This is an automated email from the ASF dual-hosted git repository.
tlopex pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git
The following commit(s) were added to refs/heads/main by this push:
new 6041e9f455 [Relax][PyTorch] Add support for antialiased bilinear
upsampling (#18500)
6041e9f455 is described below
commit 6041e9f455ebd694ea6f2dcb755897b2a9aec9fe
Author: Guan-Ming (Wesley) Chiu <[email protected]>
AuthorDate: Wed Nov 26 13:17:32 2025 +0800
[Relax][PyTorch] Add support for antialiased bilinear upsampling (#18500)
## Related Issue
closes https://github.com/apache/tvm/issues/18365
## How
- add support for antialiased bilinear upsampling
---
.../frontend/torch/exported_program_translator.py | 17 ++++++++++
.../relax/test_frontend_from_exported_program.py | 37 ++++++++++++++++++++++
2 files changed, 54 insertions(+)
diff --git a/python/tvm/relax/frontend/torch/exported_program_translator.py
b/python/tvm/relax/frontend/torch/exported_program_translator.py
index 95b0e05361..7af8774ee3 100644
--- a/python/tvm/relax/frontend/torch/exported_program_translator.py
+++ b/python/tvm/relax/frontend/torch/exported_program_translator.py
@@ -298,6 +298,22 @@ class ExportedProgramImporter(BaseFXGraphImporter):
x, size=size, scale_factor=scale_factor, method="linear",
align_corners=align_corners
)
+ def _upsample_bilinear2d_aa(self, node: fx.Node) -> relax.Var:
+ x = self.env[node.args[0]]
+ size = node.args[1] if len(node.args) > 1 else
node.kwargs.get("output_size", None)
+ align_corners = (
+ node.args[2] if len(node.args) > 2 else
node.kwargs.get("align_corners", False)
+ )
+ scale_factor = (
+ node.args[3] if len(node.args) > 3 else
node.kwargs.get("scale_factors", None)
+ )
+
+ # Note: TVM's resize2d doesn't have explicit antialias support.
+ # For upsampling, antialiasing has minimal effect, so we use regular
bilinear.
+ return self._upsample_impl(
+ x, size=size, scale_factor=scale_factor, method="linear",
align_corners=align_corners
+ )
+
def _upsample_nearest2d(self, node: fx.node) -> relax.Var:
x = self.env[node.args[0]]
size = node.args[1] if len(node.args) > 1 else node.kwargs.get("size",
None)
@@ -1218,6 +1234,7 @@ class ExportedProgramImporter(BaseFXGraphImporter):
"scaled_dot_product_attention.default":
self._scaled_dot_product_attention,
"unbind.int": self._unbind,
"upsample_bilinear2d.vec": self._upsample_bilinear2d,
+ "_upsample_bilinear2d_aa.default": self._upsample_bilinear2d_aa,
"upsample_nearest2d.vec": self._upsample_nearest2d,
"upsample_bicubic2d.vec": self._upsample_bicubic2d,
# statistical
diff --git a/tests/python/relax/test_frontend_from_exported_program.py
b/tests/python/relax/test_frontend_from_exported_program.py
index d4c23bfdd5..98c6c6d014 100644
--- a/tests/python/relax/test_frontend_from_exported_program.py
+++ b/tests/python/relax/test_frontend_from_exported_program.py
@@ -4703,6 +4703,43 @@ def test_interpolate():
verify_model(InterpolateBicubic(), example_args, {}, expected_bicubic)
+def test_interpolate_antialiased():
+ """Test bilinear interpolation with antialiasing enabled."""
+
+ class InterpolateBilinearAA(Module):
+ def forward(self, input):
+ return torch.nn.functional.interpolate(
+ input, size=(64, 64), mode="bilinear", align_corners=False,
antialias=True
+ )
+
+ @tvm.script.ir_module
+ class expected_bilinear_aa:
+ @R.function
+ def main(
+ input: R.Tensor((1, 3, 32, 32), dtype="float32")
+ ) -> R.Tuple(R.Tensor((1, 3, 64, 64), dtype="float32")):
+ with R.dataflow():
+ lv: R.Tensor((1, 3, 64, 64), dtype="float32") =
R.image.resize2d(
+ input,
+ R.shape([64, 64]),
+ roi=[T.float32(0.0), T.float32(0.0), T.float32(0.0),
T.float32(0.0)],
+ layout="NCHW",
+ method="linear",
+ coordinate_transformation_mode="half_pixel",
+ rounding_method="round",
+ cubic_alpha=-0.75,
+ cubic_exclude=0,
+ extrapolation_value=0.0,
+ out_dtype="void",
+ )
+ gv: R.Tuple(R.Tensor((1, 3, 64, 64), dtype="float32")) = (lv,)
+ R.output(gv)
+ return gv
+
+ example_args = (torch.randn(1, 3, 32, 32, dtype=torch.float32),)
+ verify_model(InterpolateBilinearAA(), example_args, {},
expected_bilinear_aa)
+
+
def test_mean():
class Mean(Module):
def forward(self, input):
