[API Compatibility No.361] Param alias for lerp using decorator -part

python/paddle/tensor/math.py

@@ -4821,8 +4821,14 @@ def logit_(
         return _C_ops.logit_(x, eps)
 
 
+@param_two_alias(["x", "input"], ["y", "end"])
 def lerp(
-    x: Tensor, y: Tensor, weight: float | Tensor, name: str | None = None
+    x: Tensor,
+    y: Tensor,
+    weight: float | Tensor,
+    name: str | None = None,
+    *,
+    out: Tensor | None = None,
 ) -> Tensor:
     r"""
     Does a linear interpolation between x and y based on weight.
@@ -4834,9 +4840,12 @@ def lerp(
 
     Args:
         x (Tensor): An N-D Tensor with starting points, the data type is bfloat16, float16, float32, float64.
+            Alias: ``input`` .
         y (Tensor): An N-D Tensor with ending points, the data type is bfloat16, float16, float32, float64.
+            Alias: ``end`` .
         weight (float|Tensor): The weight for the interpolation formula. When weight is Tensor, the data type is bfloat16, float16, float32, float64.
         name (str|None, optional): Name for the operation (optional, default is None). For more information, please refer to :ref:`api_guide_Name`.
+        out (Tensor|None, optional): The output Tensor. If provided, the result will be stored in `out`. Default is None.
 
     Returns:
         out (Tensor): An N-D Tensor, the shape and data type is the same with input.
@@ -4862,7 +4871,8 @@ def lerp(
             weight = paddle.full(shape=[], fill_value=weight, dtype=x.dtype)
 
     if in_dynamic_or_pir_mode():
-        return _C_ops.lerp(x, y, weight)
+        if out is None:
+            return _C_ops.lerp(x, y, weight, out=out)
     else:
         check_variable_and_dtype(
             x, 'x', ['uint16', 'float16', 'float32', 'float64'], 'lerp'
@@ -4879,11 +4889,13 @@ def lerp(
 
         helper = LayerHelper('lerp', **locals())
         inputs = {'X': x, 'Y': y, 'Weight': weight}
-        out = helper.create_variable_for_type_inference(dtype=x.dtype)
+        if out is None:
+            out = helper.create_variable_for_type_inference(dtype=x.dtype)
         helper.append_op(type='lerp', inputs=inputs, outputs={'Out': out})
         return out
 
 
+@param_two_alias(["x", "input"], ["y", "end"])
 @inplace_apis_in_dygraph_only
 def lerp_(
     x: Tensor, y: Tensor, weight: float | Tensor, name: str | None = None

test/legacy_test/test_lerp_op.py

@@ -234,6 +234,32 @@ def test_x_y_broadcast_w(self):
         np.testing.assert_allclose(res_ref, out.numpy(), rtol=1e-05)
         paddle.enable_static()
 
+    def test_alias(self):
+        paddle.disable_static()
+        x = paddle.to_tensor(self.x)
+        y = paddle.to_tensor(self.y)
+        w = paddle.to_tensor(self.w)
+
+        # Test with input, end
+        out1 = paddle.lerp(input=x, end=y, weight=w)
+        np.testing.assert_allclose(out1.numpy(), self.res_ref, rtol=1e-05)
+
+        # Test with x, y (alias)
+        out2 = paddle.lerp(x=x, y=y, weight=w)
+        np.testing.assert_allclose(out2.numpy(), self.res_ref, rtol=1e-05)
+        paddle.enable_static()
+
+    def test_out(self):
+        paddle.disable_static()
+        x = paddle.to_tensor(self.x)
+        y = paddle.to_tensor(self.y)
+        w = paddle.to_tensor(self.w)
+        out = paddle.empty_like(x)
+
+        paddle.lerp(x, y, w, out=out)
+        np.testing.assert_allclose(out.numpy(), self.res_ref, rtol=1e-05)
+        paddle.enable_static()
+
 
 @unittest.skipIf(
     not (core.is_compiled_with_cuda() or is_custom_device())