JAX: Autograd and XLA

Quickstart | Transformations | Install guide | Neural net libraries | Change logs | Reference docs

What is JAX?

JAX is Autograd and XLA, brought together for high-performance machine learning research.

With its updated version of Autograd, JAX can automatically differentiate native Python and NumPy functions. It can differentiate through loops, branches, recursion, and closures, and it can take derivatives of derivatives of derivatives. It supports reverse-mode differentiation (a.k.a. backpropagation) via grad as well as forward-mode differentiation, and the two can be composed arbitrarily to any order.

What’s new is that JAX uses XLA to compile and run your NumPy programs on GPUs and TPUs. Compilation happens under the hood by default, with library calls getting just-in-time compiled and executed. But JAX also lets you just-in-time compile your own Python functions into XLA-optimized kernels using a one-function API, jit. Compilation and automatic differentiation can be composed arbitrarily, so you can express sophisticated algorithms and get maximal performance without leaving Python. You can even program multiple GPUs or TPU cores at once using pmap, and differentiate through the whole thing.

Dig a little deeper, and you'll see that JAX is really an extensible system for composable function transformations. Both grad and jit are instances of such transformations. Others are vmap for automatic vectorization and pmap for single-program multiple-data (SPMD) parallel programming of multiple accelerators, with more to come.

This is a research project, not an official Google product. Expect bugs and sharp edges. Please help by trying it out, reporting bugs, and letting us know what you think!

import jax.numpy as jnp
from jax import grad, jit, vmap

def predict(params, inputs):
  for W, b in params:
    outputs = jnp.dot(inputs, W) + b
    inputs = jnp.tanh(outputs)  # inputs to the next layer
  return outputs                # no activation on last layer

def loss(params, inputs, targets):
  preds = predict(params, inputs)
  return jnp.sum((preds - targets)**2)

grad_loss = jit(grad(loss))  # compiled gradient evaluation function
perex_grads = jit(vmap(grad_loss, in_axes=(None, 0, 0)))  # fast per-example grads

Contents

• Quickstart: Colab in the Cloud
• Transformations
• Current gotchas
• Installation
• Neural net libraries
• Citing JAX
• Reference documentation

Quickstart: Colab in the Cloud

Jump right in using a notebook in your browser, connected to a Google Cloud GPU. Here are some starter notebooks:

• The basics: NumPy on accelerators, grad for differentiation, jit for compilation, and vmap for vectorization
• Training a Simple Neural Network, with TensorFlow Dataset Data Loading

JAX now runs on Cloud TPUs. To try out the preview, see the Cloud TPU Colabs.

For a deeper dive into JAX:

• The Autodiff Cookbook, Part 1: easy and powerful automatic differentiation in JAX
• Common gotchas and sharp edges
• See the full list of notebooks.

You can also take a look at the mini-libraries in jax.example_libraries, like stax for building neural networks and optimizers for first-order stochastic optimization, or the examples.

Transformations

At its core, JAX is an extensible system for transforming numerical functions. Here are four transformations of primary interest: grad, jit, vmap, and pmap.

Automatic differentiation with grad

JAX has roughly the same API as Autograd. The most popular function is grad for reverse-mode gradients:

from jax import grad
import jax.numpy as jnp

def tanh(x):  # Define a function
  y = jnp.exp(-2.0 * x)
  return (1.0 - y) / (1.0 + y)

grad_tanh = grad(tanh)  # Obtain its gradient function
print(grad_tanh(1.0))   # Evaluate it at x = 1.0
# prints 0.4199743

You can differentiate to any order with grad.

print(grad(grad(grad(tanh)))(1.0))
# prints 0.62162673

For more advanced autodiff, you can use jax.vjp for reverse-mode vector-Jacobian products and jax.jvp for forward-mode Jacobian-vector products. The two can be composed arbitrarily with one another, and with other JAX transformations. Here's one way to compose those to make a function that efficiently computes full Hessian matrices:

from jax import jit, jacfwd, jacrev

def hessian(fun):
  return jit(jacfwd(jacrev(fun)))

As with Autograd, you're free to use differentiation with Python control structures:

def abs_val(x):
  if x > 0:
    return x
  else:
    return -x

abs_val_grad = grad(abs_val)
print(abs_val_grad(1.0))   # prints 1.0
print(abs_val_grad(-1.0))  # prints -1.0 (abs_val is re-evaluated)

See the reference docs on automatic differentiation and the JAX Autodiff Cookbook for more.

Compilation with jit

You can use XLA to compile your functions end-to-end with jit, used either as an @jit decorator or as a higher-order function.

import jax.numpy as jnp
from jax import jit

def slow_f(x):
  # Element-wise ops see a large benefit from fusion
  return x * x + x * 2.0

x = jnp.ones((5000, 5000))
fast_f = jit(slow_f)
%timeit -n10 -r3 fast_f(x)  # ~ 4.5 ms / loop on Titan X
%timeit -n10 -r3 slow_f(x)  # ~ 14.5 ms / loop (also on GPU via JAX)

You can mix jit and grad and any other JAX transformation however you like.

Using jit puts constraints on the kind of Python control flow the function can use; see the Gotchas Notebook for more.

Auto-vectorization with vmap

vmap is the vectorizing map. It has the familiar semantics of mapping a function along array axes, but instead of keeping the loop on the outside, it pushes the loop down into a function’s primitive operations for better performance.

Using vmap can save you from having to carry around batch dimensions in your code. For example, consider this simple unbatched neural network prediction function:

def predict(params, input_vec):
  assert input_vec.ndim == 1
  activations = input_vec
  for W, b in params:
    outputs = jnp.dot(W, activations) + b  # `activations` on the right-hand side!
    activations = jnp.tanh(outputs)        # inputs to the next layer
  return outputs                           # no activation on last layer

We often instead write jnp.dot(activations, W) to allow for a batch dimension on the left side of activations, but we’ve written this particular prediction function to apply only to single input vectors. If we wanted to apply this function to a batch of inputs at once, semantically we could just write

from functools import partial
predictions = jnp.stack(list(map(partial(predict, params), input_batch)))

But pushing one example through the network at a time would be slow! It’s better to vectorize the computation, so that at every layer we’re doing matrix-matrix multiplication rather than matrix-vector multiplication.

The vmap function does that transformation for us. That is, if we write

from jax import vmap
predictions = vmap(partial(predict, params))(input_batch)
# or, alternatively
predictions = vmap(predict, in_axes=(None, 0))(params, input_batch)

then the vmap function will push the outer loop inside the function, and our machine will end up executing matrix-matrix multiplications exactly as if we’d done the batching by hand.

It’s easy enough to manually batch a simple neural network without vmap, but in other cases manual vectorization can be impractical or impossible. Take the problem of efficiently computing per-example gradients: that is, for a fixed set of parameters, we want to compute the gradient of our loss function evaluated separately at each example in a batch. With vmap, it’s easy:

per_example_gradients = vmap(partial(grad(loss), params))(inputs, targets)

Of course, vmap can be arbitrarily composed with jit, grad, and any other JAX transformation! We use vmap with both forward- and reverse-mode automatic differentiation for fast Jacobian and Hessian matrix calculations in jax.jacfwd, jax.jacrev, and jax.hessian.

SPMD programming with pmap

For parallel programming of multiple accelerators, like multiple GPUs, use pmap. With pmap you write single-program multiple-data (SPMD) programs, including fast parallel collective communication operations. Applying pmap will mean that the function you write is compiled by XLA (similarly to jit), then replicated and executed in parallel across devices.

Here's an example on an 8-GPU machine:

from jax import random, pmap
import jax.numpy as jnp

# Create 8 random 5000 x 6000 matrices, one per GPU
keys = random.split(random.PRNGKey(0), 8)
mats = pmap(lambda key: random.normal(key, (5000, 6000)))(keys)

# Run a local matmul on each device in parallel (no data transfer)
result = pmap(lambda x: jnp.dot(x, x.T))(mats)  # result.shape is (8, 5000, 5000)

# Compute the mean on each device in parallel and print the result
print(pmap(jnp.mean)(result))
# prints [1.1566595 1.1805978 ... 1.2321935 1.2015157]

In addition to expressing pure maps, you can use fast collective communication operations between devices:

from functools import partial
from jax import lax

@partial(pmap, axis_name='i')
def normalize(x):
  return x / lax.psum(x, 'i')

print(normalize(jnp.arange(4.)))
# prints [0.         0.16666667 0.33333334 0.5       ]

You can even nest pmap functions for more sophisticated communication patterns.

It all composes, so you're free to differentiate through parallel computations:

from jax import grad

@pmap
def f(x):
  y = jnp.sin(x)
  @pmap
  def g(z):
    return jnp.cos(z) * jnp.tan(y.sum()) * jnp.tanh(x).sum()
  return grad(lambda w: jnp.sum(g(w)))(x)

print(f(x))
# [[ 0.        , -0.7170853 ],
#  [-3.1085174 , -0.4824318 ],
#  [10.366636  , 13.135289  ],
#  [ 0.22163185, -0.52112055]]

print(grad(lambda x: jnp.sum(f(x)))(x))
# [[ -3.2369726,  -1.6356447],
#  [  4.7572474,  11.606951 ],
#  [-98.524414 ,  42.76499  ],
#  [ -1.6007166,  -1.2568436]]

When reverse-mode differentiating a pmap function (e.g. with grad), the backward pass of the computation is parallelized just like the forward pass.

See the SPMD Cookbook and the SPMD MNIST classifier from scratch example for more.

Current gotchas

For a more thorough survey of current gotchas, with examples and explanations, we highly recommend reading the Gotchas Notebook. Some standouts:

1. JAX transformations only work on pure functions, which don't have side-effects and respect referential transparency (i.e. object identity testing with is isn't preserved). If you use a JAX transformation on an impure Python function, you might see an error like Exception: Can't lift Traced... or Exception: Different traces at same level.
2. In-place mutating updates of arrays, like x[i] += y, aren't supported, but there are functional alternatives. Under a jit, those functional alternatives will reuse buffers in-place automatically.
3. Random numbers are different, but for good reasons.
4. If you're looking for convolution operators, they're in the jax.lax package.
5. JAX enforces single-precision (32-bit, e.g. float32) values by default, and to enable double-precision (64-bit, e.g. float64) one needs to set the jax_enable_x64 variable at startup (or set the environment variable JAX_ENABLE_X64=True). On TPU, JAX uses 32-bit values by default for everything except internal temporary variables in 'matmul-like' operations, such as jax.numpy.dot and lax.conv. Those ops have a precision parameter which can be used to simulate true 32-bit, with a cost of possibly slower runtime.
6. Some of NumPy's dtype promotion semantics involving a mix of Python scalars and NumPy types aren't preserved, namely np.add(1, np.array([2], np.float32)).dtype is float64 rather than float32.
7. Some transformations, like jit, constrain how you can use Python control flow. You'll always get loud errors if something goes wrong. You might have to use jit's static_argnums parameter, structured control flow primitives like lax.scan, or just use jit on smaller subfunctions.

Installation

JAX is written in pure Python, but it depends on XLA, which needs to be installed as the jaxlib package. Use the following instructions to install a binary package with pip or conda, or to build JAX from source.

We support installing or building jaxlib on Linux (Ubuntu 20.04 or later) and macOS (10.12 or later) platforms. There is also experimental native Windows support.

Windows users can use JAX on CPU and GPU via the Windows Subsystem for Linux, or alternatively they can use the experimental native Windows CPU-only support.

pip installation: CPU

We currently release jaxlib wheels for the following operating systems and architectures:

• Linux, x86-64
• Mac, Intel
• Mac, ARM
• Windows, x86-64 (experimental)

To install a CPU-only version of JAX, which might be useful for doing local development on a laptop, you can run

pip install --upgrade pip
pip install --upgrade "jax[cpu]"

On Windows, you may also need to install the Microsoft Visual Studio 2019 Redistributable if it is not already installed on your machine.

Other operating systems and architectures require building from source. Trying to pip install on other operating systems and architectures may lead to jaxlib not being installed alongside jax, although jax may successfully install (but fail at runtime).

pip installation: GPU (CUDA, installed via pip, easier)

There are two ways to install JAX with NVIDIA GPU support: using CUDA and CUDNN installed from pip wheels, and using a self-installed CUDA/CUDNN. We recommend installing CUDA and CUDNN using the pip wheels, since it is much easier!

JAX supports NVIDIA GPUs that have SM version 5.2 (Maxwell) or newer. Note that Kepler-series GPUs are no longer supported by JAX since NVIDIA has dropped support for Kepler GPUs in its software.

You must first install the NVIDIA driver. We recommend installing the newest driver available from NVIDIA, but the driver must be version >= 525.60.13 for CUDA 12 and >= 450.80.02 for CUDA 11 on Linux. If you need to use an newer CUDA toolkit with an older driver, for example on a cluster where you cannot update the NVIDIA driver easily, you may be able to use the CUDA forward compatibility packages that NVIDIA provides for this purpose.

pip install --upgrade pip

# CUDA 12 installation
# Note: wheels only available on linux.
pip install --upgrade "jax[cuda12_pip]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

# CUDA 11 installation
# Note: wheels only available on linux.
pip install --upgrade "jax[cuda11_pip]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

pip installation: GPU (CUDA, installed locally, harder)

If you prefer to use a preinstalled copy of CUDA, you must first install CUDA and CuDNN.

JAX provides pre-built CUDA-compatible wheels for Linux x86_64 only. Other combinations of operating system and architecture are possible, but require building from source.

You should use an NVIDIA driver version that is at least as new as your CUDA toolkit's corresponding driver version. If you need to use an newer CUDA toolkit with an older driver, for example on a cluster where you cannot update the NVIDIA driver easily, you may be able to use the CUDA forward compatibility packages that NVIDIA provides for this purpose.

JAX currently ships two CUDA wheel variants:

• CUDA 12.0 and CuDNN 8.9.
• CUDA 11.8 and CuDNN 8.6.

You may use a JAX wheel provided the major version of your CUDA and CuDNN installation matches, and the minor version is at least as new as the version JAX expects. For example, you would be able to use the CUDA 12.0 wheel with CUDA 12.1 and CuDNN 8.9.

Your CUDA installation must also be new enough to support your GPU. If you have an Ada Lovelace (e.g., RTX 4080) or Hopper (e.g., H100) GPU, you must use CUDA 11.8 or newer.

To install, run

pip install --upgrade pip

# Installs the wheel compatible with CUDA 12 and cuDNN 8.9 or newer.
# Note: wheels only available on linux.
pip install --upgrade "jax[cuda12_local]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

# Installs the wheel compatible with CUDA 11 and cuDNN 8.6 or newer.
# Note: wheels only available on linux.
pip install --upgrade "jax[cuda11_local]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html

These pip installations do not work with Windows, and may fail silently; see above.

You can find your CUDA version with the command:

nvcc --version

Some GPU functionality expects the CUDA installation to be at /usr/local/cuda-X.X, where X.X should be replaced with the CUDA version number (e.g. cuda-11.8). If CUDA is installed elsewhere on your system, you can either create a symlink:

sudo ln -s /path/to/cuda /usr/local/cuda-X.X

Please let us know on the issue tracker if you run into any errors or problems with the prebuilt wheels.

pip installation: Google Cloud TPU

JAX provides pre-built wheels for Google Cloud TPU. To install JAX along with appropriate versions of jaxlib and libtpu, you can run the following in your cloud TPU VM:

pip install jax[tpu] -f https://storage.googleapis.com/jax-releases/libtpu_releases.html

For interactive notebook users: Colab TPUs no longer support JAX as of JAX version 0.4. However, for an interactive TPU notebook in the cloud, you can use Kaggle TPU notebooks, which fully support JAX.

pip installation: Apple GPUs

Apple provides an experimental Metal plugin for Apple GPU hardware. For details, see Apple's JAX on Metal documentation.

There are several caveats with the Metal plugin:

• the Metal plugin is new and experimental and has a number of known issues. Please report any issues on the JAX issue tracker.
• the Metal plugin currently requires very specific versions of jax and jaxlib. This restriction will be relaxed over time as the plugin API matures.

Conda installation

There is a community-supported Conda build of jax. To install using conda, simply run

conda install jax -c conda-forge

To install on a machine with an NVIDIA GPU, run

conda install jaxlib=*=*cuda* jax cuda-nvcc -c conda-forge -c nvidia

Note the cudatoolkit distributed by conda-forge is missing ptxas, which JAX requires. You must therefore either install the cuda-nvcc package from the nvidia channel, or install CUDA on your machine separately so that ptxas is in your path. The channel order above is important (conda-forge before nvidia).

If you would like to override which release of CUDA is used by JAX, or to install the CUDA build on a machine without GPUs, follow the instructions in the Tips & tricks section of the conda-forge website.

See the conda-forge jaxlib and jax repositories for more details.

Building JAX from source

See Building JAX from source.

Neural network libraries

Multiple Google research groups develop and share libraries for training neural networks in JAX. If you want a fully featured library for neural network training with examples and how-to guides, try Flax.

In addition, DeepMind has open-sourced an ecosystem of libraries around JAX including Haiku for neural network modules, Optax for gradient processing and optimization, RLax for RL algorithms, and chex for reliable code and testing. (Watch the NeurIPS 2020 JAX Ecosystem at DeepMind talk here)

Citing JAX

To cite this repository:

@software{jax2018github,
  author = {James Bradbury and Roy Frostig and Peter Hawkins and Matthew James Johnson and Chris Leary and Dougal Maclaurin and George Necula and Adam Paszke and Jake Vander{P}las and Skye Wanderman-{M}ilne and Qiao Zhang},
  title = {{JAX}: composable transformations of {P}ython+{N}um{P}y programs},
  url = {http://github.com/google/jax},
  version = {0.3.13},
  year = {2018},
}

In the above bibtex entry, names are in alphabetical order, the version number is intended to be that from jax/version.py, and the year corresponds to the project's open-source release.

A nascent version of JAX, supporting only automatic differentiation and compilation to XLA, was described in a paper that appeared at SysML 2018. We're currently working on covering JAX's ideas and capabilities in a more comprehensive and up-to-date paper.

Reference documentation

For details about the JAX API, see the reference documentation.

For getting started as a JAX developer, see the developer documentation.