Agjax is a jax wrapper for autograd-differentiable functions. It allows existing code built with autograd to be used with the jax framework. In particular, agjax allows an arbitrary autograd function to be differentiated using jax.grad
. Several other function transformations (e.g. compilation via jax.jit
) are not supported.
Agjax is not yet a published package, but can be installed by cloning this repository and running pip install -e agjax
or run make install
.
Basic usage is as follows:
@agjax.wrap_for_jax
def fn(x, y):
return x * npa.cos(y)
grad = jax.grad(fn, argnums=(0, 1))(1.0, 0.0)
print(f"grad = {grad}")
grad = (Array(1., dtype=float32), Array(0., dtype=float32))
Agjax is intended to be quite general, and can support functions with multiple inputs and outputs as well as functions that have nondifferentiable outputs or arguments that cannot be differentiated with respect to. These should be specified using nondiff_argnums
and nondiff_outputnums
arguments to wrap_for_jax
.
@functools.partial(
agjax.wrap_for_jax, nondiff_argnums=(2,), nondiff_outputnums=(1,)
)
def fn(x, y, string_arg):
return x * npa.cos(y), string_arg * 2
(value, aux), grad = jax.value_and_grad(
fn, argnums=(0, 1), has_aux=True
)(1.0, 0.0, "test")
print(f"value = {value}")
print(f" aux = {aux}")
print(f" grad = {grad}")
value = 1.0
aux = testtest
grad = (Array(1., dtype=float32), Array(0., dtype=float32))