Nature scientific report: a deep learning model for analysis of ophthalmic images, deployed to a mobile app to enable field diagnosis. Built with Keras and TensorFlow. https://t.co/T4UCDE1TZ7
— François Chollet (@fchollet) June 30, 2022
Nature scientific report: a deep learning model for analysis of ophthalmic images, deployed to a mobile app to enable field diagnosis. Built with Keras and TensorFlow. https://t.co/T4UCDE1TZ7
— François Chollet (@fchollet) June 30, 2022
Wanted to highlight an optimizer, by Xi-Lin Li, which I believe is the most promising way to get second-order methods into ML, but I think didn't get much attention because it came from a lone signal-processing researcher rather than an ML lab. https://t.co/1ASVzhZ2DK
— Yaroslav Bulatov (@yaroslavvb) June 29, 2022
(1/5)
The {ggidst} #rstats 📦 features new ggplot geoms "designed especially for visualizing distributions and uncertainty." By @mjskay #ggplot2https://t.co/QGpAonvKBf pic.twitter.com/BknVqLkZas
— Sharon Machlis (on staycation & offline for a bit) (@sharon000) June 29, 2022
Easy correlation analysis with the {correlation} #rstats 📦 by @easystats4u. Plus simple correlation #dataviz with their {see} pkg!https://t.co/oKEdpigkfEhttps://t.co/HsGHFIQ7Lv pic.twitter.com/xC63lOsmNQ
— Sharon Machlis (on staycation & offline for a bit) (@sharon000) June 29, 2022
Announcing PyTorch 1.12: TorchArrow, functorch, accelerated PyTorch training on Mac and more. Take a look: 👇https://t.co/aPh68avBLt pic.twitter.com/q2CJNhj1X3
— PyTorch (@PyTorch) June 28, 2022
New guide on building efficient image input pipelines: https://t.co/GCr969eyyp
— Boris Dayma 🖍️ (@borisdayma) June 27, 2022
Would love any insights 🤓
Map shows travel time from Ancient Rome. Traveling by ship was faster and heaps cheaper than travelling by land. The Mediterranean Sea is relatively small and countries all along could be reached quickly for trade and war. Source: https://t.co/v0eNx2LUfp pic.twitter.com/8LqgJHfkWE
— Simon Kuestenmacher (@simongerman600) June 27, 2022
Very helpful short post I just came across about the cross-entropy loss function:
— Jeremy Howard (@jeremyphoward) June 26, 2022
» Things that confused me about cross-entropy · Chris Said https://t.co/FoP3J1K90O
Implementation of Parti, Google's pure attention-based text-to-image neural network, in Pytorch https://t.co/fAhIdq5p6v #deeplearning #machinelearning #ml #ai #neuralnetworks #datascience #pytorch
— PyTorch Best Practices (@PyTorchPractice) June 25, 2022
A quick thread on "How DALL-E 2, Imagen and Parti Architectures Differ" with breakdown into comparable modules, annotated with size 🧵#dalle2 #imagen #parti
— Rosanne Liu (@savvyRL) June 25, 2022
* figures taken from corresponding papers with slight modification
* parts used for training only are greyed out pic.twitter.com/9zsIUq3toU
MaskViT: Masked Visual Pre-Training for Video Prediction
— AK (@_akhaliq) June 24, 2022
abs: https://t.co/uhMEB6ashb
project page: https://t.co/yK4d3nx9Xj pic.twitter.com/efj7G3cjVq
Huge contribution to the research community by releasing the 20B parameter checkpoint by @m__dehghani and @YiTayML from @GoogleAI ❤️
— Hugging Face (@huggingface) June 23, 2022
Also a big thank you to @DanielHesslow for contributing the model to Transformers🤗 https://t.co/KC7oi4eHSP