Laurent Mazare

@lmazare

Machine Learning and Functional Programming,t Ex

Vrijeme pridruživanja: srpanj 2009.

Tweetovi

Blokirali ste korisnika/cu @lmazare

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @lmazare

  1. 25. sij

    Our and bindings for are now compatible with PyTorch 1.4, updated opam package/crate are available! A bunch of examples can be found on GitHub, RNNs, GAN, Reinforcement Learning...

    Poništi
  2. proslijedio/la je Tweet

    Finally made the switch from Tensorflow to PyTorch. The tch-rs bindings for libtorch are really nice. Learning the libtorch API and porting the BERT model from Hugging Face Transformers to Rust was really quick:

    Poništi
  3. 19. pro 2019.
    Prikaži ovu nit
    Poništi
  4. 19. pro 2019.

    Enjoying powerful type system and (secretly) finding / very convenient for fast iteration? Get the best of both worlds!

    Prikaži ovu nit
    Poništi
  5. 8. pro 2019.

    The nice thing about having an 8 hours flight for is getting some uninterrupted hacking time! So I used this to write a small clone of the progress bar library for .

    Poništi
  6. proslijedio/la je Tweet
    20. stu 2019.

    Hey, the next version of Jane Street's OCaml packages just landed! Lots of fun stuff, but ppx_python and pythonlib are pretty neat, as is bonsai, and the new JavaScript Zarith stubs.

    Prikaži ovu nit
    Poništi
  7. 20. kol 2019.

    Ever wanted to learn robotics and enjoy functional programming at the same time ? We gave a try at using on some Raspberry Pi .

    Poništi
  8. proslijedio/la je Tweet
    19. srp 2019.

    The Torch neural net framework is now available in Rust thanks to the tch-rs crate. Here is a first step tutorial.

    Poništi
  9. proslijedio/la je Tweet
    5. srp 2019.

    Interested in learning how to use PyTorch and TensorFlow from OCaml? Then catch 's talk on Monday, July 22nd, at Jane Street's NY office.

    Poništi
  10. proslijedio/la je Tweet
    1. srp 2019.

    Want to play with PyTorch, but prefer OCaml to Python? Here are OCaml bindings to the PyTorch C++ API. With examples for image recognition, GANs, RL, etc. By Laurent Mazare.

    Poništi
  11. proslijedio/la je Tweet
    6. ožu 2019.

    Rust Bindings for PyTorch, brought to you by (who brought to you OCaml-PyTorch previously):

    Poništi
  12. 7. velj 2019.

    A blog post on implementing some Reinforcement Learning method (DQN) using and and getting it to play Atari games.

    Poništi
  13. proslijedio/la je Tweet
    19. sij 2019.

    An OCaml front-end for the PyTorch C++ API. Frank Dellaert, this made me think of you.

    Poništi
  14. 21. pro 2018.

    The PyTorch/ocaml bindings should be usable now! In the transfer learning tutorial we show how to finetune a ResNet model pretrained on imagenet to work on a much smaller dataset.

    Poništi
  15. 8. pro 2018.

    PyTorch 1.0 is out! Time to craft some stable version for my ocaml bindings.

    Poništi
  16. 30. ruj 2018.

    A small write up on playing with (conditional) Generative Adverserial Networks in using .

    Poništi
  17. 21. ruj 2018.

    Using TensorFlow and OCaml for some deep learning fun with some type-safety :)

    Poništi
  18. proslijedio/la je Tweet
    21. ruj 2018.

    A nice writeup of using OCaml and Tensorflow for reproducing some classic deep learning results, featuring the obligatory style-transferred bactrian camel!

    Poništi
  19. proslijedio/la je Tweet
    12. lip 2018.

    Massively Parallel Video Networks -- new work in collaboration with my DeepMind colleagues: Joao Carreira, Viorica Patraucean, Laurent Mazare, & Andrew Zisserman. Potential practical applications include real-time robotics, autonomous vehicles, and AR.

    Poništi
  20. proslijedio/la je Tweet

    Swift for TensorFlow is now open source! It is a new programming model for machine learning combining the performance of graphs and the expressivity of Eager execution, with first-class autodiff built right into the compiler Learn more on the TF blog ↓

    Poništi

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.

    Možda bi vam se svidjelo i ovo:

    ·