Merge branch 'main' into stationarity_property

Author: PierreQuinton

src/torchjd/autojac/_transform/grad.py

@@ -37,12 +37,7 @@ def _differentiate(self, grad_outputs: Sequence[Tensor]) -> tuple[Tensor, ...]:
             return tuple()
 
         if len(outputs) == 0:
-            return tuple(
-                [
-                    torch.empty(input.shape, device=input.device, dtype=input.dtype)
-                    for input in inputs
-                ]
-            )
+            return tuple([torch.zeros_like(input) for input in inputs])
 
         optional_grads = torch.autograd.grad(
             outputs,

tests/unit/autojac/_transform/test_grad.py

@@ -62,6 +62,24 @@ def test_empty_inputs_2():
     assert_tensor_dicts_are_close(gradients, expected_gradients)
 
 
+def test_empty_outputs():
+    """
+    Tests that the Grad transform works correctly when the `outputs` parameter is an empty
+    `Iterable`.
+    """
+
+    a1 = torch.tensor(1.0, requires_grad=True)
+    a2 = torch.tensor([1.0, 2.0], requires_grad=True)
+    input = Gradients({})
+
+    grad = Grad(outputs=[], inputs=[a1, a2])
+
+    gradients = grad(input)
+    expected_gradients = {a1: torch.zeros_like(a1), a2: torch.zeros_like(a2)}
+
+    assert_tensor_dicts_are_close(gradients, expected_gradients)
+
+
 def test_retain_graph():
     """Tests that the `Grad` transform behaves as expected with the `retain_graph` flag."""
 

tests/unit/autojac/_transform/test_jac.py

@@ -72,6 +72,28 @@ def test_empty_inputs_2(chunk_size: int | None):
     assert_tensor_dicts_are_close(jacobians, expected_jacobians)
 
 
+@mark.parametrize("chunk_size", [1, 3, None])
+def test_empty_outputs(chunk_size: int | None):
+    """
+    Tests that the Jac transform works correctly when the `outputs` parameter is an empty
+    `Iterable`.
+    """
+
+    a1 = torch.tensor(1.0, requires_grad=True)
+    a2 = torch.tensor([1.0, 2.0], requires_grad=True)
+    input = Jacobians({})
+
+    jac = Jac(outputs=[], inputs=[a1, a2], chunk_size=chunk_size)
+
+    jacobians = jac(input)
+    expected_jacobians = {
+        a1: torch.empty_like(a1).unsqueeze(0)[:0],  # Jacobian with no row
+        a2: torch.empty_like(a2).unsqueeze(0)[:0],  # Jacobian with no row
+    }
+
+    assert_tensor_dicts_are_close(jacobians, expected_jacobians)
+
+
 def test_retain_graph():
     """Tests that the `Jac` transform behaves as expected with the `retain_graph` flag."""