
Interactive deep learning book with multi-framework code, math, and discussions. @astonzhangAZ @zacharylipton @mli65 @smolix
Joined February 2020
- Tweets28
- Following13
- Followers7.8K
- Likes25
Dive into Deep learning (D2L.ai) retweeted
Thrilled that our 'Dive into Deep Learning' book, now published by Cambridge University Press, is the Top New Release on Amazon!
To ensure accessibility and affordability, we, the authors, have waived our royalties. Plus, it's always available for free at D2L.ai
Welcome aboard!
🎉Thank all developers of PaddlePaddle for their efforts, and we really appreciate @mli65 @astonzhangAZ and every developer of @D2L_ai. #PaddlePaddle is an open source deep learning platform originated from industrial practic. Looking forward to more developers joining us!
14
Chinese version 2.0🎉 zh.d2l.ai
🙏37k stars, 248 contributors, multi-frameworks by & for our growing community:
🆙Upgraded physical book featuring @PyTorch implementation @AmazonScience
🆕New @PaddlePaddle_ implementation by @Baidu_Inc, reviewer @DavenCheung
Dive into Deep learning (D2L.ai) retweeted
Announcing the Preview Edition of Dive into Deep Learning @D2L_ai containing our nightly builds. Check them out at preview.d2l.ai/d2l-en/master…. Thanks @mli65 @astonzhangAZ @zacharylipton @gollum_here @goldpiggy95 and many volunteers for all the work!
The deep learning field has been growing fast. Our D2L book just added topics on scalability of transformers and very recent large-scale pretraining methods:
d2l.ai/chapter_attention-mec…
More updates are on the way 😃
Replying to @MoeinShariatnia
v1.0-alpha🎉
📚New topics on transformers for vision and large-scale pretraining, CNN design, generalization, etc
⚡️New API inspired by @PyTorchLightnin to streamline the presentation
🦥"Lazy" layers to simplify @PyTorch code
🈶zh/pt/tr/vi/ko/ja ongoing
D2L.ai
Thank Jared Kaplan, Ilija Radosavovic, @jiahuiyu, @Chitwan_Saharia, and Tom Brown for kind permission of using their figures (scaling law, RegNet, #parti, #Imagen, #gpt3) in our v1.0-alpha! Release notes: github.com/d2l-ai/d2l-en/rel…
Dive into Deep learning (D2L.ai) retweeted
While ConvNeXt (2022) further sparks the debate between CNNs and transformers for vision, D2L.ai added a CNN design section covering ResNeXt-itication @sainingxie et al, RegNets Radosavovic et al, ConvNeXt @liuzhuang1234 et al. Preview at: preview.d2l.ai/d2l-en/master…
Dive into Deep learning (D2L.ai) retweeted
The #Portuguese translation for #D2L is up at pt.d2l.ai. Muito obrigado a @dalcimar, Lucas Hendges e Fabio Kurpel por seu excelente trabajo! Join us on github.com/d2l-ai/d2l-pt.
Introducing a new option to run the D2L book for free: using SageMaker Studio Lab d2l.ai/
Dive into Deep learning (D2L.ai) retweeted
It's not the lack of resources that holds back people from getting started with deep learning.
It's a combination of procrastination and lack of guidance.
This is why my top recommendation is "Dive into Deep Learning".
It gets you going right away!
d2l.ai/
Dive into Deep learning (D2L.ai) retweeted
Looks like the excellent #DeepLearning book by @astonzhangAZ @zacharylipton @mli65 @smolix has been published to arXiv! arxiv.org/abs/2106.11342 While the book is thought for online reading at d2l.ai/, I still think that having a printable version is Big News!
The Chinese translation of D2L (Chap 1--8) is available, with implementations in @ApacheMXNet, @PyTorch and @TensorFlow! 《动手学深度学习》第二版预览版:zh-v2.d2l.ai/ GitHub地址:github.com/d2l-ai/d2l-zh
Dive into Deep learning (D2L.ai) retweeted
The d2l.ai chapter on Attention just got a lot more attention. d2l.ai/chapter_attention-mec… now including Bahdanau attention, Watson Nadaraya, Self-attention and much more background. Just in time for school! Thanks @astonzhangAZ and everyone on the @D2L_ai team.
Check out our brand-new attention chapter, such as simpler illustrations using Nadaraya-Watson kernel regression (1964) before Bahdanau attention. Now PyTorch adaptation is complete for Vol.1. The first 1/3 part is also translated into Chinese and Turkish. D2L.ai
Dive into Deep Learning v0.15.0 is released! More chapters are in PyTorch and TensorFlow. More chapters are close to v1.0. Adopted at 140 universities: if you use D2L to teach, apply for free computing resources by Nov 22. Check out D2L.ai
Dive into Deep Learning now supports @TensorFlow. The first 7 chapters are released today with more on their way. Special thanks to @TerryTangYuan and @NeurlAP for contributing code and our Google friends @JeffDean @DynamicWebPaige and @tomerk11. More at D2L.ai.
Dive into Deep Learning now supports @PyTorch. The first 8 chapters are ready with more on their way. Thanks to DSG IIT Roorkee @dsg_iitr, particularly @gollum_here who adapted the code into PyTorch. More at D2L.ai. @mli65 @smolix @zacharylipton @astonzhangAZ

















































