
β Radek Osmulski πΊπ¦ (@radekosmulski) February 26, 2022
Not that I don't trust @kaggle Grandmasters but wow, I never thought about this like that π pic.twitter.com/TWXzlVAlnG
On @huggingface we host more and more git repos so we are switching the Infra behind our git server to Gitaly, an open source Infra project from @gitlab
β Julien Chaumond (@julien_c) February 22, 2022
The goal is to make the Hub scalable and robust for the next 5 years of MLπ₯
more info in this thread https://t.co/rwwy5ZYYfR pic.twitter.com/ScbhkibefD
another motivational AI tweet drop π£
β Kyunghyun Cho (@kchonyc) February 15, 2022
prompt engineering is a symptom not a cure and must be treated not encouraged.
Wow this thread is amazing.
β Jeremy Howard (@jeremyphoward) February 12, 2022
Take a look, ML twitterers, and tell me whether this is the worst use of gradient boosing decision trees you've ever seen!
(And if it's not - please share in the comments any even more terrible examples you can share...) https://t.co/FEwMQZQaar
GPT-NeoX-20B, 20 billion parameter large language model made freely available to public, with candid report on strengths, limits, ecological costs, etc.
β Gary Marcus (@GaryMarcus) February 10, 2022
Genuinely Open AI https://t.co/kwcCtNQiGD
Some people requested the blog post version so here it is. Hopefully this is easier to read πhttps://t.co/pFe4HxNiFf
β Chip Huyen (@chipro) February 8, 2022
I wrote a lot of hacky scripts in my lifetime for logging, hyperparameter sweeps, and submitting my jobs to our SLURM cluster. I may be biased, but using @gridai_ now is very refreshing. Just select your Gh repo, the hardware you want to use & the hparam ranges, and off you go https://t.co/qdfYEvWGqk pic.twitter.com/u4TWrAfl7T
β Sebastian Raschka (@rasbt) February 4, 2022