Skip to content

[ONNX] Fix onnx convtranspose error #9938

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

Merged
masahi merged 3 commits into from
Jan 15, 2022
Merged

[ONNX] Fix onnx convtranspose error #9938

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
8f4593f
fix mix up of channels with conv2d-transpose
AndrewZhaoLuo Jan 14, 2022
487f035
add grouped convtranspose tests
AndrewZhaoLuo Jan 14, 2022
884b172
turn off groups for non-llvm test
AndrewZhaoLuo Jan 14, 2022
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
18 changes: 9 additions & 9 deletions python/tvm/relay/frontend/onnx.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,9 +38,9 @@
from .. import ty as _ty
from .. import vision as _vision
from .common import (
autopad,
AttrCvt,
Renamer,
autopad,
ensure_scalar_shape,
fold_constant,
get_name,
Expand Down Expand Up @@ -557,13 +557,13 @@ class ConvTranspose(OnnxOpConverter):
def _impl_v1(cls, inputs, attr, params):
# get number of channels
out_type = infer_type(inputs[1])
out_shapes = [get_const_tuple(out_type.checked_type.shape)]
channels = out_shapes[0][1]
attr["channels"] = channels
kernel_shape = [get_const_tuple(out_type.checked_type.shape)]
out_channels = kernel_shape[0][1] * attr.get("group", 1)
attr["channels"] = out_channels
groups = attr.get("group", 1)

if "kernel_shape" not in attr:
attr["kernel_shape"] = out_shapes[0][2:]
attr["kernel_shape"] = kernel_shape[0][2:]

attr["groups"] = groups
# infer pads for auto_pad
Expand Down Expand Up @@ -612,13 +612,13 @@ def _impl_v1(cls, inputs, attr, params):
def _impl_v11(cls, inputs, attr, params):
# get number of channels
out_type = infer_type(inputs[1])
out_shapes = [get_const_tuple(out_type.checked_type.shape)]
channels = out_shapes[0][1]
attr["channels"] = channels
kernel_shape = [get_const_tuple(out_type.checked_type.shape)]
out_channels = kernel_shape[0][1] * attr.get("group", 1)
attr["channels"] = out_channels
groups = attr.get("group", 1)

if "kernel_shape" not in attr:
attr["kernel_shape"] = out_shapes[0][2:]
attr["kernel_shape"] = kernel_shape[0][2:]

attr["groups"] = groups
# infer pads for auto_pad
Expand Down
9 changes: 9 additions & 0 deletions tests/python/frontend/onnx/test_forward.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -189,6 +189,7 @@ def verify_with_ort_with_inputs(
opt_level=opt_level,
convert_config=convert_config,
)

if not isinstance(tvm_out, list):
tvm_out = [tvm_out]
if not isinstance(ort_out, list):
Expand Down Expand Up @@ -2892,6 +2893,14 @@ def verify_convtranspose(x_shape, w_shape, y_shape, p, group=1):
# Test undefined groups.
verify_convtranspose((1, 1, 3, 3), (1, 2, 3, 3), (1, 2, 7, 3), [1, 2, 1, 2], group=None)

if "llvm" in target:
# GPU does not support groups != 1 for convtranspose, so only test llvm
# Test depthwise-convolution
verify_convtranspose((1, 10, 3, 3), (10, 1, 3, 3), (1, 10, 7, 3), [1, 2, 1, 2], group=10)

# Test grouped-convolution
verify_convtranspose((1, 10, 3, 3), (10, 1, 3, 3), (1, 5, 7, 3), [1, 2, 1, 2], group=5)

def repeat(N, D):
return tuple([N for _ in range(D)])

Expand Down