TensorFlowCuenta verificada

‍ Movement + AI + Friends. Our #MadeWithTFJS MoveNet model now supports multi-person pose detection and tracking, all at realtime speeds in the browser! What will you make with it? Try now → http://goo.gle/3mayz8q pic.twitter.com/P8Qo99aPAU

TensorFlow models can run faster on GPUs by using the XLA compiler. In this talk, we'll provide an overview of how to use tf.function(jit_compile=True), the XLA: Just-In-Time compiler, using XLA from TF, and benchmark results. Learn more → https://goo.gle/3CVdlRW pic.twitter.com/KDFJ7AkIgn

Due to several security vulnerabilities, we’ve released patches to TF 2.3, 2.4 and 2.5. These fixes are also included in the TF 2.6.0 release. Release notes → http://goo.gle/2XqVVMN  Security advisories → http://goo.gle/3yPiW9V  More advisories → http://goo.gle/2Xsu12Y pic.twitter.com/kHkaIvBDyF

MoveNet is now on mobile! The state-of-the-art pose estimation model finally comes to mobile via TensorFlow Lite with CPU acceleration enabled. Now you can run pose estimation in realtime on modern smartphones. Learn more https://goo.gle/3m494pn pic.twitter.com/uZz6QqXIio

How does the largest retailer in Switzerland train and serve millions of newsletters per week? They use TensorFlow Extended (TFX)! Learn how in the blog → https://goo.gle/3lS5ezu pic.twitter.com/zHbLWwzjeC

Finding the right sample size for model training is easy as , , ! With Keras, you can estimate the optimal number of images needed to achieve desired model performance. Learn how → http://goo.gle/3yRims4 pic.twitter.com/sE6pqUVU60

ScaNN is an open source tool for efficient vector similarity search at scale. In the latest "Building recommendation systems with TensorFlow," we show how to use ScaNN for the retrieval stage of recommenders. Learn more → https://goo.gle/3s4gGJv pic.twitter.com/MUu5YPMZUA

The new EfficientNet V2 models are available on #TFHub! They’re a family of image classification models that achieves greater parameter efficiency and faster training speed. Take a look → https://goo.gle/3fAfmJ3 pic.twitter.com/ojkX5BrSEF

When building AI/ML products, UX considerations are just as important as technical feasibility itself. @dqpdang introduces the People + AI Guidebook and unpacks best practices for ML product decisions. Watch now https://goo.gle/37paFh0 pic.twitter.com/bjFmqdKOBo

If you're a developer using TensorFlow Decision Forests, join our new forum! Learn more and check out the great discussions on TF Decisions Forests here → https://goo.gle/3fyLXyW pic.twitter.com/8sgD2DeNUM

Use Deep & Cross Network to learn cross features effectively in your recommendation system Tap in to the latest episode of “Building recommendation systems with TensorFlow” to learn more → https://goo.gle/3A86RwU pic.twitter.com/IPCYHvgLcl

In the latest ML Tech Talk, we discuss the basics of federated learning and applications for the mobile keyboard. Tune in → https://goo.gle/3794LAo pic.twitter.com/4Zy3etJTSp

Leverage context features and multitask learning. Improve the performance of your recommendation systems by leveraging context features multitask learning. See how → https://goo.gle/3rD6c3v pic.twitter.com/xqkNBlESSa

Transformer models in NLP Research Engineer @IuliaTurc walks through the recent history of Natural Language Processing, plus: Encoding text Language Modeling & Transformers Transfer Learning & BERT Learn more → https://goo.gle/3x1vwkC pic.twitter.com/DQqEMTm5Gt