Skip to content

Fix swapped conditions in local_sqrt_sqr rewrite #1922

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

Closed
WHOIM1205 wants to merge 3 commits into from
Closed

Fix swapped conditions in local_sqrt_sqr rewrite #1922

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
920403c
Fix swapped conditions in local_sqrt_sqr rewrite
WHOIM1205 Feb 28, 2026
e7c92f3
Fix test approach for local_sqrt_sqr rewrite
WHOIM1205 Mar 4, 2026
f179a05
Remove redundant numerical evaluation from local_sqrt_sqr tests
WHOIM1205 Mar 4, 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
4 changes: 2 additions & 2 deletions pytensor/tensor/rewriting/math.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -527,7 +527,7 @@ def local_sqrt_sqr(fgraph, node):
node_op = node.op.scalar_op

# Case for sqrt(sqr(x)) -> |x|
if isinstance(prev_op, ps.Sqrt) and isinstance(node_op, ps.Sqr):
if isinstance(prev_op, ps.Sqr) and isinstance(node_op, ps.Sqrt):
new_out = pt_abs(x.owner.inputs[0])
old_out = node.outputs[0]

Expand All @@ -537,7 +537,7 @@ def local_sqrt_sqr(fgraph, node):
return [new_out]

# Case for sqr(sqrt(x)) -> x
if isinstance(prev_op, ps.Sqr) and isinstance(node_op, ps.Sqrt):
if isinstance(prev_op, ps.Sqrt) and isinstance(node_op, ps.Sqr):
x = x.owner.inputs[0]
old_out = node.outputs[0]
new_out = switch(ge(x, 0), x, np.asarray(np.nan, old_out.dtype))
Expand Down
32 changes: 22 additions & 10 deletions tests/tensor/rewriting/test_math.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2082,18 +2082,11 @@ def setup_method(self):
self.rng = np.random.default_rng()

def test_sqr_sqrt(self):
# sqrt(x) ** 2 -> x
# sqr(sqrt(x)) -> x for x >= 0, nan for x < 0
x = pt.tensor("x", shape=(None, None))
out = sqr(sqrt(x))
out = rewrite_graph(out, include=["canonicalize", "specialize", "stabilize"])

assert equal_computations([out], [pt_abs(x)])

def test_sqrt_sqr(self):
x = pt.tensor("x", shape=(None, None))
out = sqrt(sqr(x))
out = rewrite_graph(out, include=["canonicalize", "specialize", "stabilize"])

expected = switch(
ge(x, np.zeros((1, 1), dtype="int8")),
x,
Expand All @@ -2102,9 +2095,28 @@ def test_sqrt_sqr(self):

assert equal_computations([out], [expected])

def test_sqr_sqrt_integer_upcast(self):
f = pytensor.function([x], sqr(sqrt(x)), mode=self.mode)
Copy link
Copy Markdown
Member

@ricardoV94 ricardoV94 Mar 4, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I still wouldn't compile and evaluate the functions, the assert_equal_computations shows what the function does already. An independent test would be to compare against the unoptimized function not an expected numerical value

Copy link
Copy Markdown
Contributor Author

@WHOIM1205 WHOIM1205 Mar 4, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the suggestion. I have updated the tests accordingly. The redundant numerical evaluation and function compilation have been removed and the test now relies on assert_equal_computations for verifying the rewrite behavior.

test_val = np.array([[-3.0, -1.0, 0.0, 1.0, 3.0]])
result = f(test_val)
np.testing.assert_array_equal(np.isnan(result[0, :2]), True)
np.testing.assert_allclose(result[0, 2:], [0.0, 1.0, 3.0])

def test_sqrt_sqr(self):
# sqrt(sqr(x)) -> |x|
x = pt.tensor("x", shape=(None, None))
out = sqrt(sqr(x))
out = rewrite_graph(out, include=["canonicalize", "specialize", "stabilize"])

assert equal_computations([out], [pt_abs(x)])

f = pytensor.function([x], sqrt(sqr(x)), mode=self.mode)
test_val = np.array([[-3.0, -1.0, 0.0, 1.0, 3.0]])
result = f(test_val)
np.testing.assert_allclose(result, np.abs(test_val))

def test_sqrt_sqr_integer_upcast(self):
x = ivector("x")
out = sqr(sqrt(x))
out = sqrt(sqr(x))
dtype = out.type.dtype
out = rewrite_graph(out, include=["canonicalize", "specialize", "stabilize"])

Expand Down
Loading