[torchtext v0.4]: Supervised learning datasets and baselines. Datasets for AG_NEWS, SogouNews, DBpedia, YelpReviewPolarity, YelpReviewFull, YahooAnswers, AmazonReviewPolarity, AmazonReviewFull.
— PyTorch (@PyTorch) August 8, 2019
Classification tutorials, fast and basic Tokenizerhttps://t.co/Qlpsoe5fVQ
spaCy meets PyTorch-Transformers: Fine-tune BERT, XLNet and GPT-2 · Blog · Explosion https://t.co/JSWsJgf5T0 via @rightrelevance thanks @peteskomoroch
— Bojan Tunguz (@tunguz) August 2, 2019
Selene is a deep learning library for biological sequences such as DNA, RNA or protein sequences and their measured properties
— PyTorch (@PyTorch) August 2, 2019
Kathleen Chen, @evan_cofer, @OlgaTroyanskaya and team have published it to Nature Methodshttps://t.co/mHg5zVJCTi
Get Started: https://t.co/4pWkUgref7 pic.twitter.com/WxoeM0vcIe
roberta = torch.hub.load('pytorch/fairseq', 'roberta.large') https://t.co/Sdc5UADqDY
— PyTorch (@PyTorch) July 30, 2019
new post 📈 Interest in #PyTorch among researchers is growing rapidly https://t.co/yQwZfbU3gx pic.twitter.com/BehFPnFGVm
— Ben Lorica 罗瑞卡 (@bigdata) July 25, 2019
Discovering good libraries and code might be overwhelming.
— PyTorch (@PyTorch) July 18, 2019
- our Ecosystem page has been curating high quality libraries
- Hub has been curating models that reproduce research and are easy to use
Read more about our recent additions and guidelines: https://t.co/1vezaZEF1F pic.twitter.com/oETW02dQ6y
Sequence to Sequence Models in PyTorch https://t.co/fmPefMzr0b #deeplearning #machinelearning #ml #ai #neuralnetworks #datascience #pytorch
— PyTorch Best Practices (@PyTorchPractice) July 16, 2019
Interesting new "fastai-serving" project:
— Jeremy Howard (@jeremyphoward) July 16, 2019
"A Docker image for serving fastai models, mimicking the API of Tensorflow Serving. It is designed for running batch inference at scale."https://t.co/mJgXHbdXqA
"pytorch-pretrained-bert" ➡️ "pytorch-transformers"
— Thomas Wolf (@Thom_Wolf) July 16, 2019
Install with `pip install pytorch-transformers`
Quickstart: https://t.co/rk4GZDSDWq
Release notes: https://t.co/pjmUH7Gbz5
Documentation (work in progress): https://t.co/9HronWmz1b
🔥Pytorch-Transformers 1.0🔥
— Thomas Wolf (@Thom_Wolf) July 16, 2019
Six NLU/NLG architectures: BERT, GPT, GPT-2, Transfo-XL, XLNet, XLM
Total: 27 pretrained models
Still the same
-Superfast onboarding
-SOTA scripts: GLUE, SQuAD, Text generation
New
-Unified API
-Access hidden-states, attentions...
-Torchscript
-... pic.twitter.com/eiRnLEwvFh
For all the attention the Wasserstein distance as a loss function gets, there didn't seem to be an efficient batch stable implementation. Now we have one as a @PyTorch extension. On that occasion, we look at a general technique for reductions on the GPU: https://t.co/2hkFKQ0Qww pic.twitter.com/h6NaRhcpQU
— Thomas Viehmann (@ThomasViehmann) July 8, 2019
Count the FLOPs of your PyTorch model. https://t.co/NBCXkPI2rP #deeplearning #machinelearning #ml #ai #neuralnetworks #datascience #pytorch
— PyTorch Best Practices (@PyTorchPractice) July 1, 2019
