We now have a sentencepiece tokenizer in fastai, so you can train NLP models in agglutinative languages like Turkish and non space delimited languages like Chinese.
— Jeremy Howard (@jeremyphoward) June 11, 2019
I just tried it on Turkish sentiment analysis with ULMFiT and easily beat the SoTA! :)https://t.co/6agBdk2tjW pic.twitter.com/PURrIsn7dD
TF.Text is making language models easier to build in #TensorFlow by providing tokenizers and other text-related ops.
— TensorFlow (@TensorFlow) June 10, 2019
Check out the Github repo → https://t.co/26kaIPQBzQ
Read the article to learn more ↓ https://t.co/DuJU5lF9II
Get NVIDIA text-to-speech models from new @PyTorch Hub, a repository of pre-trained models designed for reproducible research. #ICML2019 https://t.co/i6MhV0PaAi
— NVIDIA AI Developer (@NVIDIAAIDev) June 10, 2019
Announcing PyTorch Hub!
— Yann LeCun (@ylecun) June 10, 2019
Want ResNet, ResNext, BERT, GPT, PGAN, Tacotron, DenseNet, MobileNet...?
- Pull models with 1 line of code.
- Publish your models by sending a pull request.
More info here:... https://t.co/Oegem7yvYL
The Web UI and the container specs are both open source in my gpt-2-cloud-run repo, and were explicitly designed to be easily hackable. https://t.co/jVU54rLjvT
— Max Woolf (@minimaxir) June 10, 2019
I've released my web UI for GPT-2-117M that allows you to generate text from the original model, backed by Google Cloud Run for massive scalability at mostly no cost!https://t.co/LO4hHuQo2l pic.twitter.com/HNhg25x8bz
— Max Woolf (@minimaxir) June 10, 2019
tensorwatch: Debugging, Monitoring and Visualization for Deep Learning and Reinforcement Learning https://t.co/TgRBVhbX0J
— PyCoder’s Weekly (@pycoders) June 10, 2019
Attention #rstats users:
— Claus Wilke (@ClausWilke) June 9, 2019
In a few weeks, I'm going to release #cowplot 1.0, and there are going to be some important changes from the current release. I encourage you to check out the development version now and verify things work for you. Thread. pic.twitter.com/33pAELKTnd
Nice new work from FAIR on assessing bias in datasets.
— Yann LeCun (@ylecun) June 7, 2019
Turns out image recognition systems will recognize everyday objects more reliably if the picture was shot in developed countries than... https://t.co/Nq2ErHND15
I just released cloudpickle 1.2.0. If your stack depends on cloudpickle please test it with Python 3.8b1 to try the new optimized C-based implementation by @PierreGlaser (with contributions both to CPython and cloudpickle).
— Olivier Grisel (@ogrisel) June 7, 2019
lineflow: Simple NLP Data Loader for All Deep Learning Frameworks https://t.co/P7ZWiRz9Ki
— PyCoder’s Weekly (@pycoders) June 7, 2019
Become a Bayesian master with bayestestR (0.2) {https://t.co/PBKw8IU1nl} #rstats #DataScience
— R-bloggers (@Rbloggers) June 6, 2019
