Exported ONNX with Dynamic Shapes¶
The following script shows the exported program for many short cases and various l-plot-export-with-dynamic-shape to retrieve an ONNX model equivalent to the original model.
<<<
import inspect
import textwrap
import pandas
from onnx_diagnostic.helpers import string_type
from onnx_diagnostic.helpers.onnx_helper import pretty_onnx
from onnx_diagnostic.torch_export_patches.eval import discover, run_exporter
from onnx_diagnostic.ext_test_case import unit_test_going
cases = discover()
print()
print(":ref:`Summary <ledx-summary-exported-program>`")
print()
sorted_cases = sorted(cases.items())
if unit_test_going():
sorted_cases = sorted_cases[:3]
for name, cls_model in sorted_cases:
print(f"* :ref:`{name} <ledx-model-case-export-{name}>`")
print()
print()
obs = []
for name, cls_model in sorted(cases.items()):
print()
print(f".. _ledx-model-case-export-{name}:")
print()
print(name)
print("=" * len(name))
print()
print("forward")
print("+++++++")
print()
print(".. code-block:: python")
print()
src = inspect.getsource(cls_model.forward)
if src:
print(textwrap.indent(textwrap.dedent(src), " "))
else:
print(" # code is missing")
print()
print()
for exporter in ("custom", "dynamo-ir"):
expname = exporter.replace("export-", "")
print()
print(expname)
print("+" * len(expname))
print()
res = run_exporter(exporter, cls_model, True, quiet=True)
case_ref = f":ref:`{name} <ledx-model-case-export-{name}>`"
expo = exporter.split("-", maxsplit=1)[-1]
if "inputs" in res:
print(f"* **inputs:** ``{string_type(res['inputs'], with_shape=True)}``")
if "dynamic_shapes" in res:
print(f"* **shapes:** ``{string_type(res['dynamic_shapes'])}``")
print()
print()
if "onx" in res:
print(".. code-block:: text")
print()
print(textwrap.indent(pretty_onnx(res["onx"]), " "))
print()
print()
if "error" not in res:
obs.append(dict(case=case_ref, error="", exporter=expo))
if "error" in res:
print("**FAILED**")
print()
print(".. code-block:: text")
print()
err = str(res["error"])
if err:
print(textwrap.indent(err, " "))
else:
print(" # no error found for the failure")
print()
print()
obs.append(dict(case=case_ref, error="FAIL", exporter=expo))
print()
print(".. _ledx-summary-exported-program:")
print()
print("Summary")
print("+++++++")
print()
df = pandas.DataFrame(obs)
piv = df.pivot(index="case", columns="exporter", values="error")
print(piv.to_markdown(tablefmt="rst"))
print()
>>>
AtenAsStrided¶
forward¶
def forward(self, x):
y = torch.as_strided(x, (2, 2, 8, 4), (128, 8, 16, 1))
return y
custom¶
FAILED
The implementation is still incorrect, x='x', shape=('batch', 2, 8, 8), size=[2, 2, 8, 4], stride=[128, 8, 16, 1], storage_offset=None
--DEBUG--
[GraphBuilder-XRS] Message starts, there are 0 initializers, 0 nodes, 1 inputs, 1 outputs.
--CONSTRAINTS--
batch = {'s77'}
s77 = {'batch'}
--SHAPE--
_dynamic_examples=
dynamic_objects=
batch = 'batch'
s77 = 's77'
dynamic_objects_rev=
'batch' = <class 'list'>
tuple
'batch'
ERR**: <class 'torch.SymInt'>:'batch'
dynamic_dimensions_source={'batch': [{'axis': 0, 'input_name': 'x'}]}
dynamic_dimensions_source_flat=['x']
output_dynamic_dimensions_source_flat=None
dynamic_alias={'s77': 'batch'}
dynamic_shapes={'x': {0: Dim('batch', min=0)}}
_known_shapes={'x': ('batch', 2, 8, 8)}
_known_types={'x': 1}
_known_value_shape={}
_known_constants=[]
_known_ranks={}
--PARAMETERS--
_parameter_renaming=
--TORCH-USERS--
as_strided -> {output}
x -> {as_strided}
--TORCH-SHAPES--
x: ('run_node', ('', ('val', torch.float32, torch.Size([s77, 2, 8, 8])))) --- 1:4:('batch', 2, 8, 8):
as_strided: ('run_node', ('', ('val', torch.float32, torch.Size([2, 2, 8, 4])))) --- :::
--ONNX--
-- EXEPATH --
export-export_options=ExportOptions()
-- process.graph_module --
ExportedProgram:
class GraphModule(torch.nn.Module):
def forward(self, x: "f32[s77, 2, 8, 8]"):
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:186 in forward, code: y = torch.as_strided(x, (2, 2, 8, 4), (128, 8, 16, 1))
as_strided: "f32[2, 2, 8, 4]" = torch.ops.aten.as_strided.default(x, [2, 2, 8, 4], [128, 8, 16, 1]); x = None
return (as_strided,)
Graph signature:
# inputs
x: USER_INPUT
# outputs
as_strided: USER_OUTPUT
Range constraints: {s77: VR[0, int_oo]}
-- process.graph_module.graph --
graph():
%x : [num_users=1] = placeholder[target=x]
%as_strided : [num_users=1] = call_function[target=torch.ops.aten.as_strided.default](args = (%x, [2, 2, 8, 4], [128, 8, 16, 1]), kwargs = {})
return (as_strided,)
-- process.progress --
node 1/3 target=aten.as_strided.default
-- 1 INPUTS
[GraphBuilder-XRS.make_tensor_input] x[1:batchx2x8x8]
-- 0 INITIALIZERS
-- 0 OUTPUTS
[GraphBuilder-XRS] Message completed, there are 0 initializers, 0 nodes, 1 inputs, 1 outputs.
dynamo-ir¶
inputs:
#1[(T1s2x2x8x8,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,2,8,8] x) => (float[2,2,8,4] as_strided)
<int64[4] val_0, int64[4] val_1, int64[1] neg_1, int64[1] rank_tensor, int64 indices, seq(float[]) one_seq, int64 rank_0, float[?] self_flatten>
{
[node_Constant_0] val_0 = Constant <value: tensor = int64[4] {2,2,8,4}> ()
[node_Constant_1] val_1 = Constant <value: tensor = int64[4] {128,8,16,1}> ()
[n0] neg_1 = Constant <value_ints: ints = [-1]> ()
[node_Constant_2] rank_tensor = Constant <value: tensor = int64[1] rank_tensor {4}> ()
[n3] indices = Constant <value_int: int = 0> ()
[n4] one_seq = SequenceEmpty ()
[n5] rank_0 = Constant <value_int: int = 4> ()
[n6_2] one_seq_16, indices_17 = Loop (rank_0, "", one_seq, indices) <body: graph = loop_body (int64 i, bool cond_in, one_seq_1, indices_2) => (bool cond_out, one_seq_15, indices_13)
<int64 rank_3_cast, int64 tmp, int64 int64_1_cast, int64 j, int64[1] j_tensor, int64[1] size_dim_j, int64[?] size_after_j, int64[1] stride_dim_j, int64 int64_0_cast, int64 int64_1_5_cast, int64[?] tmp_6, int64[?] add_value, int64 int64_0_7_cast, bool cond, float[1] tmp_14>
{
[node_Constant_1] rank_3_cast = Constant <value: tensor = int64 rank_3_cast {4}> ()
[n2_2] tmp = Sub (rank_3_cast, i)
[node_Constant_3] int64_1_cast = Constant <value: tensor = int64 int64_1_cast {1}> ()
[n5_2] j = Sub (tmp, int64_1_cast)
[n6] j_tensor = Reshape (j, neg_1)
[n7] size_dim_j = Gather <axis: int = 0> (val_0, j_tensor)
[n8] size_after_j = Slice (val_0, j_tensor, rank_tensor)
[n9] stride_dim_j = Gather <axis: int = 0> (val_1, j_tensor)
[n10] indices_4 = Expand (indices_2, size_after_j)
[node_Constant_5] int64_0_cast = Constant <value: tensor = int64 int64_0_cast {0}> ()
[node_Constant_7] int64_1_5_cast = Constant <value: tensor = int64 int64_1_5_cast {1}> ()
[n15] tmp_6 = Range (int64_0_cast, size_dim_j, int64_1_5_cast)
[n16] add_value = Mul (tmp_6, stride_dim_j)
[node_Constant_9] int64_0_7_cast = Constant <value: tensor = int64 int64_0_7_cast {0}> ()
[n19] cond = Equal (i, int64_0_7_cast)
[n20] shape_11 = If (cond) <then_branch: graph = thenGraph_39 () => (int64[1] shape) {
[n0_3] shape = Identity (size_dim_j)
}, else_branch: graph = elseGraph_39 () => (int64[] shape_10)
<float[1] tmp_8>
{
[n0_4] ones = ConcatFromSequence <axis: int = 0> (one_seq_1)
[n1_3] tmp_8 = Cast <to: int = 1> (size_dim_j)
[n2_3] shape_9 = Concat <axis: int = 0> (tmp_8, ones)
[n3_3] shape_10 = Cast <to: int = 7> (shape_9)
}>
[n21] add_value_12 = Reshape (add_value, shape_11)
[n22] indices_13 = Add (indices_4, add_value_12)
[n23] tmp_14 = Constant <value_floats: floats = [1]> ()
[n24] one_seq_15 = SequenceInsert (one_seq_1, tmp_14)
[n25] cond_out = Identity (cond_in)
}>
[n8_2] self_flatten = Reshape (x, neg_1)
[n10_2] storage_offset_cast = CastLike (indices, indices_17)
[n11_2] indices_19 = Add (indices_17, storage_offset_cast)
[n12_2] as_strided = Gather (self_flatten, indices_19)
}
AtenInterpolate¶
forward¶
def forward(self, x):
y = torch.nn.functional.interpolate(
x,
scale_factor=2.0,
mode="bilinear",
recompute_scale_factor=False,
)
return y
custom¶
inputs:
#1[(T1s2x2x3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "16", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s2x2x3x4[-3.3515872955322266,2.3441219329833984:A-0.05102310386913208],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}", "_known_value_shapes": "{
_onx_concat_x::Shape:2: ('batch', 2, 6, 8),
x::Shape:2: ('batch', 2),
}"]
>
experiment (float[batch,2,3,4] x) => (float[batch,2,6,8] output_0)
<int64[2] init7_s2_6_8 = {6,8}, float[batch,2,6,8] upsample_bilinear2d, int64[2] "x::Shape:2">
{
[upsample_bicubic2d_vec] "x::Shape:2" = Shape <end: int = 2, start: int = 0> (x)
[upsample_bicubic2d_vec2] "_onx_concat_x::Shape:2" = Concat <axis: int = 0> ("x::Shape:2", init7_s2_6_8)
[upsample_bicubic2d_vec3] output_0 = Resize <coordinate_transformation_mode: string = "pytorch_half_pixel", mode: string = "linear", nearest_mode: string = "floor"> (x, "", "", "_onx_concat_x::Shape:2")
}
dynamo-ir¶
inputs:
#1[(T1s2x2x3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,2,3,4] x) => (float[batch,2,6,8] upsample_bilinear2d)
<float[4] val_0>
{
[node_Constant_0] val_0 = Constant <value_floats: floats = [1, 1, 2, 2]> ()
[node_upsample_bilinear2d] upsample_bilinear2d = Resize <keep_aspect_ratio_policy: string = "stretch", antialias: int = 0, extrapolation_value: float = 0, exclude_outside: int = 0, nearest_mode: string = "floor", coordinate_transformation_mode: string = "pytorch_half_pixel", cubic_coeff_a: float = -0.75, mode: string = "linear"> (x, "", val_0)
}
AtenNonZero¶
forward¶
def forward(self, x):
y = torch.nonzero(x)
return y
custom¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "0", "n_large_initializers": "0", "size_initializers": "0", "size_large_initializers": "0", "n_nodes": "2", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4[-2.22353458404541,1.8057029247283936:A0.12694389124711355],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'NEWDIM_nonzero': {'u0'},
'batch': {'s77'},
's77': {'batch'},
'u0': {'NEWDIM_nonzero'}}", "_known_value_shapes": "{
nonzero::Shape:1: ('NEWDIM_nonzero',),
sym_size_int_1: NEWDIM_nonzero,
}"]
>
experiment (float[batch,4] x) => (int64[NEWDIM_nonzero,2] output_0)
<int64 sym_size_int_1, bool sym_constrain_range_for_size_default, int64[2,NEWDIM_nonzero] _onx_nonzero_x, int64[NEWDIM_nonzero,2] nonzero, int64[1] "nonzero::Shape:1">
{
[nonzero] _onx_nonzero_x = NonZero (x)
[nonzero2] output_0 = Transpose <perm: ints = [1, 0]> (_onx_nonzero_x)
}
dynamo-ir¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (int64[u0,2] nonzero)
<int64[2,?] val_0>
{
[node_NonZero_0] val_0 = NonZero (x)
[node_nonzero] nonzero = Transpose <perm: ints = [1, 0]> (val_0)
}
AtenNonZeroTuple¶
forward¶
def forward(self, x):
y = torch.nonzero(x, as_tuple=True)
return y[0], y[1]
custom¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "8", "size_large_initializers": "0", "n_nodes": "4", "n_nodes_other_domain": "0", "mask_outputs": "[True, True]", "input_args": "(T1s3x4[-2.1581575870513916,0.9051743745803833:A-0.11730112507939339],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}", "_known_value_shapes": "{
getitem_2::Shape:1: ('u0',),
sym_size_int_1: u0,
}"]
>
experiment (float[batch,4] x) => (int64[u0] output_0, int64[u0] output_1)
<int64[1] "init7_s1_-1" = {-1}, int64 sym_size_int_1, int64[1] "getitem_2::Shape:1", bool sym_constrain_range_for_size_default, int64[2,NEWDIM_nonzero] _onx_nonzero_x, int64[u0] getitem_1, int64[u0] getitem_2, int64[?,?] _onx_split_nonzero_x_1, int64[u0] "nonzero_numpy#0", int64[?,?] _onx_split_nonzero_x_0, int64[u0] "nonzero_numpy#1">
{
[nonzero_numpy] _onx_nonzero_x = NonZero (x)
[nonzero_numpy2] _onx_split_nonzero_x_0, _onx_split_nonzero_x_1 = Split <num_outputs: int = 2> (_onx_nonzero_x)
[nonzero_numpy3] output_0 = Reshape (_onx_split_nonzero_x_0, "init7_s1_-1")
[nonzero_numpy4] output_1 = Reshape (_onx_split_nonzero_x_1, "init7_s1_-1")
}
dynamo-ir¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (int64[u0] getitem, int64[u0] getitem_1)
<int64[2,?] val_0, int64[u0,2] nonzero, int64[u0,1] unbind_split_0, int64[u0,1] unbind_split_1, int64[1] unbind_axis>
{
[node_NonZero_0] val_0 = NonZero (x)
[node_nonzero] nonzero = Transpose <perm: ints = [1, 0]> (val_0)
[node_Split_4] unbind_split_0, unbind_split_1 = Split <axis: int = 1, num_outputs: int = 2> (nonzero)
[node_Constant_5] unbind_axis = Constant <value_ints: ints = [1]> ()
[node_Squeeze_6] getitem = Squeeze (unbind_split_0, unbind_axis)
[node_Squeeze_7] getitem_1 = Squeeze (unbind_split_1, unbind_axis)
}
AtenRollPos¶
forward¶
def forward(self, x):
return torch.roll(x, 1, -1)
custom¶
inputs:
#1[(T1s2x3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "3", "n_large_initializers": "0", "size_initializers": "24", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s2x3x4[10.0,33.0:A21.5],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3,4] x) => (float[batch,3,4] output_0)
<int64[1] "init7_s1_-1" = {-1}, int64[1] init7_s1_4 = {4}, int64[1] init7_s1_0 = {0}, float[batch,3,4] roll, float[batch,3,4] _onx_concat_slice_x, float[?,?,?] _onx_slice_x2, float[?,?,?] _onx_slice_x>
{
[roll] _onx_slice_x = Slice (x, "init7_s1_-1", init7_s1_4, "init7_s1_-1")
[roll2] _onx_slice_x2 = Slice (x, init7_s1_0, "init7_s1_-1", "init7_s1_-1")
[roll3] output_0 = Concat <axis: int = -1> (_onx_slice_x, _onx_slice_x2)
}
dynamo-ir¶
inputs:
#1[(T1s2x3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3,4] x) => (float[batch,3,4] roll)
<int64[1] neg_1, int64[1] dim_tensor, int64[1] shift_tensor, int64[1] slice_length_3, int64[1] tmp_4, float[batch,3,3] suffix, int64 tmp_5, int64[1] tmp_6, float[?,?,?] prefix>
{
[n0] neg_1 = Constant <value_ints: ints = [-1]> ()
[node_Constant_1] dim_tensor = Constant <value: tensor = int64[1] dim_tensor {-1}> ()
[node_Constant_2] shift_tensor = Constant <value: tensor = int64[1] shift_tensor {1}> ()
[node_Constant_7] slice_length_3 = Constant <value: tensor = int64[1] slice_length_3 {3}> ()
[n8] tmp_4 = Constant <value_ints: ints = [0]> ()
[n9] suffix = Slice (x, tmp_4, slice_length_3, dim_tensor)
[n10] tmp_5 = Size (x)
[n11] tmp_6 = Reshape (tmp_5, neg_1)
[n12] prefix = Slice (x, slice_length_3, tmp_6, dim_tensor)
[n13] roll = Concat <axis: int = -1> (prefix, suffix)
}
AtenRollRelu¶
forward¶
def forward(self, x):
return torch.relu(torch.roll(x, -1, -1))
custom¶
inputs:
#1[(T1s2x3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "4", "n_large_initializers": "0", "size_initializers": "32", "size_large_initializers": "0", "n_nodes": "4", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s2x3x4[10.0,33.0:A21.5],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3,4] x) => (float[batch,3,4] output_0)
<int64[1] init7_s1_1 = {1}, int64[1] init7_s1_4 = {4}, int64[1] "init7_s1_-1" = {-1}, int64[1] init7_s1_0 = {0}, float[batch,3,4] roll, float[batch,3,4] _onx_concat_slice_x, float[batch,3,4] relu, float[?,?,?] _onx_slice_x2, float[?,?,?] _onx_slice_x>
{
[roll] _onx_slice_x = Slice (x, init7_s1_1, init7_s1_4, "init7_s1_-1")
[roll2] _onx_slice_x2 = Slice (x, init7_s1_0, init7_s1_1, "init7_s1_-1")
[roll3] _onx_concat_slice_x = Concat <axis: int = -1> (_onx_slice_x, _onx_slice_x2)
[relu] output_0 = Relu (_onx_concat_slice_x)
}
dynamo-ir¶
inputs:
#1[(T1s2x3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3,4] x) => (float[batch,3,4] relu)
<int64[1] neg_1, int64[1] dim_tensor, int64[1] slice_length_3, int64[1] tmp_4, float[batch,3,1] suffix, int64 tmp_5, int64[1] tmp_6, float[?,?,?] prefix, float[batch,3,4] roll>
{
[n0] neg_1 = Constant <value_ints: ints = [-1]> ()
[node_Constant_1] dim_tensor = Constant <value: tensor = int64[1] dim_tensor {-1}> ()
[node_Constant_6] slice_length_3 = Constant <value: tensor = int64[1] slice_length_3 {1}> ()
[n8] tmp_4 = Constant <value_ints: ints = [0]> ()
[n9] suffix = Slice (x, tmp_4, slice_length_3, dim_tensor)
[n10] tmp_5 = Size (x)
[n11] tmp_6 = Reshape (tmp_5, neg_1)
[n12] prefix = Slice (x, slice_length_3, tmp_6, dim_tensor)
[n13] roll = Concat <axis: int = -1> (prefix, suffix)
[node_relu] relu = Relu (roll)
}
BuildInIsInstance¶
forward¶
def forward(self, x, lx: list | torch.Tensor):
if isinstance(lx, list):
t = lx[0] * lx[1].sum(axis=1, keepdim=True)
return torch.sigmoid(self.linear(x)) - self.buff + t
return torch.sigmoid(self.linear(x)) - self.buff + lx
custom¶
inputs:
#2[(T1s4x3,#2[T1s4x1,T1s4x2]),(T1s8x3,#2[T1s8x1,T1s8x2])]shapes:
dict(x:{0:Dim(batch)},lx:#2[{0:Dim(batch)},{0:Dim(batch)}])
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "4", "n_large_initializers": "0", "size_initializers": "28", "size_large_initializers": "0", "n_nodes": "6", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x3[10.0,21.0:A15.5],#2[T1s4x1[10.0,13.0:A11.5],T1s4x2[10.0,17.0:A13.5]])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'lx': [{0: Dim('batch', min=0)}, {0: Dim('batch', min=0)}],
'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s53', 's50', 's77'},
's50': {'batch', 's53'},
's53': {'batch'},
's77': {'batch'}}"]
>
experiment (float[batch,3] x, float[batch,1] lx_0, float[batch,2] lx_1) => (float[batch,1] output_0)
<float[1] b_buff = {0.5}, int64[1] init7_s1_1 = {1}, float[1,3] "GemmTransposePattern--p_linear_weight::T10" = {0.126205,0.342849,0.132504}, float[1] "linear.bias" = {0.187118}, float[batch,1] _onx_matmul_x, float[batch,1] add, float[batch,1] mul, float[batch,1] sum_1, float[1,3] p_linear_weight, float[batch,1] linear, int64[2] "init7_s2_1_-1", float[batch,1] sigmoid, float[batch,1] sub, float[1,3] "linear.weight", float[1] p_linear_bias, float[3,1] "p_linear_weight::T10">
{
[sum] sum_1 = ReduceSum <keepdims: int = 1> (lx_1, init7_s1_1)
[mul_Tensor] mul = Mul (lx_0, sum_1)
["GemmTransposePattern--MatMulAddPattern--Opset2"] linear = Gemm <transB: int = 1> (x, "GemmTransposePattern--p_linear_weight::T10", "linear.bias")
[sigmoid] sigmoid = Sigmoid (linear)
[sub_Tensor] sub = Sub (sigmoid, b_buff)
[add_Tensor] output_0 = Add (sub, mul)
}
dynamo-ir¶
inputs:
#2[(T1s4x3,#2[T1s4x1,T1s4x2]),(T1s8x3,#2[T1s8x1,T1s8x2])]shapes:
dict(x:{0:Dim(batch)},lx:#2[{0:Dim(batch)},{0:Dim(batch)}])
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x, float[batch,1] lx_0, float[batch,2] lx_1) => (float[batch,1] add_15)
<float[1,3] "linear.weight" = {0.518632,-0.238197,-0.215139}, float[1] "linear.bias" = {-0.154246}, float[1] buff = {0.5}, float[1,3] "linear.weight", float[1] "linear.bias", float[1] buff, int64[1] val_6, float[batch,1] sum_1, float[batch,1] mul_1, float[batch,1] linear, float[batch,1] sigmoid, float[batch,1] sub_4>
{
[node_Constant_9] val_6 = Constant <value: tensor = int64[1] val_6 {1}> ()
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 1> (lx_1, val_6)
[node_mul_1] mul_1 = Mul (lx_0, sum_1)
[node_linear] linear = Gemm <beta: float = 1, transB: int = 1, alpha: float = 1, transA: int = 0> (x, "linear.weight", "linear.bias")
[node_sigmoid] sigmoid = Sigmoid (linear)
[node_sub_4] sub_4 = Sub (sigmoid, buff)
[node_add_15] add_15 = Add (sub_4, mul_1)
}
BuildInLen¶
forward¶
def forward(self, x, lx: list):
t = lx[0] * lx[1].sum(axis=1, keepdim=True)
if len(lx) > 2:
t = t + lx[2].sum(axis=1, keepdim=True)
return torch.sigmoid(self.linear(x)) - self.buff + t
custom¶
inputs:
#2[(T1s4x3,#2[T1s4x1,T1s4x2]),(T1s8x3,#3[T1s8x1,T1s8x2,T1s8x3])]shapes:
dict(x:{0:Dim(batch)},lx:#2[{0:Dim(batch)},{0:Dim(batch)}])
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "4", "n_large_initializers": "0", "size_initializers": "28", "size_large_initializers": "0", "n_nodes": "6", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x3[10.0,21.0:A15.5],#2[T1s4x1[10.0,13.0:A11.5],T1s4x2[10.0,17.0:A13.5]])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'lx': [{0: Dim('batch', min=0)}, {0: Dim('batch', min=0)}],
'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s53', 's50', 's77'},
's50': {'batch', 's53'},
's53': {'batch'},
's77': {'batch'}}"]
>
experiment (float[batch,3] x, float[batch,1] lx_0, float[batch,2] lx_1) => (float[batch,1] output_0)
<float[1] b_buff = {0.5}, int64[1] init7_s1_1 = {1}, float[1,3] "GemmTransposePattern--p_linear_weight::T10" = {-0.0998398,0.350802,0.130283}, float[1] "linear.bias" = {-0.44322}, float[batch,1] _onx_matmul_x, float[batch,1] add, float[batch,1] mul, float[batch,1] sum_1, float[1,3] p_linear_weight, float[batch,1] linear, int64[2] "init7_s2_1_-1", float[batch,1] sigmoid, float[batch,1] sub, float[1,3] "linear.weight", float[1] p_linear_bias, float[3,1] "p_linear_weight::T10">
{
[sum] sum_1 = ReduceSum <keepdims: int = 1> (lx_1, init7_s1_1)
[mul_Tensor] mul = Mul (lx_0, sum_1)
["GemmTransposePattern--MatMulAddPattern--Opset2"] linear = Gemm <transB: int = 1> (x, "GemmTransposePattern--p_linear_weight::T10", "linear.bias")
[sigmoid] sigmoid = Sigmoid (linear)
[sub_Tensor] sub = Sub (sigmoid, b_buff)
[add_Tensor] output_0 = Add (sub, mul)
}
FAILED
diff.1
dynamo-ir¶
inputs:
#2[(T1s4x3,#2[T1s4x1,T1s4x2]),(T1s8x3,#3[T1s8x1,T1s8x2,T1s8x3])]shapes:
dict(x:{0:Dim(batch)},lx:#2[{0:Dim(batch)},{0:Dim(batch)}])
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x, float[batch,1] lx_0, float[batch,2] lx_1) => (float[batch,1] add_15)
<float[1,3] "linear.weight" = {0.0222517,0.573125,0.478325}, float[1] "linear.bias" = {-0.12366}, float[1] buff = {0.5}, float[1,3] "linear.weight", float[1] "linear.bias", float[1] buff, int64[1] val_6, float[batch,1] sum_1, float[batch,1] mul_1, float[batch,1] linear, float[batch,1] sigmoid, float[batch,1] sub_4>
{
[node_Constant_9] val_6 = Constant <value: tensor = int64[1] val_6 {1}> ()
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 1> (lx_1, val_6)
[node_mul_1] mul_1 = Mul (lx_0, sum_1)
[node_linear] linear = Gemm <beta: float = 1, transB: int = 1, alpha: float = 1, transA: int = 0> (x, "linear.weight", "linear.bias")
[node_sigmoid] sigmoid = Sigmoid (linear)
[node_sub_4] sub_4 = Sub (sigmoid, buff)
[node_add_15] add_15 = Add (sub_4, mul_1)
}
FAILED
diff.1
ComplexPolar¶
forward¶
def forward(self, x, angle):
return torch.polar(x, angle)
custom¶
inputs:
#1[(T1s4x4,T1s4x4)]shapes:
dict(x:{0:Dim(batch)},angle:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "8", "size_large_initializers": "0", "n_nodes": "8", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x4[0.010826051235198975,0.9339432716369629:A0.4240015111863613],T1s4x4[0.07700252532958984,0.9605667591094971:A0.4933854453265667])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'angle': {0: Dim('batch', min=0)}, 'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77', 's96'}, 's77': {'batch'}, 's96': {'batch'}}"]
>
experiment (float[batch,4] x, float[batch,4] angle) => (complex64[batch,4] output_0)
<complex64[1] init14_s1_ = ..., complex64[batch,4] "x::C14", float[batch,4] _onx_cos_angle, complex64[batch,4] "_onx_sin_angle::C14", complex64[batch,4] "_onx_mul_sin_angle::C14", complex64[batch,4] "_onx_add_cos_angle::C14", complex64[batch,4] polar, complex64[batch,4] "_onx_cos_angle::C14", float[batch,4] _onx_sin_angle>
{
[polar] _onx_cos_angle = Cos (angle)
[polar2] "_onx_cos_angle::C14" = Cast <to: int = 14> (_onx_cos_angle)
[polar3] _onx_sin_angle = Sin (angle)
[polar4] "_onx_sin_angle::C14" = Cast <to: int = 14> (_onx_sin_angle)
[polar5] "_onx_mul_sin_angle::C14" = Mul ("_onx_sin_angle::C14", init14_s1_)
[polar6] "x::C14" = Cast <to: int = 14> (x)
[polar7] "_onx_add_cos_angle::C14" = Add ("_onx_cos_angle::C14", "_onx_mul_sin_angle::C14")
[polar8] output_0 = Mul ("x::C14", "_onx_add_cos_angle::C14")
}
FAILED
[ONNXRuntimeError] : 10 : INVALID_GRAPH : This is an invalid model. Type Error: Type 'tensor(complex64)' of input parameter (_onx_sin_angle::C14) of operator (Mul) in node (polar5) is invalid.
dynamo-ir¶
inputs:
#1[(T1s4x4,T1s4x4)]shapes:
dict(x:{0:Dim(batch)},angle:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x, float[batch,4] angle) => (float[batch,4,2] polar)
<float[batch,4] tmp, float[batch,4] tmp_0, int64[1] int64_m1_1d, float[batch,4,1] real, float[batch,4] tmp_1, float[batch,4] tmp_2, float[batch,4,1] imag>
{
[n0] tmp = Cos (angle)
[n1] tmp_0 = Mul (x, tmp)
[n2] int64_m1_1d = Constant <value: tensor = int64[1] int64_m1_1d {-1}> ()
[n3] real = Unsqueeze (tmp_0, int64_m1_1d)
[n4] tmp_1 = Sin (angle)
[n5] tmp_2 = Mul (x, tmp_1)
[n7] imag = Unsqueeze (tmp_2, int64_m1_1d)
[n8] polar = Concat <axis: int = -1> (real, imag)
}
FAILED
diff.0
ControlFlowCond¶
forward¶
def forward(self, x):
def true_fn(x):
return torch.sin(x)
def false_fn(x):
return torch.cos(x)
return torch.cond(x.sum() > 0, true_fn, false_fn, [x])
custom¶
inputs:
#1[(T1s5x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "4", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s5x3[0.130266010761261,0.9946584105491638:A0.528652552763621],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3] x) => (float[batch,3] output_0)
<float init1_s_ = {0}, float sum_1, bool gt, float[1] "sum_1::RSh1", float[1] "init1_s_::RSh1", int64[1] init7_s1_1, float[batch,3] getitem>
{
[sum] sum_1 = ReduceSum <keepdims: int = 0> (x)
[gt_Scalar3] gt = Greater (sum_1, init1_s_)
[cond] output_0 = If (gt) <else_branch: graph = experiment () => ( "cond#0") {
[cos2] "cond#0" = Cos (x)
}, then_branch: graph = experiment () => ( "cond#0") {
[sin2] "cond#0" = Sin (x)
}>
}
dynamo-ir¶
inputs:
#1[(T1s5x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x) => (float[batch,3] getitem)
<float sum_1, float scalar_tensor_default, bool gt>
{
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (x)
[node_Constant_1] scalar_tensor_default = Constant <value: tensor = float scalar_tensor_default {0}> ()
[node_gt] gt = Greater (sum_1, scalar_tensor_default)
[node_cond__0] getitem = If (gt) <then_branch: graph = true_graph_0 () => (float[batch,3] sin_true_graph_0) {
[node_sin] sin_true_graph_0 = Sin (x)
}, else_branch: graph = false_graph_0 () => (float[batch,3] cos_false_graph_0) {
[node_cos] cos_false_graph_0 = Cos (x)
}>
}
ControlFlowCond2Inputs¶
forward¶
def forward(self, x, y):
def true_fn(x, y):
return torch.sin(x), torch.cos(x) + y
def false_fn(x, y):
return torch.cos(x), torch.sin(x) + y
return torch.cond(x.sum() > 0, true_fn, false_fn, [x, y])
custom¶
inputs:
#1[(T1s5x3,T1s5x3)]shapes:
dict(x:{0:Dim(batch)},y:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "4", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True, True]", "input_args": "(T1s5x3[0.05943065881729126,0.8915067911148071:A0.5680543144543966],T1s5x3[0.016381025314331055,0.933613657951355:A0.4243027885754903])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}, 'y': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s17', 's77'}, 's17': {'batch'}, 's77': {'batch'}}"]
>
experiment (float[batch,3] x, float[batch,3] y) => (float[batch,3] output_0, float[batch,3] output_1)
<float init1_s_ = {0}, float sum_1, bool gt, float[1] "sum_1::RSh1", float[1] "init1_s_::RSh1", float[batch,3] getitem_1, int64[1] init7_s1_1, float[batch,3] getitem>
{
[sum] sum_1 = ReduceSum <keepdims: int = 0> (x)
[gt_Scalar3] gt = Greater (sum_1, init1_s_)
[cond] output_0, output_1 = If (gt) <else_branch: graph = experiment () => ( "cond#0", "cond#1") {
[cos2] "cond#0" = Cos (x)
[sin2] sin2 = Sin (x)
[add_Tensor2] "cond#1" = Add (sin2, y)
}, then_branch: graph = experiment () => ( "cond#0", "cond#1") {
[sin32] "cond#0" = Sin (x)
[cos32] cos2 = Cos (x)
[add_Tensor32] "cond#1" = Add (cos2, y)
}>
}
dynamo-ir¶
inputs:
#1[(T1s5x3,T1s5x3)]shapes:
dict(x:{0:Dim(batch)},y:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x, float[batch,3] y) => (float[batch,3] getitem, float[batch,3] getitem_1)
<float sum_1, float scalar_tensor_default, bool gt>
{
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (x)
[node_Constant_1] scalar_tensor_default = Constant <value: tensor = float scalar_tensor_default {0}> ()
[node_gt] gt = Greater (sum_1, scalar_tensor_default)
[node_cond__1] getitem, getitem_1 = If (gt) <then_branch: graph = true_graph_0 () => (float[batch,3] sin_true_graph_0, float[batch,3] add_12_true_graph_0)
<float[batch,3] cos>
{
[node_sin] sin_true_graph_0 = Sin (x)
[node_cos] cos = Cos (x)
[node_add_12] add_12_true_graph_0 = Add (cos, y)
}, else_branch: graph = false_graph_0 () => (float[batch,3] cos_false_graph_0, float[batch,3] add_12_false_graph_0)
<float[batch,3] sin_2>
{
[node_cos_2] cos_false_graph_0 = Cos (x)
[node_sin_2] sin_2 = Sin (x)
[node_add_12_2] add_12_false_graph_0 = Add (sin_2, y)
}>
}
ControlFlowCond2Outputs¶
forward¶
def forward(self, x):
def true_fn(x):
return torch.sin(x), torch.cos(x)
def false_fn(x):
return torch.cos(x), torch.sin(x)
return torch.cond(x.sum() > 0, true_fn, false_fn, [x])
custom¶
inputs:
#1[(T1s5x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "4", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True, True]", "input_args": "(T1s5x3[0.003947198390960693,0.8443840146064758:A0.48277585903803505],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3] x) => (float[batch,3] output_0, float[batch,3] output_1)
<float init1_s_ = {0}, float sum_1, bool gt, float[1] "sum_1::RSh1", float[1] "init1_s_::RSh1", float[batch,3] getitem_1, int64[1] init7_s1_1, float[batch,3] getitem>
{
[sum] sum_1 = ReduceSum <keepdims: int = 0> (x)
[gt_Scalar3] gt = Greater (sum_1, init1_s_)
[cond] output_0, output_1 = If (gt) <else_branch: graph = experiment () => ( "cond#0", "cond#1") {
[cos2] "cond#0" = Cos (x)
[sin2] "cond#1" = Sin (x)
}, then_branch: graph = experiment () => ( "cond#0", "cond#1") {
[sin32] "cond#0" = Sin (x)
[cos32] "cond#1" = Cos (x)
}>
}
dynamo-ir¶
inputs:
#1[(T1s5x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x) => (float[batch,3] getitem, float[batch,3] getitem_1)
<float sum_1, float scalar_tensor_default, bool gt>
{
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (x)
[node_Constant_1] scalar_tensor_default = Constant <value: tensor = float scalar_tensor_default {0}> ()
[node_gt] gt = Greater (sum_1, scalar_tensor_default)
[node_cond__1] getitem, getitem_1 = If (gt) <then_branch: graph = true_graph_0 () => (float[batch,3] sin_true_graph_0, float[batch,3] cos_true_graph_0) {
[node_sin] sin_true_graph_0 = Sin (x)
[node_cos] cos_true_graph_0 = Cos (x)
}, else_branch: graph = false_graph_0 () => (float[batch,3] cos_false_graph_0, float[batch,3] sin_false_graph_0) {
[node_cos_2] cos_false_graph_0 = Cos (x)
[node_sin_2] sin_false_graph_0 = Sin (x)
}>
}
ControlFlowCondConstant¶
forward¶
def forward(self, x):
def true_fn(x):
return torch.sin(x) - torch.ones(x.shape, dtype=x.dtype)
def false_fn(x):
return torch.cos(x) + torch.ones((1, 1024), dtype=x.dtype)
return torch.cond(x.sum() > 0, true_fn, false_fn, [x])
custom¶
inputs:
#1[(T1s1024x1024,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "4", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s1024x1024[1.7881393432617188e-07,0.9999999403953552:A0.4999965746537782],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,1024] x) => (float[batch,1024] output_0)
<float init1_s_ = {0}, float sum_1, bool gt, float[1] "sum_1::RSh1", float[1] "init1_s_::RSh1", int64[1] init7_s1_1, float[batch,1024] getitem>
{
[sum] sum_1 = ReduceSum <keepdims: int = 0> (x)
[gt_Scalar3] gt = Greater (sum_1, init1_s_)
[cond] output_0 = If (gt) <else_branch: graph = experiment () => ( "cond#0") {
[init2cst2] init7_s2_1_10242 = Constant <value: tensor = int64[2] init7_s2_1_1024 {1,1024}> ()
[cos2] cos2 = Cos (x)
[ones2] ones2 = ConstantOfShape <value: tensor = float[1] {1}> (init7_s2_1_10242)
[add_Tensor2] "cond#0" = Add (cos2, ones2)
}, then_branch: graph = experiment () => ( "cond#0") {
[init2cst32] init7_s1_10242 = Constant <value: tensor = int64[1] init7_s1_1024 {1024}> ()
[sin2] sin2 = Sin (x)
[sym_size_int2] "x::Shape:12" = Shape <end: int = 1, start: int = 0> (x)
[_mkshape_sym_size_int2] "_onx_concat_sym_size_int::UnSq02" = Concat <axis: int = 0> ("x::Shape:12", init7_s1_10242)
[ones32] ones32 = ConstantOfShape <value: tensor = float[1] {1}> ("_onx_concat_sym_size_int::UnSq02")
[sub_Tensor2] "cond#0" = Sub (sin2, ones32)
}>
}
dynamo-ir¶
inputs:
#1[(T1s1024x1024,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,1024] x) => (float[batch,1024] getitem)
<float[1,1024] ones_2 = {1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1}, float[1,1024] ones_2, float sum_1, float scalar_tensor_default, bool gt>
{
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (x)
[node_Constant_1] scalar_tensor_default = Constant <value: tensor = float scalar_tensor_default {0}> ()
[node_gt] gt = Greater (sum_1, scalar_tensor_default)
[node_cond__0] getitem = If (gt) <then_branch: graph = true_graph_0 () => (float[?,1024] sub_3_true_graph_0)
<int64[1] val_0_2, int64 sym_size_int, float[batch,1024] sin, int64[1] val_1, int64[1] val_2, int64[1] val_3, int64[2] val_4, float val_7, float[?,?] ones>
{
[node_Shape_0] val_0_2 = Shape <end: int = 1, start: int = 0> (x)
[node_sym_size_int] sym_size_int = Squeeze (val_0_2)
[node_sin] sin = Sin (x)
[node_Constant_1] val_1 = Constant <value: tensor = int64[1] {-1}> ()
[node_Reshape_2] val_2 = Reshape <allowzero: int = 0> (sym_size_int, val_1)
[node_Constant_3] val_3 = Constant <value: tensor = int64[1] {1024}> ()
[node_Concat_4] val_4 = Concat <axis: int = 0> (val_2, val_3)
[node_Constant_4] val_7 = Constant <value: tensor = float val_7 {1}> ()
[node_ones] ones = Expand (val_7, val_4)
[node_sub_3] sub_3_true_graph_0 = Sub (sin, ones)
}, else_branch: graph = false_graph_0 () => (float[batch,1024] add_6_false_graph_0)
<float[batch,1024] cos>
{
[node_cos] cos = Cos (x)
[node_add_6] add_6_false_graph_0 = Add (cos, ones_2)
}>
}
ControlFlowCondIdentity_153832¶
forward¶
def forward(self, x, y):
def branch_cond_then_1(x):
x = torch.abs(x) + 1
return x
def branch_cond_else_1(x):
return x # fails but succeeds with x.clone()
x = torch.cond(x.sum() > 0, branch_cond_then_1, branch_cond_else_1, [x])
return x + y
custom¶
FAILED
Cond doesn't work unless it is captured completely with torch.compile. Scroll up to find out what causes the graph break.
from user code:
File "~/vv/this312/lib/python3.12/site-packages/torch/_higher_order_ops/cond.py", line 187, in _cond_op_wrapper
return cond_op(*args, **kwargs)
Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"
dynamo-ir¶
FAILED
Failed to export the model with torch.export. [96mThis is step 1/3[0m of exporting the model to ONNX. Next steps:
- Modify the model code for `torch.export.export` to succeed. Refer to https://pytorch.org/docs/stable/generated/exportdb/index.html for more information.
- Debug `torch.export.export` and summit a PR to PyTorch.
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
## Exception summary
<class 'torch._dynamo.exc.Unsupported'>: Encountered aliasing during higher order op tracing
Explanation: Higher order ops do not support aliasing. Found in cond
Hint: Replace `return input` with `return input.clone()` to avoid aliasing.
Hint: Consider using the debug context to change user code to avoid aliasing.
Hint: Please open an issue.
Developer debug context: Input-to-output aliasing detected at nodes l_args_3_0_ and l_args_3_0_ in
graph():
%l_args_3_0_ : torch._subclasses.fake_tensor.FakeTensor [num_users=1] = placeholder[target=l_args_3_0_]
return (l_args_3_0_,)
For more details about this graph break, please visit: https://compile-graph-break-site.vercel.app/gb/GB0040
⬆️
<class 'torch._dynamo.exc.UncapturedHigherOrderOpError'>: Cond doesn't work unless it is captured completely with torch.compile. Scroll up to find out what causes the graph break.
from user code:
File "~/vv/this312/lib/python3.12/site-packages/torch/_higher_order_ops/cond.py", line 187, in _cond_op_wrapper
return cond_op(*args, **kwargs)
Set TORCHDYNAMO_VERBOSE=1 for the internal stack trace (please do this especially if you're reporting a bug to PyTorch). For even more developer context, set TORCH_LOGS="+dynamo"
(Refer to the full stack trace above for more information.)
ControlFlowCondNestedModule¶
forward¶
def forward(self, x):
def true_fn(x):
return self.submodule(x)
def false_fn(x):
return x - self.weight
y = torch.cond(x.sum() > 0, true_fn, false_fn, [x])
return y
custom¶
inputs:
#1[(T7s2,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions.0" : 1, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "3", "n_large_initializers": "0", "size_initializers": "16", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T7s2[-1,2:A0.5],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (int64[batch] x) => (float[d_output_0_0] output_0)
<int64 init7_s_0 = {0}, float[1] weight = {42}, float[1] "submodule.weight" = {100}, int64 sum_1, bool gt, int64[1] "sum_1::RSh1", int64[1] "init7_s_0::RSh1", int64[1] init7_s1_1, float[?] getitem, float[1] p_weight, float[1] p_submodule_weight>
{
[sum] sum_1 = ReduceSum <keepdims: int = 0> (x)
[gt_Scalar3] gt = Greater (sum_1, init7_s_0)
[cond] output_0 = If (gt) <else_branch: graph = experiment () => ( "cond#0") {
[Opset2] "x::C12" = Cast <to: int = 1> (x)
[sub_Tensor2] "cond#0" = Sub ("x::C12", weight)
}, then_branch: graph = experiment () => ( "cond#0") {
[init2cst2] init7_s_1002 = Constant <value: tensor = int64 init7_s_100 {100}> ()
[abs2] abs_12 = Abs (x)
[sum22] sum_122 = ReduceSum <keepdims: int = 0> (abs_12)
[gt_Scalar322] gt22 = Greater (sum_122, init7_s_1002)
[cond22] "cond#0" = If (gt22) <else_branch: graph = experiment () => ( "cond#0") {
[Opset32] "x::C132" = Cast <to: int = 1> (x)
[div_Tensor2] "cond#0" = Div ("x::C132", "submodule.weight")
}, then_branch: graph = experiment () => ( "cond#0") {
[mul_Tensor3] "x::C142" = Cast <to: int = 1> (x)
[mul_Tensor22] "cond#0" = Mul ("x::C142", "submodule.weight")
}>
}>
}
dynamo-ir¶
inputs:
#1[(T7s2,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (int64[batch] x) => (float[batch] getitem)
<float[1] weight = {42}, float[1] "submodule.weight" = {100}, float[1] weight, float[1] "submodule.weight", int64 sum_1, int64 val_0, bool gt>
{
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (x)
[node_Constant_0] val_0 = Constant <value: tensor = int64 {0}> ()
[node_gt] gt = Greater (sum_1, val_0)
[node_cond__0] getitem = If (gt) <then_branch: graph = true_graph_0 () => ( getitem_true_graph_0)
<int64[batch] abs_1, int64 sum_1_2, int64 val_0_2, bool gt_2>
{
[node_abs_1] abs_1 = Abs (x)
[node_sum_1_2] sum_1_2 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (abs_1)
[node_Constant_0_2] val_0_2 = Constant <value: tensor = int64 {100}> ()
[node_gt_2] gt_2 = Greater (sum_1_2, val_0_2)
[node_cond__0_2] getitem_true_graph_0 = If (gt_2) <then_branch: graph = true_graph_0__true_graph_0 () => (float[batch] mul_1_true_graph_0__true_graph_0)
<float[batch] convert_element_type_default>
{
[node_convert_element_type_default] convert_element_type_default = Cast <to: int = 1> (x)
[node_mul_1] mul_1_true_graph_0__true_graph_0 = Mul (convert_element_type_default, "submodule.weight")
}, else_branch: graph = true_graph_0__false_graph_0 () => (float[batch] div_true_graph_0__false_graph_0)
<float[batch] convert_element_type_default_2>
{
[node_convert_element_type_default_2] convert_element_type_default_2 = Cast <to: int = 1> (x)
[node_div] div_true_graph_0__false_graph_0 = Div (convert_element_type_default_2, "submodule.weight")
}>
}, else_branch: graph = false_graph_0 () => (float[batch] sub_1_false_graph_0)
<float[batch] convert_element_type_default_3>
{
[node_convert_element_type_default_3] convert_element_type_default_3 = Cast <to: int = 1> (x)
[node_sub_1] sub_1_false_graph_0 = Sub (convert_element_type_default_3, weight)
}>
}
ControlFlowCondNonZero¶
forward¶
def forward(self, input_ids, image_features, vocab_size):
def then_branch(input_ids, image_features, vocab_size):
input_shape = input_ids.size()
input_ids = input_ids.view(-1, input_shape[-1])
condition = (input_ids < 0) & (input_ids > -int(1e9))
positions = torch.nonzero(condition, as_tuple=True)
input_ids = input_ids.clamp_min(0).clamp_max(vocab_size)
return (input_ids, positions[0], positions[1])
def else_branch(input_ids, image_features, vocab_size):
r = torch.where(torch.zeros((1, 1), dtype=torch.bool))
return (input_ids, r[0], r[1])
a, b, c = torch.cond(
image_features.numel() > 0,
then_branch,
else_branch,
[input_ids, image_features, vocab_size],
)
return a, b, c
custom¶
FAILED
Expect operands to be a tuple of possibly nested dict/list/tuple that only consists of tensor leaves, but got [FakeTensor(..., size=(s72, 12), dtype=torch.int64), FakeTensor(..., size=(s28, s11)), 1025].
dynamo-ir¶
FAILED
Failed to export the model with torch.export. [96mThis is step 1/3[0m of exporting the model to ONNX. Next steps:
- Modify the model code for `torch.export.export` to succeed. Refer to https://pytorch.org/docs/stable/generated/exportdb/index.html for more information.
- Debug `torch.export.export` and summit a PR to PyTorch.
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
## Exception summary
<class 'RuntimeError'>: Expect operands to be a tuple of possibly nested dict/list/tuple that only consists of tensor leaves, but got [FakeTensor(..., size=(s72, 12), dtype=torch.int64), FakeTensor(..., size=(s28, s11)), 1025].
(Refer to the full stack trace above for more information.)
ControlFlowNestCond¶
forward¶
def forward(self, x):
def true_fn2(x):
def true_fn1(x):
return torch.sin(x)
def false_fn1(x):
return torch.cos(x)
return torch.cond(x.sum() < 0, true_fn1, false_fn1, [x])
def false_fn2(x):
return -x
return torch.cond(x.sum() > 0, true_fn2, false_fn2, [x])
custom¶
inputs:
#1[(T1s5x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions.0" : 1, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "4", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s5x3[0.2221928834915161,0.9708185791969299:A0.5694047530492147],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3] x) => (float[batch,3] output_0)
<float init1_s_ = {0}, float sum_1, bool gt, float[1] "sum_1::RSh1", float[1] "init1_s_::RSh1", int64[1] init7_s1_1, float[batch,3] getitem>
{
[sum] sum_1 = ReduceSum <keepdims: int = 0> (x)
[gt_Scalar3] gt = Greater (sum_1, init1_s_)
[cond] output_0 = If (gt) <else_branch: graph = experiment () => ( "cond#0") {
[neg2] "cond#0" = Neg (x)
}, then_branch: graph = experiment () => ( "cond#0") {
[init2cst2] init1_s_22 = Constant <value: tensor = float init1_s_ {0}> ()
[sum22] sum_122 = ReduceSum <keepdims: int = 0> (x)
[lt_Scalar32] lt2 = Less (sum_122, init1_s_22)
[cond22] "cond#0" = If (lt2) <else_branch: graph = experiment () => ( "cond#0") {
[cos2] "cond#0" = Cos (x)
}, then_branch: graph = experiment () => ( "cond#0") {
[sin2] "cond#0" = Sin (x)
}>
}>
}
dynamo-ir¶
inputs:
#1[(T1s5x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x) => (float[batch,3] getitem)
<float sum_1, float scalar_tensor_default, bool gt>
{
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (x)
[node_Constant_1] scalar_tensor_default = Constant <value: tensor = float scalar_tensor_default {0}> ()
[node_gt] gt = Greater (sum_1, scalar_tensor_default)
[node_cond__0] getitem = If (gt) <then_branch: graph = true_graph_0 () => ( getitem_true_graph_0)
<float sum_1_2, float scalar_tensor_default_2, bool lt>
{
[node_sum_1_2] sum_1_2 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (x)
[node_Constant_1] scalar_tensor_default_2 = Constant <value: tensor = float scalar_tensor_default_2 {0}> ()
[node_lt] lt = Less (sum_1_2, scalar_tensor_default_2)
[node_cond__0_2] getitem_true_graph_0 = If (lt) <then_branch: graph = true_graph_0__true_graph_0 () => (float[batch,3] sin_true_graph_0__true_graph_0) {
[node_sin] sin_true_graph_0__true_graph_0 = Sin (x)
}, else_branch: graph = true_graph_0__false_graph_0 () => (float[batch,3] cos_true_graph_0__false_graph_0) {
[node_cos] cos_true_graph_0__false_graph_0 = Cos (x)
}>
}, else_branch: graph = false_graph_0 () => (float[batch,3] neg_false_graph_0) {
[node_neg] neg_false_graph_0 = Neg (x)
}>
}
ControlFlowScan¶
forward¶
def forward(self, x):
init = torch.zeros_like(x[0])
carry, out = torch.ops.higher_order.scan(
ControlFlowScan.add, [init], [x], additional_inputs=[]
)
return carry
custom¶
inputs:
#1[(T1s3x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "8", "size_large_initializers": "0", "n_nodes": "2", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x3[1.0,9.0:A5.0],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3] x) => (float[3] output_0)
<int64[1] init7_s1_3 = {3}, float[3] zeros_like, float[3] "scan#0", float[batch,3] getitem_1, float[batch,3] "scan#1", int64 init7_s_0, float[3] select, float[3] getitem>
{
[zeros_likeD2] zeros_like = ConstantOfShape <value: tensor = float[1] {0}> (init7_s1_3)
[scan] output_0, "scan#1" = Scan (zeros_like, x) <body: graph = experiment ( init_0_zeros_like, scan_0_x) => ( output_0, output_1) {
[add_Tensor2] output_0 = Add (init_0_zeros_like, scan_0_x)
[".output22"] output_1 = Identity (output_0)
}, num_scan_inputs: int = 1, scan_input_directions: ints = [0], scan_output_axes: ints = [0], scan_output_directions: ints = [0]>
}
dynamo-ir¶
FAILED
Failed to decompose the FX graph for ONNX compatibility. [96mThis is step 2/3[0m of exporting the model to ONNX. Next steps:
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
- Create an error report with `torch.onnx.export(..., report=True)`, and save the ExportedProgram as a pt2 file. Create an issue in the PyTorch GitHub repository against the [96m*onnx*[0m component. Attach the error report and the pt2 model.
## Exception summary
<class 'RuntimeError'>: scan might be aliasing the input or the output!
While executing %scan : [num_users=2] = call_function[target=torch.ops.higher_order.scan](args = (%scan_combine_graph_0, [%zeros_like], [%x], ()), kwargs = {})
GraphModule: class GraphModule(torch.nn.Module):
def forward(self, x):
x: "f32[s77, 3][3, 1]";
x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec)
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:386 in forward, code: init = torch.zeros_like(x[0])
select: "f32[3][1]" = torch.ops.aten.select.int(x, 0, 0)
zeros_like: "f32[3][1]" = torch.ops.aten.zeros_like.default(select, pin_memory = False); select = None
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:387 in forward, code: carry, out = torch.ops.higher_order.scan(
scan_combine_graph_0 = self.scan_combine_graph_0
scan = torch.ops.higher_order.scan(scan_combine_graph_0, [zeros_like], [x], ()); scan_combine_graph_0 = zeros_like = x = None
getitem: "f32[3][1]" = scan[0]
getitem_1: "f32[s77, 3][3, 1]" = scan[1]; scan = getitem_1 = None
return pytree.tree_unflatten((getitem,), self._out_spec)
class scan_combine_graph_0(torch.nn.Module):
def forward(self, carry_1: "f32[3][1]", y_1: "f32[3][1]"):
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:387 in forward, code: carry, out = torch.ops.higher_order.scan(
add: "f32[3][1]" = torch.ops.aten.add.Tensor(carry_1, y_1); carry_1 = y_1 = None
return [add, add]
Original traceback:
File "~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py", line 387, in forward
carry, out = torch.ops.higher_order.scan(
(Refer to the full stack trace above for more information.)
ControlFlowScan2Carried¶
forward¶
def forward(self, x):
init1 = torch.zeros_like(x[0])
init2 = torch.ones_like(x[0])
carry1, carry2, out1, out2 = torch.ops.higher_order.scan(
ControlFlowScan2Carried.add,
[init1, init2],
[x, x * 2],
# dim=0, # 01/31/2025, not supported anymore
additional_inputs=[],
)
return carry1, carry2, out1, out2
custom¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "2", "n_large_initializers": "0", "size_initializers": "12", "size_large_initializers": "0", "n_nodes": "4", "n_nodes_other_domain": "0", "mask_outputs": "[True, True, True, True]", "input_args": "(T1s3x4[-1.0,9.0:A3.5],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,4] x) => (float[4] output_0, float[4] output_1, float[batch,4] output_2, float[batch,4] output_3)
<int64[1] init7_s1_4 = {4}, float[1] "init1_s_::RSh1" = {2}, float[4] getitem_1, int64[1] init7_s1_1, int64 init7_s_0, float init1_s_, float[4] select, float[batch,4] _onx_mul_x, float[4] "scan#0", float[4] "scan#1", float[batch,4] mul, float[4] zeros_like, float[batch,4] getitem_2, float[batch,4] getitem_3, float[4] ones_like, float[4] getitem, float[4] select_1, float[batch,4] "scan#3", float[batch,4] "scan#2">
{
[zeros_likeD2] zeros_like = ConstantOfShape <value: tensor = float[1] {0}> (init7_s1_4)
[ones_likeD2] ones_like = ConstantOfShape <value: tensor = float[1] {1}> (init7_s1_4)
[mul_Tensor2] _onx_mul_x = Mul (x, "init1_s_::RSh1")
[scan] output_0, output_1, output_2, output_3 = Scan (zeros_like, ones_like, x, _onx_mul_x) <body: graph = experiment ( init_0_zeros_like, init_1_ones_like, scan_0_x, scan_1_mul) => ( output_0, output_1, output_2, output_3) {
[add_Tensor2] output_0 = Add (init_0_zeros_like, scan_0_x)
[mul_Tensor42] output_1 = Mul (init_1_ones_like, scan_1_mul)
[".output322"] output_2 = Identity (output_0)
[".output422"] output_3 = Identity (output_1)
}, num_scan_inputs: int = 2, scan_input_directions: ints = [0, 0], scan_output_axes: ints = [0, 0], scan_output_directions: ints = [0, 0]>
}
dynamo-ir¶
FAILED
Failed to decompose the FX graph for ONNX compatibility. [96mThis is step 2/3[0m of exporting the model to ONNX. Next steps:
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
- Create an error report with `torch.onnx.export(..., report=True)`, and save the ExportedProgram as a pt2 file. Create an issue in the PyTorch GitHub repository against the [96m*onnx*[0m component. Attach the error report and the pt2 model.
## Exception summary
<class 'RuntimeError'>: scan might be aliasing the input or the output!
While executing %scan : [num_users=4] = call_function[target=torch.ops.higher_order.scan](args = (%scan_combine_graph_0, [%zeros_like, %ones_like], [%x, %mul], ()), kwargs = {})
GraphModule: class GraphModule(torch.nn.Module):
def forward(self, x):
x: "f32[s77, 4][4, 1]";
x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec)
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:404 in forward, code: init1 = torch.zeros_like(x[0])
select: "f32[4][1]" = torch.ops.aten.select.int(x, 0, 0)
zeros_like: "f32[4][1]" = torch.ops.aten.zeros_like.default(select, pin_memory = False); select = None
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:405 in forward, code: init2 = torch.ones_like(x[0])
select_1: "f32[4][1]" = torch.ops.aten.select.int(x, 0, 0)
ones_like: "f32[4][1]" = torch.ops.aten.ones_like.default(select_1, pin_memory = False); select_1 = None
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:409 in forward, code: [x, x * 2],
mul: "f32[s77, 4][4, 1]" = torch.ops.aten.mul.Tensor(x, 2)
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:406 in forward, code: carry1, carry2, out1, out2 = torch.ops.higher_order.scan(
scan_combine_graph_0 = self.scan_combine_graph_0
scan = torch.ops.higher_order.scan(scan_combine_graph_0, [zeros_like, ones_like], [x, mul], ()); scan_combine_graph_0 = zeros_like = ones_like = x = mul = None
getitem: "f32[4][1]" = scan[0]
getitem_1: "f32[4][1]" = scan[1]
getitem_2: "f32[s77, 4][4, 1]" = scan[2]
getitem_3: "f32[s77, 4][4, 1]" = scan[3]; scan = None
return pytree.tree_unflatten((getitem, getitem_1, getitem_2, getitem_3), self._out_spec)
class scan_combine_graph_0(torch.nn.Module):
def forward(self, carry1_1: "f32[4][1]", carry2_1: "f32[4][1]", y1_1: "f32[4][1]", y2_1: "f32[4][1]"):
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:406 in forward, code: carry1, carry2, out1, out2 = torch.ops.higher_order.scan(
add: "f32[4][1]" = torch.ops.aten.add.Tensor(carry1_1, y1_1); carry1_1 = y1_1 = None
mul: "f32[4][1]" = torch.ops.aten.mul.Tensor(carry2_1, y2_1); carry2_1 = y2_1 = None
return [add, mul, add, mul]
Original traceback:
File "~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py", line 406, in forward
carry1, carry2, out1, out2 = torch.ops.higher_order.scan(
(Refer to the full stack trace above for more information.)
ControlFlowScanCDist¶
forward¶
def forward(self, x):
carry, out = torch.ops.higher_order.scan(
ControlFlowScanCDist.dist,
[x],
[x],
# dim=0, # 01/31/2025, not supported anymore
additional_inputs=[],
)
return out
custom¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "0", "n_large_initializers": "0", "size_initializers": "0", "size_large_initializers": "0", "n_nodes": "1", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4[-1.0,9.0:A3.5],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,4] x) => (float[batch,batch] output_0)
<float[batch,4] "scan#0", float[batch,batch] getitem_1, float[batch,batch] "scan#1", float[batch,4] getitem>
{
[scan] "scan#0", output_0 = Scan (x, x) <body: graph = experiment ( init_0_x, scan_0_x) => ( output_0, output_1) {
[init2cst4] "init7_s2_1_-12" = Constant <value: tensor = int64[2] "init7_s2_1_-1" {1,-1}> ()
[init2cst22] init7_s1_12 = Constant <value: tensor = int64[1] init7_s1_1 {1}> ()
[init2cst32] init1_s_2 = Constant <value: tensor = float init1_s_ {0.5}> ()
[reshape2] reshape2 = Reshape (scan_0_x, "init7_s2_1_-12")
[sub_Tensor2] sub2 = Sub (init_0_x, reshape2)
[mul_Tensor2] mul2 = Mul (sub2, sub2)
[sum2] sum_12 = ReduceSum <keepdims: int = 0> (mul2, init7_s1_12)
[pow_Tensor_Scalar2] output_1 = Pow (sum_12, init1_s_2)
[".output22"] output_0 = Identity (init_0_x)
}, num_scan_inputs: int = 1, scan_input_directions: ints = [0], scan_output_axes: ints = [0], scan_output_directions: ints = [0]>
}
dynamo-ir¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (float[batch,batch] getitem_1)
<float[batch,4] scan__0>
{
[node_scan__1] scan__0, getitem_1 = Scan (x, x) <body: graph = scan_combine_graph_0 (float[s77,4] x_scan_combine_graph_0__subgraph_in, float[4] x_scan_combine_graph_0__subgraph_in) => (float[s77,4] clone_scan_combine_graph_0, float[s77] pow_1_scan_combine_graph_0)
<int64[2] val_1, float[1,4] view, float[s77,4] sub_1, float[s77,4] mul_4, int64[1] val_5, float[s77] sum_1, float val_6>
{
[node_Constant_4] val_1 = Constant <value: tensor = int64[2] val_1 {1,-1}> ()
[node_view] view = Reshape <allowzero: int = 1> (x_scan_combine_graph_0__subgraph_in, val_1)
[node_sub_1] sub_1 = Sub (x_scan_combine_graph_0__subgraph_in, view)
[node_mul_4] mul_4 = Mul (sub_1, sub_1)
[node_Constant_7] val_5 = Constant <value: tensor = int64[1] val_5 {1}> ()
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (mul_4, val_5)
[node_Constant_6] val_6 = Constant <value: tensor = float {0.5}> ()
[node_pow_1] pow_1_scan_combine_graph_0 = Pow (sum_1, val_6)
[node_clone] clone_scan_combine_graph_0 = Identity (x_scan_combine_graph_0__subgraph_in)
}, num_scan_inputs: int = 1, scan_input_directions: ints = [0], scan_output_directions: ints = [0]>
}
FAILED
[ONNXRuntimeError] : 1 : FAIL : Error: Duplicate definition-site for (x_scan_combine_graph_0__subgraph_in).
ControlFlowScanCDist2¶
forward¶
def forward(self, x):
z = torch.tensor([0], dtype=torch.float32)
y = x.clone()
out = torch.ops.higher_order.scan(
ControlFlowScanCDist2.dist,
[z],
[x],
# dim=0, # 01/31/2025, not supported anymore
additional_inputs=[y],
)
return out[1]
custom¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "4", "size_large_initializers": "0", "n_nodes": "2", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4[-1.0,9.0:A3.5],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,4] x) => (float[batch,batch] output_0)
<float[1] c_lifted_tensor_0 = {0}, float[batch,4] hidden_input_scan_0_clone, float[batch,4] clone, float[1] "scan#0", float[1] detach_, float[batch,batch] getitem_1, float[batch,batch] "scan#1", float[1] getitem, float[1] lift_fresh_copy>
{
[_DONOTREMOVE_Scan_hidden_input_0] hidden_input_scan_0_clone = Identity (x)
[scan] "scan#0", output_0 = Scan (c_lifted_tensor_0, x) <body: graph = experiment ( init_0_detach_, scan_0_x) => ( output_0, output_1) {
[init2cst3] "init7_s2_1_-12" = Constant <value: tensor = int64[2] "init7_s2_1_-1" {1,-1}> ()
[init2cst22] init7_s1_12 = Constant <value: tensor = int64[1] init7_s1_1 {1}> ()
[reshape2] reshape2 = Reshape (scan_0_x, "init7_s2_1_-12")
[sub_Tensor2] sub2 = Sub (hidden_input_scan_0_clone, reshape2)
[mul_Tensor2] mul2 = Mul (sub2, sub2)
[sum2] sum_12 = ReduceSum <keepdims: int = 0> (mul2, init7_s1_12)
[sqrt2] output_1 = Sqrt (sum_12)
[".output22"] output_0 = Identity (init_0_detach_)
}, num_scan_inputs: int = 1, scan_input_directions: ints = [0], scan_output_axes: ints = [0], scan_output_directions: ints = [0]>
}
dynamo-ir¶
inputs:
#1[(T1s3x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (float[batch,batch] getitem_1)
<float[1] clone, float[1] scan__0>
{
[node_Constant_0] clone = Constant <value: tensor = float[1] clone {0}> ()
[node_scan__1] scan__0, getitem_1 = Scan (clone, x) <body: graph = scan_combine_graph_0 (float[1] clone_scan_combine_graph_0__subgraph_in, float[4] x_scan_combine_graph_0__subgraph_in) => (float[1] clone_scan_combine_graph_0, float[batch] sqrt_scan_combine_graph_0)
<int64[2] val_1, float[1,4] view, float[batch,4] sub_1, float[batch,4] mul_4, int64[1] val_5, float[batch] sum_1>
{
[node_Constant_4] val_1 = Constant <value: tensor = int64[2] val_1 {1,-1}> ()
[node_view] view = Reshape <allowzero: int = 1> (x_scan_combine_graph_0__subgraph_in, val_1)
[node_sub_1] sub_1 = Sub (x, view)
[node_mul_4] mul_4 = Mul (sub_1, sub_1)
[node_Constant_7] val_5 = Constant <value: tensor = int64[1] val_5 {1}> ()
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (mul_4, val_5)
[node_sqrt] sqrt_scan_combine_graph_0 = Sqrt (sum_1)
[node_clone_2] clone_scan_combine_graph_0 = Identity (clone_scan_combine_graph_0__subgraph_in)
}, num_scan_inputs: int = 1, scan_input_directions: ints = [0], scan_output_directions: ints = [0]>
}
ControlFlowScanCDistXY¶
forward¶
def forward(self, x, y):
carry, out = torch.ops.higher_order.scan(
ControlFlowScanCDistXY.dist,
[y],
[x],
# dim=0, # 01/31/2025, not supported anymore
additional_inputs=[],
)
return out
custom¶
inputs:
#2[(T1s3x4,T1s5x4),(T1s13x14,T1s15x14)]shapes:
dict(x:{0:Dim(x_rows),1:Dim(dim)},y:{0:Dim(y_rows),1:Dim(dim)})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "0", "n_large_initializers": "0", "size_initializers": "0", "size_large_initializers": "0", "n_nodes": "1", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4[-1.077669382095337,1.4364134073257446:A-0.06034492305479944],T1s5x4[-2.107510566711426,2.5612235069274902:A0.26090582795441153])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('x_rows', min=0), 1: Dim('dim', min=0)},
'y': {0: Dim('y_rows', min=0), 1: Dim('dim', min=0)}}", "_discovered_shape_constraints": "{'dim': {'s94', 's27'},
's17': {'y_rows'},
's27': {'dim'},
's77': {'x_rows'},
's94': {'dim'},
'x_rows': {'s77'},
'y_rows': {'s17'}}"]
>
experiment (float[x_rows,dim] x, float[y_rows,dim] y) => (float[x_rows,y_rows] output_0)
<float[y_rows,dim] "scan#0", float[x_rows,y_rows] getitem_1, float[x_rows,y_rows] "scan#1", float[y_rows,dim] getitem>
{
[scan] "scan#0", output_0 = Scan (y, x) <body: graph = experiment ( init_0_y, scan_0_x) => ( output_0, output_1) {
[init2cst3] "init7_s2_1_-12" = Constant <value: tensor = int64[2] "init7_s2_1_-1" {1,-1}> ()
[init2cst22] init7_s1_12 = Constant <value: tensor = int64[1] init7_s1_1 {1}> ()
[reshape2] reshape2 = Reshape (scan_0_x, "init7_s2_1_-12")
[sub_Tensor2] sub2 = Sub (init_0_y, reshape2)
[mul_Tensor2] mul2 = Mul (sub2, sub2)
[sum2] sum_12 = ReduceSum <keepdims: int = 0> (mul2, init7_s1_12)
[sqrt2] output_1 = Sqrt (sum_12)
[".output22"] output_0 = Identity (init_0_y)
}, num_scan_inputs: int = 1, scan_input_directions: ints = [0], scan_output_axes: ints = [0], scan_output_directions: ints = [0]>
}
dynamo-ir¶
inputs:
#2[(T1s3x4,T1s5x4),(T1s13x14,T1s15x14)]shapes:
dict(x:{0:Dim(x_rows),1:Dim(dim)},y:{0:Dim(y_rows),1:Dim(dim)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[x_rows,dim] x, float[y_rows,dim] y) => (float[x_rows,y_rows] getitem_1)
<float[y_rows,dim] scan__0>
{
[node_scan__1] scan__0, getitem_1 = Scan (y, x) <body: graph = scan_combine_graph_0 (float[s17,s27] y_scan_combine_graph_0__subgraph_in, float[s27] x_scan_combine_graph_0__subgraph_in) => (float[s17,s27] clone_scan_combine_graph_0, float[s17] sqrt_scan_combine_graph_0)
<int64[2] val_1_2, float[1,?] view, float[s17,?] sub_4, float[s17,?] mul_7, int64[1] val_5, float[s17] sum_1>
{
[node_Constant_4] val_1_2 = Constant <value: tensor = int64[2] val_1_2 {1,-1}> ()
[node_view] view = Reshape <allowzero: int = 1> (x_scan_combine_graph_0__subgraph_in, val_1_2)
[node_sub_4] sub_4 = Sub (y_scan_combine_graph_0__subgraph_in, view)
[node_mul_7] mul_7 = Mul (sub_4, sub_4)
[node_Constant_7] val_5 = Constant <value: tensor = int64[1] val_5 {1}> ()
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 0> (mul_7, val_5)
[node_sqrt] sqrt_scan_combine_graph_0 = Sqrt (sum_1)
[node_clone] clone_scan_combine_graph_0 = Identity (y_scan_combine_graph_0__subgraph_in)
}, num_scan_inputs: int = 1, scan_input_directions: ints = [0], scan_output_directions: ints = [0]>
}
ControlFlowScanDecomposition_151564¶
forward¶
def forward(self, images, position):
return self.select_when_exporting(self.dummy_loop, self.dummy_loop_with_scan)(
images, position
)
custom¶
inputs:
#1[(T1s5x6,T7s5)]shapes:
dict(images:{0:DYNAMIC,1:DYNAMIC},position:{0:DYNAMIC})
<
ir_version: 8,
opset_import: ["" : 18, "local_functions" : 1],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "0", "n_large_initializers": "0", "size_initializers": "0", "size_large_initializers": "0", "n_nodes": "1", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s5x6[-1.8678492307662964,1.425422191619873:A-0.22642420046031475],T7s5[1,5:A3.0])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'images': {0: Dim('DYN0', min=0), 1: Dim('DYN1', min=0)},
'position': {0: Dim('DYN2', min=0)}}", "_discovered_shape_constraints": "{'DYN0': {'s34'},
'DYN1': {'s90'},
'DYN2': {'s71'},
's34': {'DYN0'},
's71': {'DYN2'},
's90': {'DYN1'}}"]
>
experiment (float[batch,channel] images, int64[batch_1] position) => (float[batch,channel] output_0)
<float[batch,channel] "scan#0", float[batch,channel] getitem>
{
[scan] output_0 = Scan (images, position) <body: graph = experiment ( scan_0_images, scan_1_position) => ( output_0) {
[init2cst2] init7_s1_02 = Constant <value: tensor = int64[1] init7_s1_0 {0}> ()
[get_dynamic_dimension_a_item2] "item::UnSq03" = Unsqueeze (scan_1_position, init7_s1_02)
[sym_size_int2] "padded_1::Shape:12" = Shape <end: int = 1, start: int = 0> (scan_0_images)
[zerosA22] zeros2 = ConstantOfShape <value: tensor = float[1] {0}> ("padded_1::Shape:12")
[slide_Tensor2] slice_12 = Slice (scan_0_images, init7_s1_02, "item::UnSq03", init7_s1_02)
[setitem_rk022] "zeros::Shape:2" = Shape (zeros2)
[setitem_rk032] _onx_slice_zeros2 = Slice (zeros2, "item::UnSq03", "zeros::Shape:2", init7_s1_02)
[setitem_rk042] output_0 = Concat <axis: int = 0> (slice_12, _onx_slice_zeros2)
}, num_scan_inputs: int = 2, scan_input_directions: ints = [0, 0], scan_output_axes: ints = [0], scan_output_directions: ints = [0]>
}
dynamo-ir¶
FAILED
Failed to decompose the FX graph for ONNX compatibility. [96mThis is step 2/3[0m of exporting the model to ONNX. Next steps:
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
- Create an error report with `torch.onnx.export(..., report=True)`, and save the ExportedProgram as a pt2 file. Create an issue in the PyTorch GitHub repository against the [96m*onnx*[0m component. Attach the error report and the pt2 model.
## Exception summary
<class 'torch.fx.experimental.symbolic_shapes.GuardOnDataDependentSymNode'>: Could not guard on data-dependent expression u1 < 0 (unhinted: u1 < 0). (Size-like symbols: none)
Caused by: (_decomp/decompositions.py:745 in slice_forward)
For more information, run with TORCH_LOGS="dynamic"
For extended logs when we create symbols, also add TORCHDYNAMO_EXTENDED_DEBUG_CREATE_SYMBOL="u1"
If you suspect the guard was triggered from C++, add TORCHDYNAMO_EXTENDED_DEBUG_CPP=1
For more debugging help, see https://docs.google.com/document/d/1HSuTTVvYH1pTew89Rtpeu84Ht3nQEFTYhAX3Ypa_xJs/edit?usp=sharing
For C++ stack trace, run with TORCHDYNAMO_EXTENDED_DEBUG_CPP=1
While executing %scan : [num_users=1] = call_function[target=torch.ops.higher_order.scan](args = (%scan_combine_graph_0, [], [%images, %position], ()), kwargs = {})
GraphModule: class GraphModule(torch.nn.Module):
def forward(self, images, position):
images: "f32[s34, s90][s90, 1]"; position: "i64[s71][1]";
images, position, = fx_pytree.tree_flatten_spec(([images, position], {}), self._in_spec)
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:568 in forward, code: return self.select_when_exporting(self.dummy_loop, self.dummy_loop_with_scan)(
scan_combine_graph_0 = self.scan_combine_graph_0
scan = torch.ops.higher_order.scan(scan_combine_graph_0, [], [images, position], ()); scan_combine_graph_0 = images = position = None
getitem: "f32[s34, s90][s90, 1]" = scan[0]; scan = None
return pytree.tree_unflatten((getitem,), self._out_spec)
class scan_combine_graph_0(torch.nn.Module):
def forward(self, padded_1: "f32[s90][1]", p_1: "i64[][]"):
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:568 in forward, code: return self.select_when_exporting(self.dummy_loop, self.dummy_loop_with_scan)(
sym_size_int: "Sym(s90)" = torch.ops.aten.sym_size.int(padded_1, 0)
zeros: "f32[s90][1]" = torch.ops.aten.zeros.default([sym_size_int], device = device(type='cpu'), pin_memory = False); sym_size_int = None
item: "Sym(u0)" = torch.ops.aten.item.default(p_1); p_1 = None
slice_1: "f32[u0][1]" = torch.ops.aten.slice.Tensor(padded_1, 0, 0, item); padded_1 = None
slice_2: "f32[u0][1]" = torch.ops.aten.slice.Tensor(zeros, 0, 0, item); item = None
copy_: "f32[u0][1]" = torch.ops.aten.copy_.default(slice_2, slice_1); slice_2 = slice_1 = copy_ = None
return (zeros,)
Original traceback:
File "~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py", line 568, in forward
return self.select_when_exporting(self.dummy_loop, self.dummy_loop_with_scan)(
(Refer to the full stack trace above for more information.)
ControlFlowScanInplace_153705¶
forward¶
def forward(self, x, y):
def loop_body_1(z, iv, x, y):
z = z.clone()
i = iv.item()
z[i, :] = ((x[i, :] - y) ** 2).sum(dim=-1)
return [z, iv]
z = torch.empty((x.shape[0], y.shape[0]))
r = torch.ops.higher_order.scan(
loop_body_1, [z], [torch.arange(x.shape[0], dtype=torch.int64)], [x, y]
)
return r[0]
custom¶
FAILED
only integers, slices (`:`), ellipsis (`...`), None and long or byte Variables are valid indices (got SymInt)
dynamo-ir¶
FAILED
Failed to decompose the FX graph for ONNX compatibility. [96mThis is step 2/3[0m of exporting the model to ONNX. Next steps:
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
- Create an error report with `torch.onnx.export(..., report=True)`, and save the ExportedProgram as a pt2 file. Create an issue in the PyTorch GitHub repository against the [96m*onnx*[0m component. Attach the error report and the pt2 model.
## Exception summary
<class 'RuntimeError'>: scan might be aliasing the input or the output!
While executing %scan : [num_users=2] = call_function[target=torch.ops.higher_order.scan](args = (%scan_combine_graph_0, [%empty], [%arange], (%x, %y)), kwargs = {})
GraphModule: class GraphModule(torch.nn.Module):
def forward(self, x, y):
x: "f32[s77, s27][s27, 1]"; y: "f32[s17, s27][s27, 1]";
x, y, = fx_pytree.tree_flatten_spec(([x, y], {}), self._in_spec)
#
sym_size_int_2: "Sym(s77)" = torch.ops.aten.sym_size.int(x, 0)
sym_size_int_3: "Sym(s27)" = torch.ops.aten.sym_size.int(x, 1)
sym_size_int_4: "Sym(s17)" = torch.ops.aten.sym_size.int(y, 0)
sym_size_int_5: "Sym(s27)" = torch.ops.aten.sym_size.int(y, 1)
eq: "Sym(True)" = sym_size_int_3 == sym_size_int_5; sym_size_int_3 = sym_size_int_5 = None
_assert_scalar_default = torch.ops.aten._assert_scalar.default(eq, "Runtime assertion failed for expression Eq(s27, s94) on node 'eq'"); eq = _assert_scalar_default = None
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:517 in forward, code: z = torch.empty((x.shape[0], y.shape[0]))
empty: "f32[s77, s17][s17, 1]" = torch.ops.aten.empty.memory_format([sym_size_int_2, sym_size_int_4], device = device(type='cpu'), pin_memory = False); sym_size_int_4 = None
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:519 in forward, code: loop_body_1, [z], [torch.arange(x.shape[0], dtype=torch.int64)], [x, y]
arange: "i64[s77][1]" = torch.ops.aten.arange.default(sym_size_int_2, dtype = torch.int64, device = device(type='cpu'), pin_memory = False); sym_size_int_2 = None
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:518 in forward, code: r = torch.ops.higher_order.scan(
scan_combine_graph_0 = self.scan_combine_graph_0
scan = torch.ops.higher_order.scan(scan_combine_graph_0, [empty], [arange], (x, y)); scan_combine_graph_0 = empty = arange = x = y = None
getitem: "f32[s77, s17][s17, 1]" = scan[0]
getitem_1: "i64[s77][1]" = scan[1]; scan = getitem_1 = None
return pytree.tree_unflatten((getitem,), self._out_spec)
class scan_combine_graph_0(torch.nn.Module):
def forward(self, z_1: "f32[s77, s17][s17, 1]", iv_1: "i64[][]", x_1: "f32[s77, s27][s27, 1]", y_1: "f32[s17, s27][s27, 1]"):
# File: ~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py:518 in forward, code: r = torch.ops.higher_order.scan(
clone: "f32[s77, s17][s17, 1]" = torch.ops.aten.clone.default(z_1); z_1 = None
select: "f32[s27][1]" = torch.ops.aten.select.int(x_1, 0, 0); x_1 = None
slice_1: "f32[s27][1]" = torch.ops.aten.slice.Tensor(select, 0, 0, 9223372036854775807); select = None
sub: "f32[s17, s27][s27, 1]" = torch.ops.aten.sub.Tensor(slice_1, y_1); slice_1 = y_1 = None
pow_1: "f32[s17, s27][s27, 1]" = torch.ops.aten.pow.Tensor_Scalar(sub, 2); sub = None
sum_1: "f32[s17][1]" = torch.ops.aten.sum.dim_IntList(pow_1, [-1]); pow_1 = None
select_1: "f32[s17][1]" = torch.ops.aten.select.int(clone, 0, 0)
slice_2: "f32[s17][1]" = torch.ops.aten.slice.Tensor(select_1, 0, 0, 9223372036854775807); select_1 = None
copy_: "f32[s17][1]" = torch.ops.aten.copy_.default(slice_2, sum_1); slice_2 = sum_1 = copy_ = None
return [clone, iv_1]
Original traceback:
File "~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py", line 518, in forward
r = torch.ops.higher_order.scan(
(Refer to the full stack trace above for more information.)
CreateFromShape¶
forward¶
def forward(self, x):
y = torch.ones((x.shape[0], x.shape[1] + 1))
return y
custom¶
inputs:
#2[(T1s4x4,),(T1s5x5,)]shapes:
dict(x:{0:Dim(dx),1:Dim(dy)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "2", "n_large_initializers": "0", "size_initializers": "16", "size_large_initializers": "0", "n_nodes": "7", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x4[0.028293073177337646,0.9380697011947632:A0.5150020532310009],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('dx', min=0), 1: Dim('dy', min=0)}}", "_discovered_shape_constraints": "{'dx': {'s77'}, 'dy': {'s27'}, 's27': {'dy'}, 's77': {'dx'}}", "_known_value_shapes": "{
_onx_add_sym_size_int_3: dy+1,
_onx_concat_sym_size_int_2::UnSq0: ('dx', 'add'),
add::UnSq0: ('add',),
sym_size_int_2: dx,
sym_size_int_2::UnSq0: ('dx',),
sym_size_int_3: dy,
x::Shape1:2: ('dy',),
x::Shape:1: ('dx',),
}"]
>
experiment (float[dx,dy] x) => (float[dx,dy+1] output_0)
<int64 init7_s_1 = {1}, int64[1] init7_s1_0 = {0}, int64[1] "init7_s_1::RSh1", int64 _onx_add_sym_size_int_3, int64 sym_size_int_3, int64[1] "sym_size_int_3::RSh1", float[dx,dy+1] ones, int64[1] "x::Shape1:2", int64[1] "add::UnSq0", int64[1] "sym_size_int_2::UnSq0", int64[1] init7_s1_1, int64 add, int64[1] "x::Shape:1", int64 sym_size_int_2>
{
[sym_size_int] "x::Shape:1" = Shape <end: int = 1, start: int = 0> (x)
[sym_size_int3] "x::Shape1:2" = Shape <end: int = 2, start: int = 1> (x)
[sym_size_int4] sym_size_int_3 = Squeeze ("x::Shape1:2")
[add3] _onx_add_sym_size_int_3 = Add (sym_size_int_3, init7_s_1)
[_mkshape1_add] "add::UnSq0" = Unsqueeze (_onx_add_sym_size_int_3, init7_s1_0)
[_mkshape_add] "_onx_concat_sym_size_int_2::UnSq0" = Concat <axis: int = 0> ("x::Shape:1", "add::UnSq0")
[ones] output_0 = ConstantOfShape <value: tensor = float[1] {1}> ("_onx_concat_sym_size_int_2::UnSq0")
}
dynamo-ir¶
inputs:
#2[(T1s4x4,),(T1s5x5,)]shapes:
dict(x:{0:Dim(dx),1:Dim(dy)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[dx,dy] x) => (float[dx,dy + 1] ones)
<int64[1] val_0, int64[1] val_1, int64 sym_size_int_3, int64 val_2, int64 add, int64[1] val_5, int64[1] val_6, int64[2] val_7, float val_10>
{
[node_Shape_0] val_0 = Shape <end: int = 1, start: int = 0> (x)
[node_Shape_1] val_1 = Shape <end: int = 2, start: int = 1> (x)
[node_sym_size_int_3] sym_size_int_3 = Squeeze (val_1)
[node_Constant_2] val_2 = Constant <value: tensor = int64 {1}> ()
[node_add] add = Add (sym_size_int_3, val_2)
[node_Constant_5] val_5 = Constant <value: tensor = int64[1] {-1}> ()
[node_Reshape_6] val_6 = Reshape <allowzero: int = 0> (add, val_5)
[node_Concat_7] val_7 = Concat <axis: int = 0> (val_0, val_6)
[node_Constant_13] val_10 = Constant <value: tensor = float val_10 {1}> ()
[node_ones] ones = Expand (val_10, val_7)
}
CreateFromShapeThroughFunction¶
forward¶
def forward(self, x):
dy1 = CreateFromShapeThroughFunction.add_one(x.shape[1])
y = torch.ones((x.shape[0], dy1))
return y
custom¶
inputs:
#2[(T1s4x4,),(T1s5x5,)]shapes:
dict(x:{0:Dim(dx),1:Dim(dy)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "2", "n_large_initializers": "0", "size_initializers": "16", "size_large_initializers": "0", "n_nodes": "7", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x4[0.021807074546813965,0.965708315372467:A0.5018488876521587],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('dx', min=0), 1: Dim('dy', min=0)}}", "_discovered_shape_constraints": "{'dx': {'s77'}, 'dy': {'s27'}, 's27': {'dy'}, 's77': {'dx'}}", "_known_value_shapes": "{
_onx_add_sym_size_int_3: dy+1,
_onx_concat_sym_size_int_2::UnSq0: ('dx', 'add'),
add::UnSq0: ('add',),
sym_size_int_2: dx,
sym_size_int_2::UnSq0: ('dx',),
sym_size_int_3: dy,
x::Shape1:2: ('dy',),
x::Shape:1: ('dx',),
}"]
>
experiment (float[dx,dy] x) => (float[dx,dy+1] output_0)
<int64 init7_s_1 = {1}, int64[1] init7_s1_0 = {0}, int64[1] "init7_s_1::RSh1", int64 _onx_add_sym_size_int_3, int64 sym_size_int_3, int64[1] "sym_size_int_3::RSh1", float[dx,dy+1] ones, int64[1] "x::Shape1:2", int64[1] "add::UnSq0", int64[1] "sym_size_int_2::UnSq0", int64[1] init7_s1_1, int64 add, int64[1] "x::Shape:1", int64 sym_size_int_2>
{
[sym_size_int] "x::Shape:1" = Shape <end: int = 1, start: int = 0> (x)
[sym_size_int3] "x::Shape1:2" = Shape <end: int = 2, start: int = 1> (x)
[sym_size_int4] sym_size_int_3 = Squeeze ("x::Shape1:2")
[add3] _onx_add_sym_size_int_3 = Add (sym_size_int_3, init7_s_1)
[_mkshape1_add] "add::UnSq0" = Unsqueeze (_onx_add_sym_size_int_3, init7_s1_0)
[_mkshape_add] "_onx_concat_sym_size_int_2::UnSq0" = Concat <axis: int = 0> ("x::Shape:1", "add::UnSq0")
[ones] output_0 = ConstantOfShape <value: tensor = float[1] {1}> ("_onx_concat_sym_size_int_2::UnSq0")
}
dynamo-ir¶
inputs:
#2[(T1s4x4,),(T1s5x5,)]shapes:
dict(x:{0:Dim(dx),1:Dim(dy)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[dx,dy] x) => (float[dx,dy + 1] ones)
<int64[1] val_0, int64[1] val_1, int64 sym_size_int_3, int64 val_2, int64 add, int64[1] val_5, int64[1] val_6, int64[2] val_7, float val_10>
{
[node_Shape_0] val_0 = Shape <end: int = 1, start: int = 0> (x)
[node_Shape_1] val_1 = Shape <end: int = 2, start: int = 1> (x)
[node_sym_size_int_3] sym_size_int_3 = Squeeze (val_1)
[node_Constant_2] val_2 = Constant <value: tensor = int64 {1}> ()
[node_add] add = Add (sym_size_int_3, val_2)
[node_Constant_5] val_5 = Constant <value: tensor = int64[1] {-1}> ()
[node_Reshape_6] val_6 = Reshape <allowzero: int = 0> (add, val_5)
[node_Concat_7] val_7 = Concat <axis: int = 0> (val_0, val_6)
[node_Constant_13] val_10 = Constant <value: tensor = float val_10 {1}> ()
[node_ones] ones = Expand (val_10, val_7)
}
CropLastDimensionWithTensorContent¶
forward¶
def forward(self, x, shape):
return x[..., : shape[0]]
custom¶
FAILED
Could not extract specialized integer from data-dependent expression u0 (unhinted: u0). (Size-like symbols: none)
Caused by: (_export/non_strict_utils.py:1054 in __torch_function__)
For more information, run with TORCH_LOGS="dynamic"
For extended logs when we create symbols, also add TORCHDYNAMO_EXTENDED_DEBUG_CREATE_SYMBOL="u0"
If you suspect the guard was triggered from C++, add TORCHDYNAMO_EXTENDED_DEBUG_CPP=1
For more debugging help, see https://docs.google.com/document/d/1HSuTTVvYH1pTew89Rtpeu84Ht3nQEFTYhAX3Ypa_xJs/edit?usp=sharing
For C++ stack trace, run with TORCHDYNAMO_EXTENDED_DEBUG_CPP=1
The following call raised this error:
File "~/github/onnx-diagnostic/onnx_diagnostic/torch_export_patches/eval/model_cases.py", line 818, in forward
return x[..., : shape[0]]
The error above occurred when calling torch.export.export. If you would like to view some more information about this error, and get a list of all other errors that may occur in your export call, you can replace your `export()` call with `draft_export()`.
dynamo-ir¶
inputs:
#2[(T1s3x4x4,T7s1),(T1s6x4x4,T7s1)]shapes:
dict(x:{0:Dim(batch)},shape:{})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4,4] x, int64[1] shape) => (float[batch,4,2] slice_1)
<int64[1] val_7, int64[1] val_10, int64[1] val_14>
{
[node_Constant_17] val_7 = Constant <value: tensor = int64[1] val_7 {0}> ()
[node_Constant_20] val_10 = Constant <value: tensor = int64[1] val_10 {2}> ()
[node_Constant_14] val_14 = Constant <value_ints: ints = [1]> ()
[node_slice_1] slice_1 = Slice (x, val_7, val_10, val_10, val_14)
}
FAILED
diff.1
CropLastDimensionWithTensorShape¶
forward¶
def forward(self, x, y):
return x[..., : y.shape[0]]
custom¶
inputs:
#2[(T1s3x4x4,T1s2),(T1s6x4x4,T1s3)]shapes:
dict(x:{0:Dim(batch)},y:{0:Dim(crop)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "2", "n_large_initializers": "0", "size_initializers": "16", "size_large_initializers": "0", "n_nodes": "2", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4x4[0.003647148609161377,0.9746490716934204:A0.49173058196902275],T1s2[0.036011695861816406,0.23263895511627197:A0.1343253254890442])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}, 'y': {0: Dim('crop', min=1, max=3)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 'crop': {'s17'}, 's17': {'crop'}, 's77': {'batch'}}", "_known_value_shapes": "{
sym_size_int_2: crop,
sym_size_int_2::UnSq0: ('crop',),
y::Shape:1: ('crop',),
}"]
>
experiment (float[batch,4,4] x, float[crop] y) => (float[batch,4,crop] output_0)
<int64[1] init7_s1_0 = {0}, int64[1] init7_s1_2 = {2}, int64[1] "y::Shape:1", int64[1] "sym_size_int_2::UnSq0", float[batch,4,crop] slice_1, int64 sym_size_int_2>
{
[sym_size_int] "y::Shape:1" = Shape <end: int = 1, start: int = 0> (y)
[slide_Tensor] output_0 = Slice (x, init7_s1_0, "y::Shape:1", init7_s1_2)
}
dynamo-ir¶
inputs:
#2[(T1s3x4x4,T1s2),(T1s6x4x4,T1s3)]shapes:
dict(x:{0:Dim(batch)},y:{0:Dim(crop)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4,4] x, float[crop] y) => (float[batch,4,crop] slice_1)
<int64[1] val_0, int64[1] val_4, int64[1] val_11, int64[1] val_12>
{
[node_Shape_0] val_0 = Shape <end: int = 1, start: int = 0> (y)
[node_Constant_15] val_4 = Constant <value: tensor = int64[1] val_4 {0}> ()
[node_Constant_19] val_11 = Constant <value: tensor = int64[1] val_11 {2}> ()
[node_Constant_12] val_12 = Constant <value_ints: ints = [1]> ()
[node_slice_1] slice_1 = Slice (x, val_4, val_0, val_11, val_12)
}
InplaceAdd¶
forward¶
def forward(self, x):
x += self.bias
return x
custom¶
inputs:
#2[(T1s3x4,),(T1s5x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "16", "size_large_initializers": "0", "n_nodes": "1", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4[6.029128551483154,6.933902263641357:A6.421988209088643],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,4] x) => (float[batch,4] output_0)
<float[1,4] c_bias = {1,1,1,1}, float[batch,4] add_>
{
[add__Tensor] output_0 = Add (x, c_bias)
}
dynamo-ir¶
inputs:
#2[(T1s3x4,),(T1s5x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (float[batch,4] add_3)
<float[1,4] bias = {1,1,1,1}, float[1,4] bias>
{
[node_add_3] add_3 = Add (x, bias)
}
InplaceAdd2¶
forward¶
def forward(self, x):
x.add_(self.bias)
return x
custom¶
inputs:
#2[(T1s3x4,),(T1s5x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "16", "size_large_initializers": "0", "n_nodes": "1", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4[8.156997680664062,8.949931144714355:A8.431713183720907],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,4] x) => (float[batch,4] output_0)
<float[1,4] c_bias = {1,1,1,1}, float[batch,4] add_>
{
[add__Tensor] output_0 = Add (x, c_bias)
}
dynamo-ir¶
inputs:
#2[(T1s3x4,),(T1s5x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (float[batch,4] add_3)
<float[1,4] bias = {1,1,1,1}, float[1,4] bias>
{
[node_add_3] add_3 = Add (x, bias)
}
InplaceAdd_Mul¶
forward¶
def forward(self, x):
x.add_(self.bias)
return x * 2
custom¶
inputs:
#2[(T1s3x4,),(T1s5x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "2", "n_large_initializers": "0", "size_initializers": "20", "size_large_initializers": "0", "n_nodes": "2", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4[8.001041412353516,8.841524124145508:A8.332729657491049],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,4] x) => (float[batch,4] output_0)
<float[1,4] c_bias = {1,1,1,1}, float[1] "init1_s_::RSh1" = {2}, float[batch,4] add_, float[batch,4] _onx_mul_add_, int64[1] init7_s1_1, float init1_s_, float[batch,4] mul>
{
[add__Tensor] add_ = Add (x, c_bias)
[mul_Tensor2] output_0 = Mul (add_, "init1_s_::RSh1")
}
dynamo-ir¶
inputs:
#2[(T1s3x4,),(T1s5x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (float[batch,4] mul_4)
<float[1,4] bias = {1,1,1,1}, float[1,4] bias, float[batch,4] add_3, float scalar_tensor_default>
{
[node_add_3] add_3 = Add (x, bias)
[node_Constant_1] scalar_tensor_default = Constant <value: tensor = float scalar_tensor_default {2}> ()
[node_mul_4] mul_4 = Mul (add_3, scalar_tensor_default)
}
InplaceCloneAdd¶
forward¶
def forward(self, x):
x = x.clone()
x.add_(self.bias)
return x
custom¶
inputs:
#2[(T1s3x4,),(T1s5x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "16", "size_large_initializers": "0", "n_nodes": "1", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s3x4[0.026060938835144043,0.8989080190658569:A0.3850300113360087],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,4] x) => (float[batch,4] output_0)
<float[1,4] c_bias = {1,1,1,1}, float[batch,4] clone, float[batch,4] add_>
{
[add__Tensor] output_0 = Add (x, c_bias)
}
dynamo-ir¶
inputs:
#2[(T1s3x4,),(T1s5x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (float[batch,4] add_6)
<float[1,4] bias = {1,1,1,1}, float[1,4] bias>
{
[node_add_6] add_6 = Add (x, bias)
}
InplaceSetItemEllipsis_1¶
forward¶
def forward(self, index, update):
copy = self.params.clone()
copy[..., index] = update
return copy
custom¶
FAILED
L['update'].size()[0] = 8192 is not equal to L['index'].size()[0] = 4
The error above occurred when calling torch.export.export. If you would like to view some more information about this error, and get a list of all other errors that may occur in your export call, you can replace your `export()` call with `draft_export()`.
dynamo-ir¶
FAILED
Failed to convert the exported program to an ONNX model. [96mThis is step 3/3[0m of exporting the model to ONNX. Next steps:
- If there is a missing ONNX function, implement it and register it to the registry.
- If there is an internal error during ONNX conversion, debug the error and summit a PR to PyTorch.
- Create an error report with `torch.onnx.export(..., report=True)`, and save the ExportedProgram as a pt2 file. Create an issue in the PyTorch GitHub repository against the [96m*onnx*[0m component. Attach the error report and the pt2 model.
## Exception summary
<class 'TypeError'>: int() argument must be a string, a bytes-like object or a real number, not 'SymbolicDim'
⬆️
<class 'torch.onnx._internal.exporter._errors.GraphConstructionError'>: Error processing Python constants for operator '::Expand'. named_inputs={'input': SymbolicTensor(name='anonymous:132313144740592', producer=anonymous_node:132316699979952, index=0), 'shape': (SymbolicDim(s5), SymbolicDim(s32))}, named_attrs={}, opset=, op_signature=''::Expand(input: T, shape: shape) -> (T) where T=UINT64 | COMPLEX64 | UINT8 | DOUBLE | UINT16 | BFLOAT16 | INT16 | STRING | FLOAT | INT32 | FLOAT16 | UINT32 | COMPLEX128 | INT8 | INT64 | BOOL, shape=INT64.
⬆️
<class 'torch.onnx._internal.exporter._errors.GraphConstructionError'>: Error calling operator 'Expand' with args (SymbolicTensor(name='anonymous:132313144740592', producer=anonymous_node:132316699979952, index=0), (SymbolicDim(s5), SymbolicDim(s32))) and kwargs {}.
⬆️
<class 'torch.onnx._internal.exporter._errors.GraphConstructionError'>: Error when calling function 'TracedOnnxFunction(<function aten_index_put at 0x7856941954e0>)' with args '[SymbolicTensor(name='clone', type=Tensor(FLOAT), shape=Shape([1, 8192, 4]), producer='node_clone', index=0), [None, None, SymbolicTensor(name='index', type=Tensor(INT64), shape=Shape([SymbolicDim(s91)]))], SymbolicTensor(name='update', type=Tensor(FLOAT), shape=Shape([SymbolicDim(s5), SymbolicDim(s32)]))]' and kwargs '{}'
⬆️
<class 'torch.onnx._internal.exporter._errors.ConversionError'>: Error when translating node %index_put : [num_users=1] = call_function[target=torch.ops.aten.index_put.default](args = (%clone, [None, None, %index], %update), kwargs = {}). See the stack trace for more information.
(Refer to the full stack trace above for more information.)
InplaceSetItemEllipsis_2¶
forward¶
def forward(self, index, update):
copy = self.params.clone()
copy[..., index] = update
return copy
custom¶
FAILED
L['update'].size()[0] = 8192 is not equal to L['index'].size()[0] = 4
The error above occurred when calling torch.export.export. If you would like to view some more information about this error, and get a list of all other errors that may occur in your export call, you can replace your `export()` call with `draft_export()`.
dynamo-ir¶
FAILED
Failed to convert the exported program to an ONNX model. [96mThis is step 3/3[0m of exporting the model to ONNX. Next steps:
- If there is a missing ONNX function, implement it and register it to the registry.
- If there is an internal error during ONNX conversion, debug the error and summit a PR to PyTorch.
- Create an error report with `torch.onnx.export(..., report=True)`, and save the ExportedProgram as a pt2 file. Create an issue in the PyTorch GitHub repository against the [96m*onnx*[0m component. Attach the error report and the pt2 model.
## Exception summary
<class 'TypeError'>: int() argument must be a string, a bytes-like object or a real number, not 'SymbolicDim'
⬆️
<class 'torch.onnx._internal.exporter._errors.GraphConstructionError'>: Error processing Python constants for operator '::Expand'. named_inputs={'input': SymbolicTensor(name='anonymous:132313147898640', producer=anonymous_node:132313170417504, index=0), 'shape': (SymbolicDim(s5), SymbolicDim(s32))}, named_attrs={}, opset=, op_signature=''::Expand(input: T, shape: shape) -> (T) where T=UINT64 | COMPLEX64 | UINT8 | DOUBLE | UINT16 | BFLOAT16 | INT16 | STRING | FLOAT | INT32 | FLOAT16 | UINT32 | COMPLEX128 | INT8 | INT64 | BOOL, shape=INT64.
⬆️
<class 'torch.onnx._internal.exporter._errors.GraphConstructionError'>: Error calling operator 'Expand' with args (SymbolicTensor(name='anonymous:132313147898640', producer=anonymous_node:132313170417504, index=0), (SymbolicDim(s5), SymbolicDim(s32))) and kwargs {}.
⬆️
<class 'torch.onnx._internal.exporter._errors.GraphConstructionError'>: Error when calling function 'TracedOnnxFunction(<function aten_index_put at 0x7856941954e0>)' with args '[SymbolicTensor(name='clone', type=Tensor(FLOAT), shape=Shape([1, 8192, 6]), producer='node_clone', index=0), [None, None, SymbolicTensor(name='index', type=Tensor(INT64), shape=Shape([SymbolicDim(s91)]))], SymbolicTensor(name='update', type=Tensor(FLOAT), shape=Shape([SymbolicDim(s5), SymbolicDim(s32)]))]' and kwargs '{}'
⬆️
<class 'torch.onnx._internal.exporter._errors.ConversionError'>: Error when translating node %index_put : [num_users=1] = call_function[target=torch.ops.aten.index_put.default](args = (%clone, [None, None, %index], %update), kwargs = {}). See the stack trace for more information.
(Refer to the full stack trace above for more information.)
InplaceSetItemMask¶
forward¶
def forward(self, x):
mask = x.to(bool)
x[mask] = 2
return x
custom¶
inputs:
#2[(T1s2x3x3,),(T1s3x3x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "1", "n_large_initializers": "0", "size_initializers": "4", "size_large_initializers": "0", "n_nodes": "2", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s2x3x3[2.0,2.0:A2.0],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3,3] x) => (float[batch,3,3] output_0)
<float c_lifted_tensor_0 = {2}, bool[batch,3,3] to, float[batch,3,3] index_put_, float lift_fresh_copy>
{
[to_dtype] to = Cast <to: int = 9> (x)
[index_put_1b__where] output_0 = Where (to, c_lifted_tensor_0, x)
}
dynamo-ir¶
inputs:
#2[(T1s2x3x3,),(T1s3x3x3,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3,3] x) => (float[batch,3,3] index_put)
<bool[batch,3,3] _to_copy, float clone>
{
[node__to_copy] _to_copy = Cast <to: int = 9> (x)
[node_Constant_0] clone = Constant <value: tensor = float clone {2}> ()
[node_index_put] index_put = Where (_to_copy, clone, x)
}
InplaceSetItemSquare¶
forward¶
def forward(self, x):
x[:2, :3] = 1
return x
custom¶
inputs:
#2[(T1s5x5,),(T1s7x5,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "6", "n_large_initializers": "0", "size_initializers": "88", "size_large_initializers": "0", "n_nodes": "11", "n_nodes_other_domain": "0", "mask_outputs": "[False, True]", "input_args": "(T1s5x5[0.04932647943496704,1.0:A0.5737357687950134],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}", "_known_value_shapes": "{
_onx_gather_x::Shape:: ('batch',),
slice_2::Shape:: (2, 3),
x::Shape:: ('batch', 5),
}"]
>
experiment (float[batch,5] x) => (float[batch,5] output_1)
<int64[1] init7_s1_0 = {0}, int64[1] init7_s1_2 = {2}, int64[1] init7_s1_1 = {1}, int64[2] "init7_s2_-1_1" = {-1,1}, int64[3,1] "init7_s3_0_1_2::RSh-1x1" = {0,1,2}, float[3,2] "fill::T10" = {1,1,1,1,1,1}, float[batch,5] output_0, float[5,2] "slice_3::T10", int64[2] "x::Shape:", float[5,2] "_onx_scatternd_slice_3::T10", int64[?] _onx_range_init7_s1_0, float c_lifted_tensor_0, float clone, int64[1] "_onx_gather_x::Shape:", int64[?,?] "_onx_slice_range_init7_s1_0::RSh-1x1", float[2,5] slice_scatter, float[2,5] slice_3, int64[3] init7_s3_0_1_2, int64[1] init7_s1_3, float[batch,5] slice_scatter_1, float[2,3] fill, float[2,3] slice_2, int64[2] "slice_2::Shape:", int64[?] _onx_slice_range_init7_s1_0, float[2,5] slice_1>
{
[slide_Tensor3] slice_3 = Slice (x, init7_s1_0, init7_s1_2, init7_s1_0)
[slice_scatter_static2] "slice_3::T10" = Transpose <perm: ints = [1, 0]> (slice_3)
[slice_scatter_static4] "_onx_scatternd_slice_3::T10" = ScatterND ("slice_3::T10", "init7_s3_0_1_2::RSh-1x1", "fill::T10")
[slice_scatter_static5] slice_scatter = Transpose <perm: ints = [1, 0]> ("_onx_scatternd_slice_3::T10")
[slice_scatter_dynamic] "x::Shape:" = Shape (x)
[slice_scatter_dynamic2] "_onx_gather_x::Shape:" = Gather ("x::Shape:", init7_s1_0)
[slice_scatter_dynamic3] _onx_range_init7_s1_0 = Range (init7_s1_0, "_onx_gather_x::Shape:", init7_s1_1)
[slice_scatter_dynamic4] _onx_slice_range_init7_s1_0 = Slice (_onx_range_init7_s1_0, init7_s1_0, init7_s1_2, init7_s1_0, init7_s1_1)
[slice_scatter_dynamic5] "_onx_slice_range_init7_s1_0::RSh-1x1" = Reshape (_onx_slice_range_init7_s1_0, "init7_s2_-1_1")
[slice_scatter_dynamic6] output_0 = ScatterND (x, "_onx_slice_range_init7_s1_0::RSh-1x1", slice_scatter)
[".output2"] output_1 = Identity (output_0)
}
dynamo-ir¶
inputs:
#2[(T1s5x5,),(T1s7x5,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,5] x) => (float[batch,5] slice_scatter_1)
<int64 val_0, int64 val_19, int64[1] val_26, int64[1] val_29, int64[1] val_33, float[2,5] slice_3, int64[1] val_42, int64[3,1] val_43, float[3,2] val_44, float[5,2] val_45, float[5,2] val_46, float[2,5] slice_scatter, int64[1] val_47, int64[2] val_48, int64 val_49, int64[?] val_50, int64[1] val_53, int64[?] val_54, int64[?,1] val_55>
{
[node_Constant_0] val_0 = Constant <value: tensor = int64 {0}> ()
[node_Constant_19] val_19 = Constant <value: tensor = int64 {1}> ()
[node_Constant_81] val_26 = Constant <value: tensor = int64[1] val_26 {0}> ()
[node_Constant_84] val_29 = Constant <value: tensor = int64[1] val_29 {2}> ()
[node_Constant_33] val_33 = Constant <value_ints: ints = [1]> ()
[node_slice_3] slice_3 = Slice (x, val_26, val_29, val_26, val_33)
[node_Constant_42] val_42 = Constant <value: tensor = int64[1] {-1}> ()
[node_Constant_95] val_43 = Constant <value: tensor = int64[3,1] val_43 {0,1,2}> ()
[node_Constant_96] val_44 = Constant <value: tensor = float[3,2] val_44 {1,1,1,1,1,1}> ()
[node_Transpose_45] val_45 = Transpose <perm: ints = [1, 0]> (slice_3)
[node_ScatterND_46] val_46 = ScatterND <reduction: string = "none"> (val_45, val_43, val_44)
[node_slice_scatter] slice_scatter = Transpose <perm: ints = [1, 0]> (val_46)
[node_Constant_47] val_47 = Constant <value_ints: ints = [0]> ()
[node_Shape_48] val_48 = Shape <start: int = 0> (x)
[node_Gather_49] val_49 = Gather <axis: int = 0> (val_48, val_0)
[node_Range_50] val_50 = Range (val_0, val_49, val_19)
[node_Constant_99] val_53 = Constant <value: tensor = int64[1] val_53 {1}> ()
[node_Slice_54] val_54 = Slice (val_50, val_26, val_29, val_47, val_53)
[node_Unsqueeze_55] val_55 = Unsqueeze (val_54, val_42)
[node_slice_scatter_1] slice_scatter_1 = ScatterND <reduction: string = "none"> (x, val_55, slice_scatter)
}
InplaceSetItemSquareAdd¶
forward¶
def forward(self, x):
x[:2, :3] = 1
return x + 2
custom¶
inputs:
#2[(T1s5x5,),(T1s7x5,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "7", "n_large_initializers": "0", "size_initializers": "92", "size_large_initializers": "0", "n_nodes": "11", "n_nodes_other_domain": "0", "mask_outputs": "[False, True]", "input_args": "(T1s5x5[0.06415653228759766,1.0:A0.6122418832778931],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}", "_known_value_shapes": "{
_onx_gather_x::Shape:: ('batch',),
slice_2::Shape:: (2, 3),
x::Shape:: ('batch', 5),
}"]
>
experiment (float[batch,5] x) => (float[batch,5] output_1)
<int64[1] init7_s1_0 = {0}, int64[1] init7_s1_2 = {2}, int64[1] init7_s1_1 = {1}, int64[2] "init7_s2_-1_1" = {-1,1}, int64[3,1] "init7_s3_0_1_2::RSh-1x1" = {0,1,2}, float[3,2] "fill::T10" = {1,1,1,1,1,1}, float[1] "init1_s_::RSh1" = {2}, float[batch,5] output_0, float[5,2] "slice_3::T10", float init1_s_, int64[2] "x::Shape:", float[5,2] "_onx_scatternd_slice_3::T10", int64[?] _onx_range_init7_s1_0, float c_lifted_tensor_0, float clone, int64[1] "_onx_gather_x::Shape:", int64[?,?] "_onx_slice_range_init7_s1_0::RSh-1x1", float[2,5] slice_scatter, float[2,5] slice_3, float[batch,5] add, int64[3] init7_s3_0_1_2, int64[1] init7_s1_3, float[batch,5] slice_scatter_1, float[2,3] fill, float[2,3] slice_2, int64[2] "slice_2::Shape:", int64[?] _onx_slice_range_init7_s1_0, float[2,5] slice_1>
{
[slide_Tensor3] slice_3 = Slice (x, init7_s1_0, init7_s1_2, init7_s1_0)
[slice_scatter_static2] "slice_3::T10" = Transpose <perm: ints = [1, 0]> (slice_3)
[slice_scatter_static4] "_onx_scatternd_slice_3::T10" = ScatterND ("slice_3::T10", "init7_s3_0_1_2::RSh-1x1", "fill::T10")
[slice_scatter_static5] slice_scatter = Transpose <perm: ints = [1, 0]> ("_onx_scatternd_slice_3::T10")
[slice_scatter_dynamic] "x::Shape:" = Shape (x)
[slice_scatter_dynamic2] "_onx_gather_x::Shape:" = Gather ("x::Shape:", init7_s1_0)
[slice_scatter_dynamic3] _onx_range_init7_s1_0 = Range (init7_s1_0, "_onx_gather_x::Shape:", init7_s1_1)
[slice_scatter_dynamic4] _onx_slice_range_init7_s1_0 = Slice (_onx_range_init7_s1_0, init7_s1_0, init7_s1_2, init7_s1_0, init7_s1_1)
[slice_scatter_dynamic5] "_onx_slice_range_init7_s1_0::RSh-1x1" = Reshape (_onx_slice_range_init7_s1_0, "init7_s2_-1_1")
[slice_scatter_dynamic6] output_0 = ScatterND (x, "_onx_slice_range_init7_s1_0::RSh-1x1", slice_scatter)
[add_Tensor] output_1 = Add (output_0, "init1_s_::RSh1")
}
dynamo-ir¶
inputs:
#2[(T1s5x5,),(T1s7x5,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,5] x) => (float[batch,5] add)
<int64 val_0, int64 val_19, int64[1] val_26, int64[1] val_29, int64[1] val_33, float[2,5] slice_3, int64[1] val_42, int64[3,1] val_43, float[3,2] val_44, float[5,2] val_45, float[5,2] val_46, float[2,5] slice_scatter, int64[1] val_47, int64[2] val_48, int64 val_49, int64[?] val_50, int64[1] val_53, int64[?] val_54, int64[?,1] val_55, float[batch,5] slice_scatter_1, float scalar_tensor_default>
{
[node_Constant_0] val_0 = Constant <value: tensor = int64 {0}> ()
[node_Constant_19] val_19 = Constant <value: tensor = int64 {1}> ()
[node_Constant_81] val_26 = Constant <value: tensor = int64[1] val_26 {0}> ()
[node_Constant_84] val_29 = Constant <value: tensor = int64[1] val_29 {2}> ()
[node_Constant_33] val_33 = Constant <value_ints: ints = [1]> ()
[node_slice_3] slice_3 = Slice (x, val_26, val_29, val_26, val_33)
[node_Constant_42] val_42 = Constant <value: tensor = int64[1] {-1}> ()
[node_Constant_95] val_43 = Constant <value: tensor = int64[3,1] val_43 {0,1,2}> ()
[node_Constant_96] val_44 = Constant <value: tensor = float[3,2] val_44 {1,1,1,1,1,1}> ()
[node_Transpose_45] val_45 = Transpose <perm: ints = [1, 0]> (slice_3)
[node_ScatterND_46] val_46 = ScatterND <reduction: string = "none"> (val_45, val_43, val_44)
[node_slice_scatter] slice_scatter = Transpose <perm: ints = [1, 0]> (val_46)
[node_Constant_47] val_47 = Constant <value_ints: ints = [0]> ()
[node_Shape_48] val_48 = Shape <start: int = 0> (x)
[node_Gather_49] val_49 = Gather <axis: int = 0> (val_48, val_0)
[node_Range_50] val_50 = Range (val_0, val_49, val_19)
[node_Constant_99] val_53 = Constant <value: tensor = int64[1] val_53 {1}> ()
[node_Slice_54] val_54 = Slice (val_50, val_26, val_29, val_47, val_53)
[node_Unsqueeze_55] val_55 = Unsqueeze (val_54, val_42)
[node_slice_scatter_1] slice_scatter_1 = ScatterND <reduction: string = "none"> (x, val_55, slice_scatter)
[node_Constant_100] scalar_tensor_default = Constant <value: tensor = float scalar_tensor_default {2}> ()
[node_add] add = Add (slice_scatter_1, scalar_tensor_default)
}
InplaceSetItemSquareAdd2¶
forward¶
def forward(self, x):
x[:2, :3] = 1
return x + 2, x + 3
custom¶
inputs:
#2[(T1s5x5,),(T1s7x5,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "8", "n_large_initializers": "0", "size_initializers": "96", "size_large_initializers": "0", "n_nodes": "12", "n_nodes_other_domain": "0", "mask_outputs": "[False, True, True]", "input_args": "(T1s5x5[0.011756062507629395,1.0:A0.6521409487724305],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}", "_known_value_shapes": "{
_onx_gather_x::Shape:: ('batch',),
slice_2::Shape:: (2, 3),
x::Shape:: ('batch', 5),
}"]
>
experiment (float[batch,5] x) => (float[batch,5] output_1, float[batch,5] output_2)
<int64[1] init7_s1_0 = {0}, int64[1] init7_s1_2 = {2}, int64[1] init7_s1_1 = {1}, int64[2] "init7_s2_-1_1" = {-1,1}, int64[3,1] "init7_s3_0_1_2::RSh-1x1" = {0,1,2}, float[3,2] "fill::T10" = {1,1,1,1,1,1}, float[1] "init1_s_::RSh1" = {2}, float[1] "init1_s_2::RSh1" = {3}, float[batch,5] output_0, float[5,2] "slice_3::T10", float init1_s_, int64[2] "x::Shape:", float[5,2] "_onx_scatternd_slice_3::T10", int64[?] _onx_range_init7_s1_0, float c_lifted_tensor_0, float clone, int64[1] "_onx_gather_x::Shape:", int64[?,?] "_onx_slice_range_init7_s1_0::RSh-1x1", float[2,5] slice_scatter, float[2,5] slice_3, float[batch,5] add, float[batch,5] add_4, int64[3] init7_s3_0_1_2, int64[1] init7_s1_3, float[batch,5] slice_scatter_1, float[2,3] fill, float[2,3] slice_2, int64[2] "slice_2::Shape:", int64[?] _onx_slice_range_init7_s1_0, float init1_s_2, float[2,5] slice_1>
{
[slide_Tensor3] slice_3 = Slice (x, init7_s1_0, init7_s1_2, init7_s1_0)
[slice_scatter_static2] "slice_3::T10" = Transpose <perm: ints = [1, 0]> (slice_3)
[slice_scatter_static4] "_onx_scatternd_slice_3::T10" = ScatterND ("slice_3::T10", "init7_s3_0_1_2::RSh-1x1", "fill::T10")
[slice_scatter_static5] slice_scatter = Transpose <perm: ints = [1, 0]> ("_onx_scatternd_slice_3::T10")
[slice_scatter_dynamic] "x::Shape:" = Shape (x)
[slice_scatter_dynamic2] "_onx_gather_x::Shape:" = Gather ("x::Shape:", init7_s1_0)
[slice_scatter_dynamic3] _onx_range_init7_s1_0 = Range (init7_s1_0, "_onx_gather_x::Shape:", init7_s1_1)
[slice_scatter_dynamic4] _onx_slice_range_init7_s1_0 = Slice (_onx_range_init7_s1_0, init7_s1_0, init7_s1_2, init7_s1_0, init7_s1_1)
[slice_scatter_dynamic5] "_onx_slice_range_init7_s1_0::RSh-1x1" = Reshape (_onx_slice_range_init7_s1_0, "init7_s2_-1_1")
[slice_scatter_dynamic6] output_0 = ScatterND (x, "_onx_slice_range_init7_s1_0::RSh-1x1", slice_scatter)
[add_Tensor] output_1 = Add (output_0, "init1_s_::RSh1")
[add_Tensor2] output_2 = Add (output_0, "init1_s_2::RSh1")
}
dynamo-ir¶
inputs:
#2[(T1s5x5,),(T1s7x5,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,5] x) => (float[batch,5] add, float[batch,5] add_4)
<int64 val_0, int64 val_19, int64[1] val_26, int64[1] val_29, int64[1] val_33, float[2,5] slice_3, int64[1] val_42, int64[3,1] val_43, float[3,2] val_44, float[5,2] val_45, float[5,2] val_46, float[2,5] slice_scatter, int64[1] val_47, int64[2] val_48, int64 val_49, int64[?] val_50, int64[1] val_53, int64[?] val_54, int64[?,1] val_55, float[batch,5] slice_scatter_1, float scalar_tensor_default, float scalar_tensor_default_1>
{
[node_Constant_0] val_0 = Constant <value: tensor = int64 {0}> ()
[node_Constant_19] val_19 = Constant <value: tensor = int64 {1}> ()
[node_Constant_81] val_26 = Constant <value: tensor = int64[1] val_26 {0}> ()
[node_Constant_84] val_29 = Constant <value: tensor = int64[1] val_29 {2}> ()
[node_Constant_33] val_33 = Constant <value_ints: ints = [1]> ()
[node_slice_3] slice_3 = Slice (x, val_26, val_29, val_26, val_33)
[node_Constant_42] val_42 = Constant <value: tensor = int64[1] {-1}> ()
[node_Constant_95] val_43 = Constant <value: tensor = int64[3,1] val_43 {0,1,2}> ()
[node_Constant_96] val_44 = Constant <value: tensor = float[3,2] val_44 {1,1,1,1,1,1}> ()
[node_Transpose_45] val_45 = Transpose <perm: ints = [1, 0]> (slice_3)
[node_ScatterND_46] val_46 = ScatterND <reduction: string = "none"> (val_45, val_43, val_44)
[node_slice_scatter] slice_scatter = Transpose <perm: ints = [1, 0]> (val_46)
[node_Constant_47] val_47 = Constant <value_ints: ints = [0]> ()
[node_Shape_48] val_48 = Shape <start: int = 0> (x)
[node_Gather_49] val_49 = Gather <axis: int = 0> (val_48, val_0)
[node_Range_50] val_50 = Range (val_0, val_49, val_19)
[node_Constant_99] val_53 = Constant <value: tensor = int64[1] val_53 {1}> ()
[node_Slice_54] val_54 = Slice (val_50, val_26, val_29, val_47, val_53)
[node_Unsqueeze_55] val_55 = Unsqueeze (val_54, val_42)
[node_slice_scatter_1] slice_scatter_1 = ScatterND <reduction: string = "none"> (x, val_55, slice_scatter)
[node_Constant_100] scalar_tensor_default = Constant <value: tensor = float scalar_tensor_default {2}> ()
[node_add] add = Add (slice_scatter_1, scalar_tensor_default)
[node_Constant_101] scalar_tensor_default_1 = Constant <value: tensor = float scalar_tensor_default_1 {3}> ()
[node_add_4] add_4 = Add (slice_scatter_1, scalar_tensor_default_1)
}
SignatureFloat1¶
forward¶
def forward(self, x, alpha: float = 2.0):
return torch.sigmoid(self.linear(x)) - self.buff * alpha
custom¶
inputs:
#2[(T1s4x3,float),(T1s8x3,float)]shapes:
({0:Dim(batch)},None)
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "3", "n_large_initializers": "0", "size_initializers": "20", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x3[10.0,21.0:A15.5],float=1.5)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "({0: Dim('batch', min=1, max=1024)}, None)", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3] x, float alpha) => (float[batch,1] output_0)
<float[1] mul = {0.75}, float[1,3] "GemmTransposePattern--p_linear_weight::T10" = {-0.488475,0.0411319,-0.2934}, float[1] "linear.bias" = {0.322306}, float[batch,1] _onx_matmul_x, int64[1] init7_s1_1, float init1_s_, float[1] b_buff, float[1,3] p_linear_weight, float[batch,1] linear, float[1] "init1_s_::RSh1", int64[2] "init7_s2_1_-1", float[batch,1] sigmoid, float[batch,1] sub, float[1,3] "linear.weight", float[1] _onx_mul_b_buff, float[1] p_linear_bias, float[3,1] "p_linear_weight::T10">
{
["GemmTransposePattern--MatMulAddPattern--Opset2"] linear = Gemm <transB: int = 1> (x, "GemmTransposePattern--p_linear_weight::T10", "linear.bias")
[sigmoid] sigmoid = Sigmoid (linear)
[sub_Tensor] output_0 = Sub (sigmoid, mul)
}
FAILED
diff.1
dynamo-ir¶
inputs:
#2[(T1s4x3,float),(T1s8x3,float)]shapes:
({0:Dim(batch)},None)
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[s77,3] x) => (float[s77,1] sub_2)
<float[1,3] "linear.weight" = {0.442887,-0.540267,0.247224}, float[1] "linear.bias" = {-0.530752}, float[1,3] "linear.weight", float[1] "linear.bias", float[s77,1] linear, float[s77,1] sigmoid, float[1] mul_2>
{
[node_linear] linear = Gemm <beta: float = 1, transB: int = 1, alpha: float = 1, transA: int = 0> (x, "linear.weight", "linear.bias")
[node_sigmoid] sigmoid = Sigmoid (linear)
[node_Constant_1] mul_2 = Constant <value: tensor = float[1] mul_2 {0.75}> ()
[node_sub_2] sub_2 = Sub (sigmoid, mul_2)
}
FAILED
Input mismatch, inputs[0]=(T1r2,float) but names=['x'], model=SignatureFloat1, export='dynamo-ir'
SignatureInt1¶
forward¶
def forward(self, x, i: int = 2):
return torch.sigmoid(self.linear(x)) - self.buff + x[:, i : i + 1]
custom¶
inputs:
#2[(T1s4x3,int),(T1s8x3,int)]shapes:
({0:Dim(batch)},None)
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "5", "n_large_initializers": "0", "size_initializers": "36", "size_large_initializers": "0", "n_nodes": "5", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x3[10.0,21.0:A15.5],int=1)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "({0: Dim('batch', min=1, max=1024)}, None)", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3] x, int64 i) => (float[batch,1] output_0)
<float[1] b_buff = {0.5}, int64[1] init7_s1_1 = {1}, int64[1] init7_s1_2 = {2}, float[1,3] "GemmTransposePattern--p_linear_weight::T10" = {-0.287984,0.108205,0.00799918}, float[1] "linear.bias" = {-0.153188}, float[batch,1] _onx_matmul_x, float[batch,1] add, float[1,3] p_linear_weight, float[batch,1] linear, int64[2] "init7_s2_1_-1", float[batch,1] sigmoid, float[batch,1] slice_2, float[batch,1] sub, float[1,3] "linear.weight", float[1] p_linear_bias, float[3,1] "p_linear_weight::T10", float[batch,3] slice_1>
{
["GemmTransposePattern--MatMulAddPattern--Opset2"] linear = Gemm <transB: int = 1> (x, "GemmTransposePattern--p_linear_weight::T10", "linear.bias")
[sigmoid] sigmoid = Sigmoid (linear)
[sub_Tensor] sub = Sub (sigmoid, b_buff)
[slide_Tensor] slice_2 = Slice (x, init7_s1_1, init7_s1_2, init7_s1_1)
[add_Tensor] output_0 = Add (sub, slice_2)
}
FAILED
diff.1
dynamo-ir¶
inputs:
#2[(T1s4x3,int),(T1s8x3,int)]shapes:
({0:Dim(batch)},None)
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[s77,3] x) => (float[s77,1] add_15)
<float[1,3] "linear.weight" = {0.303618,-0.0922485,-0.57155}, float[1] "linear.bias" = {0.0983027}, float[1] buff = {0.5}, float[1,3] "linear.weight", float[1] "linear.bias", float[1] buff, float[s77,1] linear, float[s77,1] sigmoid, float[s77,1] sub_2, int64[1] val_10, int64[1] val_14, int64[1] val_18, float[s77,1] slice_2>
{
[node_linear] linear = Gemm <beta: float = 1, transB: int = 1, alpha: float = 1, transA: int = 0> (x, "linear.weight", "linear.bias")
[node_sigmoid] sigmoid = Sigmoid (linear)
[node_sub_2] sub_2 = Sub (sigmoid, buff)
[node_Constant_24] val_10 = Constant <value: tensor = int64[1] val_10 {1}> ()
[node_Constant_27] val_14 = Constant <value: tensor = int64[1] val_14 {2}> ()
[node_Constant_18] val_18 = Constant <value_ints: ints = [1]> ()
[node_slice_2] slice_2 = Slice (x, val_10, val_14, val_10, val_18)
[node_add_15] add_15 = Add (sub_2, slice_2)
}
FAILED
Input mismatch, inputs[0]=(T1r2,int) but names=['x'], model=SignatureInt1, export='dynamo-ir'
SignatureInt2¶
forward¶
def forward(self, x, i: int = 2):
return torch.sigmoid(self.linear(x)) - self.buff + x[:, i]
custom¶
inputs:
#1[(T1s4x3,int)]shapes:
dict(x:{0:Dim(batch)},i:None)
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "4", "n_large_initializers": "0", "size_initializers": "28", "size_large_initializers": "0", "n_nodes": "5", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x3[10.0,21.0:A15.5],int=1)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'i': None, 'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,3] x, int64 i) => (float[batch,batch] output_0)
<float[1] b_buff = {0.5}, int64 init7_s_1 = {1}, float[1,3] "GemmTransposePattern--p_linear_weight::T10" = {-0.430331,0.400767,0.422815}, float[1] "linear.bias" = {-0.433012}, float[batch,1] _onx_matmul_x, float[batch] select, float[batch,batch] add, float[1,3] p_linear_weight, float[batch,1] linear, int64[2] "init7_s2_1_-1", float[batch,1] sigmoid, float[batch,1] sub, float[1,3] "linear.weight", float[1] p_linear_bias, float[3,1] "p_linear_weight::T10", float[batch,3] slice_1>
{
["GemmTransposePattern--MatMulAddPattern--Opset2"] linear = Gemm <transB: int = 1> (x, "GemmTransposePattern--p_linear_weight::T10", "linear.bias")
[sigmoid] sigmoid = Sigmoid (linear)
[sub_Tensor] sub = Sub (sigmoid, b_buff)
[select_int] select = Gather <axis: int = 1> (x, init7_s_1)
[add_Tensor] output_0 = Add (sub, select)
}
dynamo-ir¶
inputs:
#1[(T1s4x3,int)]shapes:
dict(x:{0:Dim(batch)},i:None)
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[s77,3] x) => (float[s77,s77] add_14)
<float[1,3] "linear.weight" = {-0.47754,-0.0452508,-0.016264}, float[1] "linear.bias" = {-0.496291}, float[1] buff = {0.5}, float[1,3] "linear.weight", float[1] "linear.bias", float[1] buff, float[s77,1] linear, float[s77,1] sigmoid, float[s77,1] sub_2, int64 val_7, float[s77] select>
{
[node_linear] linear = Gemm <beta: float = 1, transB: int = 1, alpha: float = 1, transA: int = 0> (x, "linear.weight", "linear.bias")
[node_sigmoid] sigmoid = Sigmoid (linear)
[node_sub_2] sub_2 = Sub (sigmoid, buff)
[node_Constant_7] val_7 = Constant <value: tensor = int64 {1}> ()
[node_select] select = Gather <axis: int = 1> (x, val_7)
[node_add_14] add_14 = Add (sub_2, select)
}
FAILED
Input mismatch, inputs[0]=(T1r2,int) but names=['x'], model=SignatureInt2, export='dynamo-ir'
SignatureListFixedLength¶
forward¶
def forward(self, x, lx: list):
return (
torch.sigmoid(self.linear(x)) - self.buff + lx[0] * lx[1].sum(axis=1, keepdim=True)
)
custom¶
inputs:
#2[(T1s4x3,#2[T1s4x1,T1s4x2]),(T1s8x3,#2[T1s8x1,T1s8x2])]shapes:
dict(x:{0:Dim(batch)},lx:#2[{0:Dim(batch)},{0:Dim(batch)}])
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "4", "n_large_initializers": "0", "size_initializers": "28", "size_large_initializers": "0", "n_nodes": "6", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x3[10.0,21.0:A15.5],#2[T1s4x1[10.0,13.0:A11.5],T1s4x2[10.0,17.0:A13.5]])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'lx': [{0: Dim('batch', min=0)}, {0: Dim('batch', min=0)}],
'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s53', 's50', 's77'},
's50': {'batch', 's53'},
's53': {'batch'},
's77': {'batch'}}"]
>
experiment (float[batch,3] x, float[batch,1] lx_0, float[batch,2] lx_1) => (float[batch,1] output_0)
<float[1] b_buff = {0.5}, int64[1] init7_s1_1 = {1}, float[1,3] "GemmTransposePattern--p_linear_weight::T10" = {-0.405241,-0.17025,-0.320338}, float[1] "linear.bias" = {0.380358}, float[batch,1] _onx_matmul_x, float[batch,1] add, float[batch,1] mul, float[batch,1] sum_1, float[1,3] p_linear_weight, float[batch,1] linear, int64[2] "init7_s2_1_-1", float[batch,1] sigmoid, float[batch,1] sub, float[1,3] "linear.weight", float[1] p_linear_bias, float[3,1] "p_linear_weight::T10">
{
["GemmTransposePattern--MatMulAddPattern--Opset2"] linear = Gemm <transB: int = 1> (x, "GemmTransposePattern--p_linear_weight::T10", "linear.bias")
[sigmoid] sigmoid = Sigmoid (linear)
[sub_Tensor] sub = Sub (sigmoid, b_buff)
[sum] sum_1 = ReduceSum <keepdims: int = 1> (lx_1, init7_s1_1)
[mul_Tensor] mul = Mul (lx_0, sum_1)
[add_Tensor] output_0 = Add (sub, mul)
}
dynamo-ir¶
inputs:
#2[(T1s4x3,#2[T1s4x1,T1s4x2]),(T1s8x3,#2[T1s8x1,T1s8x2])]shapes:
dict(x:{0:Dim(batch)},lx:#2[{0:Dim(batch)},{0:Dim(batch)}])
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x, float[batch,1] lx_0, float[batch,2] lx_1) => (float[batch,1] add_15)
<float[1,3] "linear.weight" = {0.296968,0.325305,0.272611}, float[1] "linear.bias" = {0.254532}, float[1] buff = {0.5}, float[1,3] "linear.weight", float[1] "linear.bias", float[1] buff, float[batch,1] linear, float[batch,1] sigmoid, float[batch,1] sub_2, int64[1] val_6, float[batch,1] sum_1, float[batch,1] mul_4>
{
[node_linear] linear = Gemm <beta: float = 1, transB: int = 1, alpha: float = 1, transA: int = 0> (x, "linear.weight", "linear.bias")
[node_sigmoid] sigmoid = Sigmoid (linear)
[node_sub_2] sub_2 = Sub (sigmoid, buff)
[node_Constant_9] val_6 = Constant <value: tensor = int64[1] val_6 {1}> ()
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 1> (lx_1, val_6)
[node_mul_4] mul_4 = Mul (lx_0, sum_1)
[node_add_15] add_15 = Add (sub_2, mul_4)
}
SignatureListFixedWithNone¶
forward¶
def forward(self, lx):
x = lx[0]
if lx[1] is not None:
x += lx[1]
if lx[2] is not None:
x += lx[2]
return x
custom¶
FAILED
Detected mismatch between the structure of `inputs` and `dynamic_shapes`: `inputs['lx']` has 3 elements, but `dynamic_shapes['lx']` has 2 elements
For more information about this error, see: https://pytorch.org/docs/main/generated/exportdb/index.html#dynamic-shapes-validation
The error above occurred when calling torch.export.export. If you would like to view some more information about this error, and get a list of all other errors that may occur in your export call, you can replace your `export()` call with `draft_export()`.
dynamo-ir¶
FAILED
Failed to export the model with torch.export. [96mThis is step 1/3[0m of exporting the model to ONNX. Next steps:
- Modify the model code for `torch.export.export` to succeed. Refer to https://pytorch.org/docs/stable/generated/exportdb/index.html for more information.
- Debug `torch.export.export` and summit a PR to PyTorch.
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
## Exception summary
<class 'torch._dynamo.exc.UserError'>: Detected mismatch between the structure of `inputs` and `dynamic_shapes`: `inputs['lx']` has 3 elements, but `dynamic_shapes['lx']` has 2 elements
For more information about this error, see: https://pytorch.org/docs/main/generated/exportdb/index.html#dynamic-shapes-validation
The error above occurred when calling torch.export.export. If you would like to view some more information about this error, and get a list of all other errors that may occur in your export call, you can replace your `export()` call with `draft_export()`.
(Refer to the full stack trace above for more information.)
SignatureListVariableLength¶
forward¶
def forward(self, x, lx: list):
t = torch.cat(lx, dim=1).sum(axis=1, keepdim=True)
return torch.sigmoid(self.linear(x)) - self.buff + t
custom¶
inputs:
#2[(T1s4x3,#2[T1s4x1,T1s4x2]),(T1s8x3,#3[T1s8x1,T1s8x2,T1s8x3])]shapes:
dict(x:{0:Dim(batch)},lx:#2[{0:Dim(batch)},{0:Dim(batch)}])
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "4", "n_large_initializers": "0", "size_initializers": "28", "size_large_initializers": "0", "n_nodes": "6", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x3[10.0,21.0:A15.5],#2[T1s4x1[10.0,13.0:A11.5],T1s4x2[10.0,17.0:A13.5]])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'lx': [{0: Dim('batch', min=0)}, {0: Dim('batch', min=0)}],
'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s53', 's50', 's77'},
's50': {'batch'},
's53': {'batch'},
's77': {'batch'}}"]
>
experiment (float[batch,3] x, float[batch,1] lx_0, float[batch,2] lx_1) => (float[batch,1] output_0)
<float[1] b_buff = {0.5}, int64[1] init7_s1_1 = {1}, float[1,3] "GemmTransposePattern--p_linear_weight::T10" = {0.0289861,-0.0736461,0.263717}, float[1] "linear.bias" = {-0.320988}, float[batch,1] _onx_matmul_x, float[batch,1] add, float[batch,1] sum_1, float[1,3] p_linear_weight, float[batch,1] linear, int64[2] "init7_s2_1_-1", float[batch,1] sigmoid, float[batch,3] cat, float[batch,1] sub, float[1,3] "linear.weight", float[1] p_linear_bias, float[3,1] "p_linear_weight::T10">
{
[cat] cat = Concat <axis: int = 1> (lx_0, lx_1)
[sum] sum_1 = ReduceSum <keepdims: int = 1> (cat, init7_s1_1)
["GemmTransposePattern--MatMulAddPattern--Opset2"] linear = Gemm <transB: int = 1> (x, "GemmTransposePattern--p_linear_weight::T10", "linear.bias")
[sigmoid] sigmoid = Sigmoid (linear)
[sub_Tensor] sub = Sub (sigmoid, b_buff)
[add_Tensor] output_0 = Add (sub, sum_1)
}
FAILED
diff.1
dynamo-ir¶
inputs:
#2[(T1s4x3,#2[T1s4x1,T1s4x2]),(T1s8x3,#3[T1s8x1,T1s8x2,T1s8x3])]shapes:
dict(x:{0:Dim(batch)},lx:#2[{0:Dim(batch)},{0:Dim(batch)}])
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x, float[batch,1] lx_0, float[batch,2] lx_1) => (float[batch,1] add_15)
<float[1,3] "linear.weight" = {-0.155996,-0.00216091,-0.157249}, float[1] "linear.bias" = {-0.312999}, float[1] buff = {0.5}, float[1,3] "linear.weight", float[1] "linear.bias", float[1] buff, float[batch,3] cat, int64[1] val_6, float[batch,1] sum_1, float[batch,1] linear, float[batch,1] sigmoid, float[batch,1] sub_4>
{
[node_cat] cat = Concat <axis: int = 1> (lx_0, lx_1)
[node_Constant_9] val_6 = Constant <value: tensor = int64[1] val_6 {1}> ()
[node_sum_1] sum_1 = ReduceSum <noop_with_empty_axes: int = 0, keepdims: int = 1> (cat, val_6)
[node_linear] linear = Gemm <beta: float = 1, transB: int = 1, alpha: float = 1, transA: int = 0> (x, "linear.weight", "linear.bias")
[node_sigmoid] sigmoid = Sigmoid (linear)
[node_sub_4] sub_4 = Sub (sigmoid, buff)
[node_add_15] add_15 = Add (sub_4, sum_1)
}
FAILED
diff.1
SignatureShapeAsIndex¶
forward¶
def forward(self, x, y):
t = torch.sigmoid(self.linear(x)) + x
return t[:, : y.shape[1]]
custom¶
inputs:
#1[(T1s4x3,T1s4x2)]shapes:
dict(x:{0:Dim(batch)},y:{0:Dim(batch),1:Dim(length)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "4", "n_large_initializers": "0", "size_initializers": "32", "size_large_initializers": "0", "n_nodes": "5", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x3[10.0,21.0:A15.5],T1s4x2[10.0,17.0:A13.5])", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0, max=1024)},
'y': {0: Dim('batch', min=0, max=1024), 1: Dim('length', min=0, max=2)}}", "_discovered_shape_constraints": "{'batch': {'s17', 's77'},
'length': {'s94'},
's17': {'batch'},
's77': {'batch'},
's94': {'length'}}", "_known_value_shapes": "{
sym_size_int_3: length,
sym_size_int_3::UnSq0: ('length',),
y::Shape1:2: ('length',),
}"]
>
experiment (float[batch,3] x, float[batch,length] y) => (float[batch,length] output_0)
<int64[1] init7_s1_0 = {0}, int64[1] init7_s1_1 = {1}, float[1,3] "GemmTransposePattern--p_linear_weight::T10" = {-0.226019,-0.247899,0.459164}, float[1] "linear.bias" = {-0.5307}, float[batch,1] _onx_matmul_x, int64 sym_size_int_3, int64[1] "sym_size_int_3::UnSq0", float[batch,3] add, float[1] b_buff, float[1,3] p_linear_weight, float[batch,1] linear, int64[2] "init7_s2_1_-1", int64[1] "y::Shape1:2", float[batch,1] sigmoid, float[batch,length] slice_2, float[1,3] "linear.weight", float[1] p_linear_bias, float[3,1] "p_linear_weight::T10", float[batch,3] slice_1>
{
[sym_size_int] "y::Shape1:2" = Shape <end: int = 2, start: int = 1> (y)
["GemmTransposePattern--MatMulAddPattern--Opset2"] linear = Gemm <transB: int = 1> (x, "GemmTransposePattern--p_linear_weight::T10", "linear.bias")
[sigmoid] sigmoid = Sigmoid (linear)
[add_Tensor] add = Add (sigmoid, x)
[slide_Tensor] output_0 = Slice (add, init7_s1_0, "y::Shape1:2", init7_s1_1)
}
dynamo-ir¶
inputs:
#1[(T1s4x3,T1s4x2)]shapes:
dict(x:{0:Dim(batch)},y:{0:Dim(batch),1:Dim(length)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,3] x, float[batch,length] y) => (float[batch,length] slice_2)
<float[1,3] "linear.weight" = {0.242793,0.552866,0.0156701}, float[1] "linear.bias" = {-0.41803}, float[1,3] "linear.weight", float[1] "linear.bias", int64[1] val_0, float[batch,1] linear, float[batch,1] sigmoid, float[batch,3] add_6, int64[1] val_8, int64[1] val_15, int64[1] val_16>
{
[node_Shape_0] val_0 = Shape <end: int = 2, start: int = 1> (y)
[node_linear] linear = Gemm <beta: float = 1, transB: int = 1, alpha: float = 1, transA: int = 0> (x, "linear.weight", "linear.bias")
[node_sigmoid] sigmoid = Sigmoid (linear)
[node_add_6] add_6 = Add (sigmoid, x)
[node_Constant_8] val_8 = Constant <value_ints: ints = [0]> ()
[node_Constant_23] val_15 = Constant <value: tensor = int64[1] val_15 {1}> ()
[node_Constant_16] val_16 = Constant <value_ints: ints = [1]> ()
[node_slice_2] slice_2 = Slice (add_6, val_8, val_0, val_15, val_16)
}
TypeBFloat16¶
forward¶
def forward(self, x):
xb = x.to(torch.bfloat16)
return (xb + xb).to(torch.float32)
custom¶
inputs:
#1[(T1s4x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 8,
opset_import: ["" : 18],
metadata_props: ["large_model": "False", "inline": "True", "external_threshold": "1024", "function_options": "FunctionOptions()", "optimize": "True", "n_initializers": "0", "n_large_initializers": "0", "size_initializers": "0", "size_large_initializers": "0", "n_nodes": "3", "n_nodes_other_domain": "0", "mask_outputs": "[True]", "input_args": "(T1s4x4[0.17821824550628662,0.9225718379020691:A0.5503441654145718],)", "input_kwargs": "None", "optimizations": "OptimizationOptions(constant_folding={'Transpose', 'Cast', 'Reshape', 'Concat'}, patterns=[BatchNormalizationPattern(), BatchNormalizationTrainingPattern(), CastLayerNormalizationCastPattern(), CastPattern(), CastCastBinaryPattern(), CastOpCastPattern(), ClipClipPattern(), ComputationCastOpCastPattern(), ConcatGatherPattern(), ConvBiasNullPattern(), DropoutPattern(), ExpandPattern(), ExpandBroadcastPattern(), ExpandSwapPattern(), GeluPattern(), IdentityPattern(), LayerNormalizationPattern(), LayerNormalizationScalePattern(), LeakyReluPattern(), MulMulMulScalarPattern(), ReduceReshapePattern(), ReduceSumNormalizePattern(), ReshapePattern(), ReshapeMatMulReshapePattern(), Reshape2Of3Pattern(), ReshapeReshapeBinaryPattern(), MatMulAddPattern(), GemmTransposePattern(), MatMulReshape2Of3Pattern(), MulMulMatMulPattern(), ReshapeReshapePattern(), RotaryConcatPartPattern(), SameChildrenPattern(), SequenceConstructAtPattern(), SliceSlicePattern(), SlicesSplitPattern(), SoftmaxCrossEntropyLossCastPattern(), SplitConcatPattern(), SqueezeAddPattern(), SqueezeUnsqueezePattern(), Sub1MulPattern(), SwitchOrderBinaryPattern(), SwitchReshapeActivationPattern(), TransposeEqualReshapePattern(), TransposeMatMulPattern(), TransposeReshapeMatMulPattern(), TransposeReshapeTransposePattern(), TransposeTransposePattern(), UnsqueezeEqualPattern(), UnsqueezeUnsqueezePattern()])", "dynamic_shapes": "{'x': {0: Dim('batch', min=0)}}", "_discovered_shape_constraints": "{'batch': {'s77'}, 's77': {'batch'}}"]
>
experiment (float[batch,4] x) => (float[batch,4] output_0)
<float[batch,4] "add-x", bfloat16[batch,4] to, float[batch,4] to_1, bfloat16[batch,4] add>
{
["CastCastBinaryPattern--add_Tensor"] "add-x" = Add (x, x)
["CastCastBinaryPattern--add_Tensor2"] add = Cast <to: int = 16> ("add-x")
[to_dtype2] output_0 = Cast <to: int = 1> (add)
}
dynamo-ir¶
inputs:
#1[(T1s4x4,)]shapes:
dict(x:{0:Dim(batch)})
<
ir_version: 10,
opset_import: ["" : 18],
producer_name: "pytorch",
producer_version: "2.9.0.dev20250701+cu126"
>
main_graph (float[batch,4] x) => (float[batch,4] _to_copy_1)
<bfloat16[batch,4] _to_copy, bfloat16[batch,4] add_3>
{
[node__to_copy] _to_copy = Cast <to: int = 16> (x)
[node_add_3] add_3 = Add (_to_copy, _to_copy)
[node__to_copy_1] _to_copy_1 = Cast <to: int = 1> (add_3)
}
FAILED
[ONNXRuntimeError] : 9 : NOT_IMPLEMENTED : Could not find an implementation for Add(14) node with name 'node_add_3'
Vmap¶
forward¶
def forward(self, x, y):
f = lambda x, y: x * y + 1 # noqa: E731
return torch.vmap(f)(x, y)
custom¶
FAILED
# no error found for the failure
dynamo-ir¶
FAILED
Failed to export the model with torch.export. [96mThis is step 1/3[0m of exporting the model to ONNX. Next steps:
- Modify the model code for `torch.export.export` to succeed. Refer to https://pytorch.org/docs/stable/generated/exportdb/index.html for more information.
- Debug `torch.export.export` and summit a PR to PyTorch.
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
## Exception summary
<class 'AssertionError'>:
(Refer to the full stack trace above for more information.)
VmapPython¶
forward¶
def forward(self, x, y):
f = lambda x, y: x * y + 1 # noqa: E731
return patched_vmap(f)(x, y)
custom¶
FAILED
object of type 'Node' has no len()
dynamo-ir¶
FAILED
Failed to export the model with torch.export. [96mThis is step 1/3[0m of exporting the model to ONNX. Next steps:
- Modify the model code for `torch.export.export` to succeed. Refer to https://pytorch.org/docs/stable/generated/exportdb/index.html for more information.
- Debug `torch.export.export` and summit a PR to PyTorch.
- Create an issue in the PyTorch GitHub repository against the [96m*torch.export*[0m component and attach the full error stack as well as reproduction scripts.
## Exception summary
<class 'TypeError'>: object of type 'Node' has no len()
(Refer to the full stack trace above for more information.)
Summary¶
case |
custom |
ir |
|---|---|---|
FAIL |
||
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
||
FAIL |
||
FAIL |
||
FAIL |
||
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
||
FAIL |
FAIL |
|
FAIL |
FAIL |
|
FAIL |
||
FAIL |
FAIL |
|
FAIL |
FAIL |