Rezultati pretraživanja
  1. 6. sij

    Welcome to the New and Improved TensorFlow Hub! "We added search features and visual cues that enable you to find and download the right model for your use case." by via

  2. We recently announced a lot of new upgrades for TensorFlow Hub ✨ Find out what’s new, and how is helping you get what you need for your ML projects, faster. ⏰ Read more here →

  3. Stylize images with arbitrary styles in . We released a model on , try it out yourself in our new Colab! Details here →

  4. We are proud to announce that the FGVCx image embedding model is now live on TensorFlow Hub using training images of >1,000 species from 🇩🇰🍄

  5. SPIRAL is an RL agent that generates images by interacting with off-the-shelf graphics software. We've released 9 agents trained to paint portraits of imaginary people in just 19 brush strokes on ! For more, see: →

  6. Hello World! ≈ Bonjour le monde! ≈ こんにちは世界! Explore multilingual semantic retrieval → and retrieval QA → using our latest multilingual Universal Sentence Encoder models →

    Multilingual semantic retrieval
  7. Hey, I recognize those two Gravatars... 😀 Very proud of and for being top contributors to Hub over the last few months! 🙌✨

  8. So many tweets in so many languages about the cross language universal sentence encoder... If only there was a way to embed them in the same space with one line of python... 🤷‍♂️

  9. 12. ožu 2019.

    Analyzing the semantic similarity between title tags and keywords w/ Universal Sentence Encoder 👉 cc in a way the more I try to automate the more I find myself doing *traditional SEO* 🙈

  10. BERT is in and here's a runnable notebook! 🎉 More BERT modules at

  11. ✨🛰️ Day #54 in spaaaaaaaaace! Machine Learning Space Applications on SmallSat Platforms with , via the good folks at and the Center for Space, High-Performance, and Resilient Computing (SHREC). 📰Check out the paper:

    Prikaži ovu nit
  12. 🧠 Learn more about how you can *reuse* model components (like LEGO bricks!) for free on : uses transfer learning to bootstrap model creation: which means you can achieve higher accuracy with less data, minimal hardware, and in record time.

    Prikaži ovu nit
  13. 2) Hub If you want to train your model on a small data set, or improve generalization, you'll need to use something called transfer learning. modules make it easy—and are available in an marketplace: . site:

    Prikaži ovu nit
  14. 10. stu 2018.

    What better way to celebrate 's 3rd birthday than giving a talk about its magical Tensorflow Hub to 300+ devs at DevFest Rift Valley. Happy birthday Tensorflow!

  15. ✨🤯 Attempting to get a handle on the full list of offerings; each new discovery makes me more excited. The ecosystem expanded so much in 3 years! Swift for TF Edge TPUs tf.keras eager mode

    Prikaži ovu nit
  16. Image augmentation lets you get the most of your dataset. We released state-of-the-art AutoAugment Modules on allowing you to train better image models with less data! Check it out here →

  17. Perform object detection with the brand new module from , trained on the OpenImages Dataset V4! Join at the Google booth Wednesday 12th at for a demo. Check it out here ↓

  18. Check out 's demonstration of how pre-trained modules from TensorFlow Hub can be leveraged to solve sentiment analysis challenges on Kaggle →

  19. 15. kol 2018.

    With , , , and more, TensorFlow is no longer a library or framework, it is an ecosystem. BTW, this article doesn't mention the recent performance gains in TensorFlow:

  20. 10. svi 2018.

    First module on ! Use I3D for video actions recognition, for details about this model read CVPR paper

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.