Tweetovi
- Tweetovi, trenutna stranica.
- Tweetovi i odgovori
- Medijski sadržaj
Blokirali ste korisnika/cu @misterkardas
Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @misterkardas
-
Prikvačeni tweet
ULMFiT + sentencepiece = state-of-the-art perplexity for polish language. Our (me and
@PiotrCzapla) model (ppl. 118) easily won in PolEval'18 competition (second best ppl. 147). Thank you@jeremyphoward,@seb_ruder and@fastdotai for your invaluable contribution to DL & NLP.Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Interesting work! I believe that sooner than later someone will use neural networks to translate textbooks and papers into theorem prover language like lean or coq.https://twitter.com/GuillaumeLample/status/1202178956063064064 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
Want to export your conda environment/requirements but don't list all dependencies but only those you directly depend on / have installed explicitly, use `--from-history`pic.twitter.com/vop60Nv6Y3
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Great statistics of deep learning frameworks over time based on inspecting repos (by
@paperswithcode). Interesting to see how research trends compare to search query trends. https://twitter.com/paperswithcode/status/1189900999277850625 …pic.twitter.com/8uUnnZX66i
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
I've always (in deep learning timescale) thought that SentencePiece unigram model has an advantage over BPE by not keeping intermediate tokens and thus better utilizing vocabulary space. However, this redundancy of BPE may be crucial when subword regularization is used.https://twitter.com/lena_voita/status/1189546512491134977 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
CVXPY - Differentiable Convex Optimization Layers for pytorch and tensorflow. Read blog post by
@akshaykagrawal with a how-to and an interesting perspective on standard activation functions as convex optimization problems. https://twitter.com/akshaykagrawal/status/1188875869663576064 …pic.twitter.com/p06vyUhpNC
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
Most of the world’s text is not in English. We are releasing MultiFiT to train and fine-tune language models efficiently in any language. Post: http://nlp.fast.ai/classification/2019/09/10/multifit.html … Paper: https://arxiv.org/abs/1909.04761 With
@eisenjulian@PiotrCzapla Marcin Kadras@GuggerSylvain@jeremyphowardpic.twitter.com/QtcWhKqxyL
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
And since 20 min, you can see how much faster the DistilGPT-2 is compared to the original GPT-2 small, thanks to http://sotabench.com an new service by @paperwithcode https://sotabench.com/benchmarks/language-modelling-on-wikitext-103 …pic.twitter.com/XrTU0yJsey
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Speed vs BLEU score trade-off for state-of-the-art English-German translation models. See more on just launched http://sotabench.com pic.twitter.com/RNi7LrvwCz
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
Introducing sotabench : a new service with the mission of benchmarking every open source ML model. We run GitHub repos on free GPU servers to capture their results: compare to papers, other models and see speed/accuracy trade-offs. Check it out:http://sotabench.com Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
Join us next Thursday at the
@PyTorch developer conference for an exciting update on Papers With Code and where we are headed next... https://pytorch.fbreg.com/schedule pic.twitter.com/chijONJUk3
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
An Animated History of ImageNet : from AlexNet to FixResNeXt-101. See the full table and add more results here: https://www.paperswithcode.com/sota/image-classification-on-imagenet …pic.twitter.com/zmyXWrXyAJ
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
If there's one thing that's impossible for Jeremy it's... :) Really appreciate. Now, stop scrolling and start contributing. http://forums.fast.ai is a great place to start.https://twitter.com/jeremyphoward/status/1155954320451334144 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
Resources from my talk on Applied NLP Lessons at
#spaCyIRL. Thanks to everyone who attended!https://pmbaumgartner.github.io/blog/applied-nlp-lessons/ …Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
After the top down approach from part 1, part 2 is a comprehensive bottoms-up approach that shows every single part needed to train a SOTA model on ImageNet.https://twitter.com/jeremyphoward/status/1144643670009774080 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
Thank you everyone for your patience - our new deep learning MOOC is here! Includes 5 lessons diving into the foundations with
@PyTorch, and 2 lessons (co-taught with@clattner_llvm) building a new deep learning library in#swifthttps://www.fast.ai/2019/06/28/course-p2v3/ …Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Happy to see
@fastdotai featured in Technology Radar (https://www.thoughtworks.com/radar/languages-and-frameworks …) by@thoughtworks.pic.twitter.com/A6eCwgNhAm
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
Amazing work by the
@PyTorch team. Easier than ever before to discover, build upon and extend deep learning models!https://twitter.com/PyTorch/status/1138134785640075265 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Marcin Kardas proslijedio/la je Tweet
My &
@misterkardas presentation@fastdotai ’s ULMFiT at AI NLP day was received well. I had a lot of fun presenting. I hope we see increased use of the library and some contributions back, especially implementation of Unsupervised Data Augmentation in ULMFiT.pic.twitter.com/y5Yq2qkaIB
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
TIL about
@sveltejs component framework, the top trending js project of this month, by inspecting what was used to implement Activation Atlas (https://distill.pub/2019/activation-atlas/ …). The great thing is that@distillpub posts are open source (https://github.com/distillpub/post--activation-atlas …).Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Together with
@PiotrCzapla we meet weekly with students from local CS club to teach them DL and@fastdotai. I quickly realized that I'm much better at coding than drawing :) so for a lesson about ResNet internals I used ipyvolume to visualize a few tensors. Hope you like it.pic.twitter.com/sqgdyJpLqj
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.