[BUG] The pass "eliminate_nop_transpose" causes output mismatch

[BUG] The pass "eliminate_nop_transpose" causes output mismatch

Issue
Running the single pass eliminate_nop_transpose with onnxoptimizer 0.4.2 changes numerical outputs.

Environment

• Ubuntu 20.04
• Python 3.10
• onnx==1.19.0
• onnxruntime==1.23.2
• onnxoptimizer==0.4.2(latest)

Repro steps (run from this folder)
eliminate_nop_transpose_repro.tar.gz

1. Create a Python environment (Python 3.10) and install dependencies:

python3 -m venv .venv
source .venv/bin/activate
pip install -U pip
pip install -r requirements.txt

2. Optimize a case with only eliminate_nop_transpose (writes model.opt.onnx next to model.onnx):

• python optimize_model.py --case ./case_03140_seed45116679

3. Differential test original and optimized models with the same oracle inputs (ORT graph optimization disabled):

• python diff_test.py --case ./case_03140_seed45116679

Expected
eliminate_nop_transpose should be semantics-preserving. Applying this pass alone should not change any output values.

Differential Test Output Details

1. case_03140_seed45116679 — mismatch; output[14] max_abs=4.022e+04 max_rel=4.022e+12, output[15] max_abs=5.190e+02 max_rel=5.120e+10; overall mismatch.
2. case_09334_seed94166965 — mismatch; outputs[3,4,9,10,11] non-zero; overall max_rel=3.682e+02; overall mismatch.

Case: case_03140_seed45116679 (ORT graph optimizations disabled)
  output[0]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 21)
  output[1]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(2, 1, 24)
  output[2]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(42336,)
  output[3]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 21, 24)
  output[4]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(21, 1)
  output[5]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(21, 1)
  output[6]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(21, 1)
  output[7]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(2, 24)
  output[8]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 24)
  output[9]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 24)
  output[10]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 24)
  output[11]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 24)
  output[12]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(21, 24)
  output[13]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(24, 21, 1)
  output[14]: max_abs=4.022e+04, max_rel=4.022e+12, shape=(1, 21, 24)
  output[15]: max_abs=5.190e+02, max_rel=5.120e+10, shape=(32, 21, 24)
Overall: max_abs=4.022e+04, max_rel=4.022e+12

Case: case_09334_seed94166965 (ORT graph optimizations disabled)
  output[0]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 2, 1, 1, 1)
  output[1]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(5, 1, 1, 1, 1)
  output[2]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(2, 1, 62, 1, 1)
  output[3]: max_abs=6.211e-01, max_rel=2.543e-01, shape=(2, 1, 1, 1, 62)
  output[4]: max_abs=1.000e+00, max_rel=1.000e+00, shape=(2, 1, 1, 1, 62)
  output[5]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(58, 1, 1, 1)
  output[6]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(58, 1, 1, 1)
  output[7]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(58, 1, 1, 1)
  output[8]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 58, 1, 62)
  output[9]: max_abs=2.009e-01, max_rel=5.917e-01, shape=(2, 58, 1, 62)
  output[10]: max_abs=3.922e-01, max_rel=3.682e+02, shape=(58, 2, 62, 1)
  output[11]: max_abs=1.792e-01, max_rel=5.014e-01, shape=(2, 58, 1, 62)
  output[12]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(58, 1, 62)
  output[13]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(4, 62)
  output[14]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(1, 58, 1, 62)
  output[15]: max_abs=0.000e+00, max_rel=0.000e+00, shape=(21576,)
Overall: max_abs=1.000e+00, max_rel=3.682e+02

Attachments

• requirements.txt
• optimize_model.py
• diff_test.py (runs original and optimized with same oracle inputs; ORT graph optimizations disabled)
• run_optimized_model.py (executes optimized model with oracle inputs)
• case_03140_seed45116679/,case_09334_seed94166965/ (original models and oracle inputs)

Start with the attached optimize_model.py and diff_test.py, then reproduce both listed cases using the provided requirements.txt and commands. Inspect how the eliminate_nop_transpose pass transforms each model and compare the original and optimized outputs with ORT graph optimizations disabled. Done means both cases remain numerically identical after the pass.