Skip to content

JAX: fix Alloc failing under jax.jit when shape inputs are constants #2050

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jaj42 wants to merge 3 commits into
base: main
Choose a base branch
from
Open

JAX: fix Alloc failing under jax.jit when shape inputs are constants #2050

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
cee7141
JAX: fix Alloc failing under jax.jit when shape inputs are constants
jaj42 Apr 13, 2026
109b55b
Simplify the fix: break for-loop early, remove additional if statemen…
jaj42 Apr 13, 2026
108df00
fix linting
jaj42 Apr 14, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 13 additions & 1 deletion pytensor/link/jax/dispatch/tensor_basic.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,8 +41,20 @@ def allocempty(*shape):

@jax_funcify.register(Alloc)
def jax_funcify_Alloc(op, node, **kwargs):
static_shapes = []
for shape_input in node.inputs[1:]:
try:
static_shapes.append(int(get_scalar_constant_value(shape_input)))
except NotScalarConstantError:
concrete_shape = None
break
else:
concrete_shape = tuple(static_shapes)

def alloc(x, *shape):
res = jnp.broadcast_to(x, shape)
res = jnp.broadcast_to(
x, concrete_shape if concrete_shape is not None else shape
)
Alloc._check_runtime_broadcast(node, jnp.asarray(x), res.shape)
return res

Expand Down
Loading