Tweetovi
- Tweetovi, trenutna stranica.
- Tweetovi i odgovori
- Medijski sadržaj
Blokirali ste korisnika/cu @Thom_Wolf
Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @Thom_Wolf
-
Thomas Wolf proslijedio/la je Tweet
#nlphighlights 104:@SanhEstPasMoi and@Thom_Wolf talk to us about model distillation, when you try to approximate a large model's decision boundary with a smaller model. After talking about the general area, we dive into DistilBERT.https://soundcloud.com/nlp-highlights/104-model-distillation-with-victor-sanh-and-thomas-wolf …Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Thomas Wolf proslijedio/la je Tweet
The 2.4.0 release of transformers is **𝐌𝐀𝐒𝐒𝐈𝐕𝐄** thanks to our amazing community of contributors.
https://github.com/huggingface/transformers/releases/tag/v2.4.0 …Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Thomas Wolf proslijedio/la je Tweet
I'm impressed by the work Hugging Face is doing.https://twitter.com/ClementDelangue/status/1223747179480780800 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Thomas Wolf proslijedio/la je Tweet
Turkish-
#NLP: Anyone interested in a Turkish BERT and wants to evaluate it on downstream tasks? I did evaluation only for UD PoS tagging - any help is really appreciated! Would really like to have a proper evaluation before adding it to the@huggingface Transformers hub
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Thomas Wolf proslijedio/la je Tweet
Supervised multimodal bitransformers now available in the awesome HuggingFace Transformers library!https://twitter.com/huggingface/status/1223267748542808064 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Thomas Wolf proslijedio/la je Tweet
Transformers 2.4.0 is out
- Training transformers from scratch is now supported
- New models, including *FlauBERT*, Dutch BERT, *UmBERTo*
- Revamped documentation
- First multi-modal model, MMBT from @facebookai, text & images Bye bye Python 2
https://github.com/huggingface/transformers/releases …Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Thomas Wolf proslijedio/la je Tweet
Learn to build an interactive Transformer attention visualization based on
@huggingface and@d3js_org in under 30 minutes! We developed a minimal teaching example for our@MIT_CSAIL IAP class, publicly available here: http://bit.ly/attnvis@sebgehr@davidbau#NLProc#XAI#Vispic.twitter.com/CLOHbRT0vAPrikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
People asking me to teach classes clearly give zero fuck to the imposter syndrome of a former physics PhD turned lawyer before joining AI Anyway I'll co-teach NLPL Winter School w Yoav Goldberg talking transfer learning, its limits & where the field might head Will share slidespic.twitter.com/M1QFXBT5CP
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Let me highlight this amazing work I've read recently on
#compositionality in NLP, in which you'll find both: - a deep discussion of what it means for a neural model to be compositional - a deep and insightful comparison of LSTM, ConvNet & Transformers!
https://arxiv.org/abs/1908.08351 pic.twitter.com/LX9JQE1Ira
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
JAX's ecosystem is growing fast! https://github.com/google-research/flax/tree/prerelease …pic.twitter.com/bx0zHFk7QF
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
That’s a neat use of Transfer Learning to leverage a pretrained GAN (here BigGAN)https://twitter.com/anh_ng8/status/1217222528386494464 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
=> you need to keep everything clear & visible. No unnecessary user-facing abstractions or layers. Direct access to the core. Each user-facing abstraction is a mask that can hide some ML-bug, a potential source of misunderstandings, and a steeper learning curve for users. [End]
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
D. Open-sourcing ML can be very different from other types of open-sourcing: - ML bugs are silent => researchers need to know exactly what's happening inside your code. - Researchers will create things you have no ideas about => they'll want to dive in your code and modify it.
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
7. Now if you want to build a large-scale tool like
Transformers? Here are a few additional tips
A. focus on one essential feature that your community really needs and no one provides
B. do it well
C. keep putting yourself in the shoes of people using your tool for the 1st timePrikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
5. Spend 4 days to do it well. Open-sourcing a good code base takes some time but you should consider it as important as your paper 6. Consider merging with a larger repo: are you working on language models?
Transformers is probably happy to help you
https://github.com/huggingface/transformers …Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
3. Give clear instructions on how to run the code, at least evaluation, in such a way that, combined with pretrained models, it allows for fast test/debug 4. Use the least amount of dependencies: if you are using an internal framework to build the model => copy the relevant part
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
2. Put yourself in the shoes of a master student who has to start from scratch with your code: - give them a ride up to the end with pre-trained models - focus examples/code on open-access datasets (not everybody can pay for CoNLL-2003)
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
1. Consider sharing your code as a tool to build on more than a snapshot of your work: -other will build stuff that you can't imagine => give them easy access to the core elements -don't over-do it => no need for one-liner abstractions that won't fit other's need – clean & simple
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
I often meet research scientists interested in open-sourcing their code/research and asking for advice. Here is a thread for you. First: why should you open-source models along with your paper? Because science is a virtuous circle of knowledge sharing not a zero-sum competitionpic.twitter.com/x16jgKmLFr
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Thomas Wolf proslijedio/la je Tweet
Great intro to the modern landscape of Deep Learning &
#nlproc by@lexfridman@MIT. Including sweet mentions of the models inside@huggingface transformers, write with transformers &@seb_ruder's NLP progress repo! https://youtu.be/0VH1Lim8gL8?t=1114 … https://youtu.be/0VH1Lim8gL8?t=1282 …pic.twitter.com/pB4iyeCuAE
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.