RUNTIME_EXCEPTION : Non-zero status code returned while running If node.

[FFchopon]

Describe the issue

I got an error while running the onnx model: Non-zero status code returned while running If node.

• Specific error report:

onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running If node. Name:'' Status Message: Non-zero status code returned while running If node. Name:'' Status Message: Non-zero status code returned while running GatherElements node. Name:'' Status Message: /software/onnxruntime/onnxruntime/core/providers/common.h:31 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) IsAxisInRange(axis, tensor_rank) was false. axis 1 is not in valid range [-1,0]

To reproduce

1. Download the model
2. Run the following script:

import onnx
import onnxruntime as ort
from onnxruntime.transformers import optimizer
import numpy as np

model_path = "38012.onnx"
optimized_model_path = f"./opt.onnx"

input_data = {
    'x': np.array([0.05135667, 0.49169028, 0.2317685, 0.7714388, 0.10556535,
                   0.7446228, 0.77673155, 0.4430063, 0.46662223, 0.4861915,
                   0.20035234, 0.5316502, 0.8579882, 0.01025127, 0.6977761,
                   0.7261769, 0.43401167, 0.77205604, 0.7047838, 0.8162092,
                   0.6132042, 0.59994316, 0.47380182, 0.89255065, 0.8315158,
                   0.7442334, 0.04432015, 0.9065669, 0.40642294, 0.8992343,
                   0.51915145, 0.96065384, 0.88932925, 0.7611128, 0.9960299,
                   0.04980307, 0.28150338, 0.6282024, 0.29980934, 0.6425377,
                   0.14627998, 0.38970673, 0.8462834, 0.8327852, 0.58551437,
                   0.5901391, 0.10345234, 0.1530145, 0.34184808, 0.96655446,
                   0.7688564, 0.10754307, 0.8349921, 0.46405357, 0.41953513,
                   0.24148946, 0.6070621, 0.00942784, 0.4898656, 0.3826074,
                   0.23482858, 0.5481194, 0.03267029, 0.36253762, 0.09667406,
                   0.5429725, 0.45386177, 0.66851056, 0.51004875, 0.39172414,
                   0.23835072, 0.02985721, 0.918535, 0.55690295, 0.97696245,
                   0.98216826, 0.6946321, 0.8859541, 0.1622831, 0.83600485,
                   0.4745072, 0.70302343, 0.3033251, 0.2943358, 0.77564985,
                   0.29016948, 0.84607005, 0.27696252, 0.9643625, 0.19356592,
                   0.78589076, 0.6827836, 0.98737943, 0.37815085, 0.1211899,
                   0.02344177, 0.97916144, 0.9867203, 0.7446272, 0.75813687,
                   0.31773388, 0.41267744, 0.12573875, 0.63623524, 0.09663095,
                   0.49160004, 0.6418833, 0.75377125, 0.48768246, 0.06855919,
                   0.4702471, 0.255228, 0.8079538, 0.5095185, 0.58212304,
                   0.06267849, 0.4565444, 0.00950742, 0.7498734, 0.04434598,
                   0.48962507, 0.3139298, 0.48399472, 0.44127202, 0.4732648,
                   0.3804463, 0.40799254, 0.24919167], dtype=np.float32),
    'x1': np.array([0.8426625, 0.9732153, 0.49775425, 0.05435705, 0.4693269,
                    0.2900393, 0.6734157, 0.6896115, 0.8811082, 0.11899561,
                    0.9244948, 0.94079465, 0.5876591, 0.23305634, 0.78063804,
                    0.17882146, 0.6678079, 0.70737696, 0.08595871, 0.05268361,
                    0.01278743, 0.25570008, 0.7130087, 0.1399794, 0.08106553,
                    0.5992047, 0.588875, 0.7871804, 0.7853509, 0.26299697,
                    0.8193554, 0.67199385, 0.6101456, 0.95636225, 0.5152923,
                    0.6044122, 0.44106615, 0.82251936, 0.54130244, 0.2778342,
                    0.601269, 0.6048449, 0.20572579, 0.3961332, 0.26576704,
                    0.24089175, 0.92432624, 0.5886368, 0.2728472, 0.01720504,
                    0.65580326, 0.91351014, 0.77888834, 0.60864544, 0.61413944,
                    0.7032979, 0.65464437, 0.1084903, 0.49285117, 0.9979988,
                    0.26293004, 0.38058266, 0.56481045, 0.7391961, 0.98462343,
                    0.02746766, 0.1915805, 0.799147, 0.29056203, 0.7198771,
                    0.79346496, 0.4845838, 0.2524755, 0.6142809, 0.29809123,
                    0.8227626, 0.78785723, 0.62629646, 0.8279695, 0.44274712,
                    0.76114076, 0.26292846, 0.00214652, 0.29157782, 0.33320805,
                    0.43552852, 0.03375685, 0.7057689, 0.75814784, 0.31626043,
                    0.24448082, 0.01732731, 0.3749923, 0.8667468, 0.7575453,
                    0.17516032, 0.33060876, 0.22861947, 0.4026713, 0.17343079,
                    0.691345, 0.62467605, 0.38594428, 0.3417037, 0.871786,
                    0.3767675, 0.9026966, 0.39513087, 0.98681647, 0.04550003,
                    0.5636926, 0.8291888, 0.93976754, 0.874003, 0.66336656,
                    0.76403767, 0.26931816, 0.8255282, 0.9449286, 0.22858198,
                    0.07249299, 0.5257493, 0.28457695, 0.08677769, 0.8126051,
                    0.78178006, 0.2609097, 0.28725886], dtype=np.float32)
}

sess_options = ort.SessionOptions()
sess_options.graph_optimization_level = ort.GraphOptimizationLevel.ORT_DISABLE_ALL
original_session = ort.InferenceSession(model_path, sess_options, providers=["CPUExecutionProvider"])
original_output_names = [output.name for output in original_session.get_outputs()]
original_result = original_session.run(original_output_names, input_data)

optimized_model = optimizer.optimize_model(model_path, opt_level=99)
optimized_model.save_model_to_file(optimized_model_path)
optimized_session = ort.InferenceSession(optimized_model_path, providers=["CPUExecutionProvider"])
optimized_output_names = [output.name for output in optimized_session.get_outputs()]
optimized_result = optimized_session.run(optimized_output_names, input_data)
for r1, r2 in zip(original_result, optimized_result):
    np.testing.assert_allclose(r1, r2, atol=1e-3, rtol=1e-3)

Urgency

No response

Platform

Linux

OS Version

Ubuntu 20.04

ONNX Runtime Installation

Built from Source

ONNX Runtime Version or Commit ID

5c1b7cc

ONNX Runtime API

Python

Architecture

X64

Execution Provider

CUDA

Execution Provider Library Version

No response

[github-actions]

This issue has been automatically marked as stale due to inactivity and will be closed in 30 days if no further activity occurs. If further support is needed, please provide an update and/or more details.