jax-ml/jax on Hacker News

▲ 1

How to Tame Your Jax: Navigating TPU VMs with TFRC in 2021

story by sillysaurusx · 2021-06-23 · 2 comments

…hashtag anchors. The goal is to submit to HN one of the replies in that thread, not the main page: https://github.com/google/jax/issues/2108#issuecomment-86623... Without that anchor, I see no possible way to submit it to HN, because the…

comment by sillysaurusx · 2021-06-23 · in How to Tame Your Jax: Navigating TPU VMs with TFRC in 2021

▲ 3

How to Tame Your Jax: Navigating TPU VMs with TFRC in 2021

story by sillysaurusx · 2021-06-22 · 4 comments

…). Then the gradient calculations don't mix well with GPUs, at least for ODEs ( https://github.com/google/jax/issues/5006 ). That lack of composability compared to Julia is the main issue that then creeps up in the bigger problems.

comment by ChrisRackauckas · 2021-06-21 · in Julia: Faster than Fortran, cleaner than Numpy

…Or this person just happened to pick it up for some reason (for, say, better TPU support)? [1] https://github.com/google/jax

comment by flakiness · 2021-06-09 · in GPT-J-6B: 6B JAX-Based Transformer

…shoyer on Oct 26, 2019 [–] The practical counterpoint to Dex is Jax, by many of the same authors: https://github.com/google/jax I’m optimistic that we’ll eventually be able to figure out how to build practical tools [..] The biggest…

comment by kelas · 2021-03-13 · in Dex: A research language for typed, functional array processing

…do well) Currently Jax has an in-progress stiff ODE solver that's about 200x slower than SciPy https://github.com/google/jax/issues/3686#issuecomment-65709... and SciPy (with JIT) is about 50x-100x slower than the pure Julia methods…

comment by ChrisRackauckas · 2020-09-09 · in Fast.ai and why Python is not the future of ML with Jeremy Howard

what do you think of something like https://github.com/google/jax that i guess claims to be a jack of all trades, in the sense that it'll transform programs (through XLA???) to various architectures?

comment by throwlaplace · 2020-04-17 · in A Taste of GPU Compute [video]

…on different things. We haven't tried combining them yet, but we think it would be fun to explore ( https://github.com/google/jax/issues/1870 ). For example, you could use Numba to hand write a numerical kernel that then participates in a…

comment by emu · 2020-04-08 · in Jax – Composable transformations of Python and NumPy programs

▲ 235

Jax – Composable transformations of Python and NumPy programs

story by lelf · 2020-04-08 · 65 comments

…so you get those automatically under jax.jit. There's also the jax.checkpoint decorator ( https://github.com/google/jax/pull/1749 ) which lets you control reverse-mode checkpointing yourself; you can use it recursively to implement…

comment by mattjjatgoogle · 2020-03-16 · in Using JAX, numpy, and optimization techniques to improve separable image filters

…to discover that Jax can't use a TPU's CPU, and that there are no plans to add this to Jax. https://github.com/google/jax/issues/2108#issuecomment-58154... A TPU's CPU is the only reason that TPUs are able to get such high performance on…

comment by sillysaurusx · 2020-02-17 · in Google/Trax – Understand and explore advanced deep learning

…programs. Design patterns, rules of thumb, and standard practices have yet to be defined. [1] https://github.com/google/jax [2] https://github.com/HIPS/autograd

comment by aaron-santos · 2020-02-16 · in Demystifying Differentiable Programming (2018)

Do these help? https://github.com/google/jax/tree/master/cloud_tpu_colabs The ODE solver might be close to what you want. I use Tensorflow 1.15. The world has been steadily pushing for Tensorflow 2.0 or Jax, but I like the simplicity of…

comment by sillysaurusx · 2020-01-07 · in AMD’s 64-Core Threadripper 3990X, only $3990 Coming February 7th

▲ 2

JAX now runs on Cloud TPUs (with example collab notebooks)

story by ArtWomb · 2019-12-15 · 0 comments

…functions). For example, there must be at least 4 different neural network libraries in JAX. (e.g., https://github.com/google/jax/blob/master/jax/experimental/s... , https://github.com/google/trax ).

comment by nihilest · 2019-12-08 · in Tensorflow User Experience

…flow interesting since it tries to be more pythonic while also being possible to jit-compile. ( https://github.com/google/jax/blob/master/jax/experimental/l... )

comment by schsam · 2019-12-07 · in Tensorflow User Experience

I really like the UX of JAX, https://github.com/google/jax The API is numpy plus like four functions. That's the beginning and end. It does JIT compilation under the hood, so can run quite fast.

comment by jlebar · 2019-12-07 · in Tensorflow User Experience

The practical counterpoint to Dex is Jax, by many of the same authors: https://github.com/google/jax I’m optimistic that we’ll eventually be able to figure out how to build practical tools for deep learning (and scientific computing for…

comment by shoyer · 2019-10-26 · in Dex: A research language for array processing in the Haskell/ML family

It has stax, which is a minimal example of how to build a high level library: https://github.com/google/jax/blob/master/jax/experimental/s... It seems that the JAX developers are focusing their time on making the core framework better and…

comment by ibab · 2019-10-13 · in Deep Learning with PyTorch: A 60 Minute Blitz [video]

I really like JAX as well: https://github.com/google/jax . It's younger than PyTorch and TF, but feels cleaner and more expressive. It has a very nice autodiff implementation (based on https://github.com/HIPS/autograd ) and performance is…

comment by ibab · 2019-10-13 · in Deep Learning with PyTorch: A 60 Minute Blitz [video]

…to transition from TF1 to TF2, consider doing the TF1 to pytorch transition instead. Or consider https://github.com/google/jax !

comment by jlebar · 2019-09-30 · in Tensorflow 2.0

For what it's worth, Jax (github.com/google/jax) now lets you use XLA to compile Numpy code. It'd be cool to see how that would stack up in a modern comparison.

comment by davmre · 2019-09-22 · in Weld: Accelerating numpy, scikit and pandas as much as 100x with Rust and LLVM

How does this compare to Jax? https://github.com/google/jax Why do this in the language instead of a library?

comment by hsaliak · 2019-09-06 · in Differentiable Programming Mega-Proposal

…case, you should be able to get really good performance without sacrificing readability. Also Jax - https://github.com/google/jax

comment by hsaliak · 2019-08-18 · in Python vs. Rust for Neural Networks

…capabilities. There is recent work to put numpy on the same footing for deep learning. Check https://github.com/google/jax

comment by jimmy_dean · 2019-08-18 · in Python vs. Rust for Neural Networks

JAX (github.com/google/jax), which is being developed by many of the authors of Autograd, is a probably a better comparison. At the cost of requiring you to rewrite control flow in a functional way, it eliminates Python's overhead by…

comment by shoyer · 2019-07-19 · in A Differentiable Programming System to Bridge ML and Scientific Computing

…that there are only 2 people working on the project, if that. The name of the project is JAX: https://github.com/google/jax . It's arguably a lower level framework on top of which something like PyTorch could be built.

comment by m0zg · 2019-01-11 · in NLP Architect by Intel AI Lab

Check out github.com/google/jax, it’s NumPy on the GPU with automatic differentiation, JIT and autobatching.

comment by alexbw · 2018-12-08 · in PyTorch 1.0 is out

▲ 132

JAX: Numpy with Gradients, GPUs and TPUs

story by one-more-minute · 2018-12-08 · 29 comments

▲ 3

Google jax

story by sahin-boydas · 2018-11-22 · 0 comments

Those code examples look very clean: https://github.com/google/jax/tree/master/examples

comment by nabla9 · 2018-11-21 · in Compiling machine learning programs via high-level tracing [pdf]