# Erorr with onnx to trt

**URL:** <https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796>\
**Category:** Jetson Xavier NX\
**Tags:** tensorrt\
**Created:** [March 18, 2022, 9:29am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796 "2022-03-18T09:29:49Z")\
**Posts on this page:** 9\
**Page:** 1

<div class="post-metadata">

**Author:** ![wang\_jin\_1999](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@wang\_jin\_1999](https://forums.developer.nvidia.com/u/wang_jin_1999)\
**Post date:** [March 18, 2022, 9:29am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/1 "2022-03-18T09:29:49Z")

</div>

## Description

A clear and concise description of the bug or issue.

## Environment

**TensorRT Version** : 8.2.1.8  
**GPU Type** : jetson xarive nx 16g  
**Nvidia Driver Version** : jetpack4.6.1  
**CUDA Version** : 10.2  
**CUDNN Version** : I dont kown,its installed by jetpack4.6.1.  
**Operating System + Version** : ubuntu 18.04LTS  
**Python Version (if applicable)**: 3.6.9

I use a .py to realize onnx to trt, error:

```auto
trt version 8.2.1.8

[03/18/2022-16:54:16] [TRT] [W] onnx2trt_utils.cpp:366: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.

[03/18/2022-16:54:16] [TRT] [W] onnx2trt_utils.cpp:392: One or more weights outside the range of INT32 was clamped

[03/18/2022-16:54:18] [TRT] [E] 2: [utils.cpp::checkMemLimit::380] Error Code 2: Internal Error (Assertion upperBound != 0 failed. Unknown embedded device detected. Please update the table with the entry: {{1794, 6, 16}, 12653},)

Traceback (most recent call last):

  File "tools/export_trt.py", line 77, in <module>

    f.write(engine.serialize())

AttributeError: 'NoneType' object has no attribute 'serialize'

```

.py:

```auto
import tensorrt as trt
import sys
import argparse

"""
takes in onnx model
converts to tensorrt
"""

if __name__ == ' __main__':

    desc = 'compile Onnx model to TensorRT'
    parser = argparse.ArgumentParser(description=desc)
    parser.add_argument('--model', help='onnx file')
    parser.add_argument('--out', type=str, default='', help='name of trt output file')
    parser.add_argument('--fp', type=int, default=16, help='floating point precision. 16 or 32')
    parser.add_argument('--batch', type=int, default=1)
    parser.add_argument('--verbose', action='store_true')
    parser.add_argument('--jetson_nano', action='store_true')
    opt = parser.parse_args()
    
    batch_size = opt.batch
    model = opt.model
    fp = opt.fp
    output = opt.out if opt.out else opt.model.replace('.onnx', '.trt')
    assert fp in (16, 32)

    if opt.jetson_nano:
        workspace = 1 << 28
    else:
        workspace = 4 * 1 << 30
    
    logger = trt.Logger(trt.Logger.WARNING)
    if opt.verbose:
        logger.min_severity = trt.Logger.VERBOSE
    
    print('trt version', trt. __version__ )
    assert trt. __version__ [0] >= '7'
    EXPLICIT_BATCH = 1 << (int)(trt.NetworkDefinitionCreationFlag.EXPLICIT_BATCH)

    with trt.Builder(logger) as builder, builder.create_network(EXPLICIT_BATCH) as network, trt.OnnxParser(network, logger) as parser:
        if trt. __version__ [0] == '7':
            builder.max_workspace_size = workspace
            builder.max_batch_size = batch_size
            if fp == 16:
                builder.fp16_mode = True

            with open(model, 'rb') as f:
                if not parser.parse(f.read()):
                    for error in range(parser.num_errors):
                        print('ERROR', parser.get_error(error))
            
            # if your onnx has a dynamic input...
            # network.get_input(0).shape = [1, 3, 352, 608]
            
            engine = builder.build_cuda_engine(network)
            with open(output, 'wb') as f:
                f.write(engine.serialize())
            print('Done')
        else:
            # https://github.com/NVIDIA-AI-IOT/torch2trt/issues/557
            # https://github.com/NVIDIA-AI-IOT/torch2trt/commit/8f742904d603fcde4fe521baa31bdc18002c23cb#diff-f682ce583d8646e112002fb3631f7d205b63aae7b0ca673b020fed7244d4ed38
            
            config = builder.create_builder_config()
            config.max_workspace_size = workspace
            if fp == 16:
                config.set_flag(trt.BuilderFlag.FP16)
            builder.max_batch_size = batch_size
            
            with open(model, 'rb') as f:
                if not parser.parse(f.read()):
                    for error in range(parser.num_errors):
                        print('ERROR', parser.get_error(error))
                        
            engine = builder.build_engine(network, config)
            with open(output, 'wb') as f:
                f.write(engine.serialize())
            print('Done')

```

it can work in jetson nano with jetpack4.5, should I install jetpack 4.5 to nx?

---

<div class="post-metadata">

**Author:** ![wang\_jin\_1999](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@wang\_jin\_1999](https://forums.developer.nvidia.com/u/wang_jin_1999)\
**Post date:** [March 19, 2022, 1:32pm UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/2 "2022-03-19T13:32:32Z")

</div>

> [@wang\_jin\_1999](#):
>
> `engine.serialize()`

hello?

---

<div class="post-metadata">

**Author:** ![wang\_jin\_1999](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@wang\_jin\_1999](https://forums.developer.nvidia.com/u/wang_jin_1999)\
**Post date:** [March 21, 2022, 9:12am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/3 "2022-03-21T09:12:27Z")

</div>

Plus: It could not work in trtexec.  
jetson nano developkit with jetpack4.6.1 can run it.  
I get the verbose:

```auto
fwav@ubuntu:/usr/src/tensorrt/bin$ ./trtexec --onnx=/media/fwav/1706-63D2/yolo-fastestv2.onnx --saveEngine=/media/fwav/1706-63D2/test.trt
&&&& RUNNING TensorRT.trtexec [TensorRT v8201] # ./trtexec --onnx=/media/fwav/1706-63D2/yolo-fastestv2.onnx --saveEngine=/media/fwav/1706-63D2/test.trt
[03/21/2022-15:24:04] [I] === Model Options ===
[03/21/2022-15:24:04] [I] Format: ONNX
[03/21/2022-15:24:04] [I] Model: /media/fwav/1706-63D2/yolo-fastestv2.onnx
[03/21/2022-15:24:04] [I] Output:
[03/21/2022-15:24:04] [I] === Build Options ===
[03/21/2022-15:24:04] [I] Max batch: explicit batch
[03/21/2022-15:24:04] [I] Workspace: 16 MiB
[03/21/2022-15:24:04] [I] minTiming: 1
[03/21/2022-15:24:04] [I] avgTiming: 8
[03/21/2022-15:24:04] [I] Precision: FP32
[03/21/2022-15:24:04] [I] Calibration: 
[03/21/2022-15:24:04] [I] Refit: Disabled
[03/21/2022-15:24:04] [I] Sparsity: Disabled
[03/21/2022-15:24:04] [I] Safe mode: Disabled
[03/21/2022-15:24:04] [I] DirectIO mode: Disabled
[03/21/2022-15:24:04] [I] Restricted mode: Disabled
[03/21/2022-15:24:04] [I] Save engine: /media/fwav/1706-63D2/test.trt
[03/21/2022-15:24:04] [I] Load engine: 
[03/21/2022-15:24:04] [I] Profiling verbosity: 0
[03/21/2022-15:24:04] [I] Tactic sources: Using default tactic sources
[03/21/2022-15:24:04] [I] timingCacheMode: local
[03/21/2022-15:24:04] [I] timingCacheFile: 
[03/21/2022-15:24:04] [I] Input(s)s format: fp32:CHW
[03/21/2022-15:24:04] [I] Output(s)s format: fp32:CHW
[03/21/2022-15:24:04] [I] Input build shapes: model
[03/21/2022-15:24:04] [I] Input calibration shapes: model
[03/21/2022-15:24:04] [I] === System Options ===
[03/21/2022-15:24:04] [I] Device: 0
[03/21/2022-15:24:04] [I] DLACore: 
[03/21/2022-15:24:04] [I] Plugins:
[03/21/2022-15:24:04] [I] === Inference Options ===
[03/21/2022-15:24:04] [I] Batch: Explicit
[03/21/2022-15:24:04] [I] Input inference shapes: model
[03/21/2022-15:24:04] [I] Iterations: 10
[03/21/2022-15:24:04] [I] Duration: 3s (+ 200ms warm up)
[03/21/2022-15:24:04] [I] Sleep time: 0ms
[03/21/2022-15:24:04] [I] Idle time: 0ms
[03/21/2022-15:24:04] [I] Streams: 1
[03/21/2022-15:24:04] [I] ExposeDMA: Disabled
[03/21/2022-15:24:04] [I] Data transfers: Enabled
[03/21/2022-15:24:04] [I] Spin-wait: Disabled
[03/21/2022-15:24:04] [I] Multithreading: Disabled
[03/21/2022-15:24:04] [I] CUDA Graph: Disabled
[03/21/2022-15:24:04] [I] Separate profiling: Disabled
[03/21/2022-15:24:04] [I] Time Deserialize: Disabled
[03/21/2022-15:24:04] [I] Time Refit: Disabled
[03/21/2022-15:24:04] [I] Skip inference: Disabled
[03/21/2022-15:24:04] [I] Inputs:
[03/21/2022-15:24:04] [I] === Reporting Options ===
[03/21/2022-15:24:04] [I] Verbose: Disabled
[03/21/2022-15:24:04] [I] Averages: 10 inferences
[03/21/2022-15:24:04] [I] Percentile: 99
[03/21/2022-15:24:04] [I] Dump refittable layers:Disabled
[03/21/2022-15:24:04] [I] Dump output: Disabled
[03/21/2022-15:24:04] [I] Profile: Disabled
[03/21/2022-15:24:04] [I] Export timing to JSON file: 
[03/21/2022-15:24:04] [I] Export output to JSON file: 
[03/21/2022-15:24:04] [I] Export profile to JSON file: 
[03/21/2022-15:24:04] [I] 
[03/21/2022-15:24:04] [I] === Device Information ===
[03/21/2022-15:24:04] [I] Selected Device: Xavier
[03/21/2022-15:24:04] [I] Compute Capability: 7.2
[03/21/2022-15:24:04] [I] SMs: 6
[03/21/2022-15:24:04] [I] Compute Clock Rate: 1.109 GHz
[03/21/2022-15:24:04] [I] Device Global Memory: 15825 MiB
[03/21/2022-15:24:04] [I] Shared Memory per SM: 96 KiB
[03/21/2022-15:24:04] [I] Memory Bus Width: 256 bits (ECC disabled)
[03/21/2022-15:24:04] [I] Memory Clock Rate: 1.109 GHz
[03/21/2022-15:24:04] [I] 
[03/21/2022-15:24:04] [I] TensorRT version: 8.2.1
[03/21/2022-15:24:06] [I] [TRT] [MemUsageChange] Init CUDA: CPU +362, GPU +0, now: CPU 381, GPU 4022 (MiB)
[03/21/2022-15:24:06] [I] [TRT] [MemUsageSnapshot] Begin constructing builder kernel library: CPU 381 MiB, GPU 4022 MiB
[03/21/2022-15:24:06] [I] [TRT] [MemUsageSnapshot] End constructing builder kernel library: CPU 486 MiB, GPU 4127 MiB
[03/21/2022-15:24:06] [I] Start parsing network model
[03/21/2022-15:24:06] [I] [TRT] ----------------------------------------------------------------
[03/21/2022-15:24:06] [I] [TRT] Input filename: /media/fwav/1706-63D2/yolo-fastestv2.onnx
[03/21/2022-15:24:06] [I] [TRT] ONNX IR version: 0.0.6
[03/21/2022-15:24:06] [I] [TRT] Opset version: 11
[03/21/2022-15:24:06] [I] [TRT] Producer name: pytorch
[03/21/2022-15:24:06] [I] [TRT] Producer version: 1.9
[03/21/2022-15:24:06] [I] [TRT] Domain:           
[03/21/2022-15:24:06] [I] [TRT] Model version: 0
[03/21/2022-15:24:06] [I] [TRT] Doc string:       
[03/21/2022-15:24:06] [I] [TRT] ----------------------------------------------------------------
[03/21/2022-15:24:06] [W] [TRT] onnx2trt_utils.cpp:366: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
[03/21/2022-15:24:07] [I] Finish parsing network model
[03/21/2022-15:24:07] [I] [TRT] ---------- Layers Running on DLA ----------
[03/21/2022-15:24:07] [I] [TRT] ---------- Layers Running on GPU ----------
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_0 + Relu_1
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] MaxPool_2
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_6 + Relu_7
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_3
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_4
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_8
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_9
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[Relu_5...Gather_20]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_21 + Relu_22
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_23
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_24
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[468...Gather_35]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_36 + Relu_37
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_38
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_39
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[486...Gather_48]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_51 + Relu_52
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_53
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_54 + Relu_55
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] 505 copy
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_60 + Relu_61
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_57
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_58
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_62
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_63
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[Relu_59...Gather_74]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_75 + Relu_76
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_77
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_78
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[536...Gather_89]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_90 + Relu_91
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_92
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_93
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[554...Gather_104]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_105 + Relu_106
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_107
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_108
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[572...Gather_119]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_120 + Relu_121
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_122
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_123
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[590...Gather_134]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_135 + Relu_136
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_137
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_138
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[608...Gather_149]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_150 + Relu_151
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_152
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_153
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[626...Gather_162]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_165 + Relu_166
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_167
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_168 + Relu_169
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] 645 copy
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_174 + Relu_175
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_171
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_172
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_176
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_177
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[Relu_173...Gather_188]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_189 + Relu_190
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_191
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_192
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[676...Gather_203]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_204 + Relu_205
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_206
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_207
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] {ForeignNode[694...Gather_216]}
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_219 + Relu_220
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_221
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_222 + Relu_223
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] 713 copy
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Resize_240
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_225 + Relu_226
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] 752 copy
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_233 + Relu_234
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_227 + Relu_228
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_242 + Relu_243
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_250 + Relu_251
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_244 + Relu_245
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_229
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_230 + Relu_231
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_235
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_236 + Relu_237
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_238
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_259
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] PWN(Sigmoid_267)
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_232
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_261 || Conv_260
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Transpose_269 + (Unnamed Layer* 245) [Shuffle]
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Softmax_270
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] (Unnamed Layer* 247) [Shuffle] + Transpose_271
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] PWN(Sigmoid_268)
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] 789 copy
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Transpose_275
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_246
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_247 + Relu_248
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_252
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_253 + Relu_254
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_255
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_256
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] PWN(Sigmoid_262)
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_249
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Conv_258 || Conv_257
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Transpose_264 + (Unnamed Layer* 238) [Shuffle]
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Softmax_265
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] (Unnamed Layer* 240) [Shuffle] + Transpose_266
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] PWN(Sigmoid_263)
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] 784 copy
[03/21/2022-15:24:07] [I] [TRT] [GpuLayer] Transpose_273
[03/21/2022-15:24:08] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +227, GPU +228, now: CPU 715, GPU 4359 (MiB)
[03/21/2022-15:24:09] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +307, GPU +322, now: CPU 1022, GPU 4681 (MiB)
[03/21/2022-15:24:09] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
[03/21/2022-15:24:09] [E] Error[2]: [utils.cpp::checkMemLimit::380] Error Code 2: Internal Error (Assertion upperBound != 0 failed. Unknown embedded device detected. Please update the table with the entry: {{1794, 6, 16}, 12660},)
[03/21/2022-15:24:09] [E] Error[2]: [builder.cpp::buildSerializedNetwork::609] Error Code 2: Internal Error (Assertion enginePtr != nullptr failed. )
[03/21/2022-15:24:09] [E] Engine could not be created from network
[03/21/2022-15:24:09] [E] Building engine failed
[03/21/2022-15:24:09] [E] Failed to create engine from model.
[03/21/2022-15:24:09] [E] Engine set up failed
&&&& FAILED TensorRT.trtexec [TensorRT v8201] # ./trtexec --onnx=/media/fwav/1706-63D2/yolo-fastestv2.onnx --saveEngine=/media/fwav/1706-63D2/test.trt

```

---

<div class="post-metadata">

**Author:** ![spolisetty](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/spolisetty/32/14047_2.png) [@spolisetty](https://forums.developer.nvidia.com/u/spolisetty)\
**Post date:** [March 23, 2022, 11:26am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/4 "2022-03-23T11:26:36Z")

</div>

> [@wang\_jin\_1999](#):
>
> it can work in jetson nano with jetpack4.5, should I install jetpack 4.5 to nx?

We are moving this to the Jetson Xavier forum to get better help.

Thank you.

---

<div class="post-metadata">

**Author:** ![AastaLLL](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aastalll/32/14043_2.png) [@AastaLLL](https://forums.developer.nvidia.com/u/AastaLLL)\
**Post date:** [March 24, 2022, 3:23am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/6 "2022-03-24T03:23:54Z")

</div>

Duplicate to the following topic:

> [@ERORR with ONNX2TRT : Unknown embedded device detected](https://forums.developer.nvidia.com/t/erorr-with-onnx2trt-unknown-embedded-device-detected/208933/):
>
> BOARD:Jetson Xariver NX 16g emmc + jetson nano B01 carrier board. Environment TensorRT Version : 8.2.1.8 GPU Type : jetson xarive nx 16g Nvidia Driver Version : jetpack4.6.1 CUDA Version : 10.2 CUDNN Version : I dont kown,its installed by jetpack4.6.1. Operating System + Version : ubuntu 18.04LTS Python Version (if applicable) : 3.6.9 I use a .py to realize onnx to trt, error: trt version 8.2.1.8 [03/18/2022-16:54:16] [TRT] [W] onnx2trt\_utils.cpp:366: Your ONNX model has been generated…

---

<div class="post-metadata">

**Author:** ![wang\_jin\_1999](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@wang\_jin\_1999](https://forums.developer.nvidia.com/u/wang_jin_1999)\
**Post date:** [March 24, 2022, 7:28am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/7 "2022-03-24T07:28:38Z")

</div>

This topic was initiated by me in tensorRT section, and it was moved over.  
I suspect my problem is because I used 16g nx not 8g RAM. I hope you can solve it in the next version of jetpack.

---

<div class="post-metadata">

**Author:** ![AastaLLL](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aastalll/32/14043_2.png) [@AastaLLL](https://forums.developer.nvidia.com/u/AastaLLL)\
**Post date:** [March 28, 2022, 5:05am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/8 "2022-03-28T05:05:17Z")

</div>

Hi,

Is this the same issue of [topic 208933](https://forums.developer.nvidia.com/t/erorr-with-onnx2trt-unknown-embedded-device-detected/208933)?  
If yes, let’s discuss the following on that topic for clarity.

Thanks.

---

<div class="post-metadata">

**Author:** ![wang\_jin\_1999](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@wang\_jin\_1999](https://forums.developer.nvidia.com/u/wang_jin_1999)\
**Post date:** [March 30, 2022, 10:53am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/9 "2022-03-30T10:53:25Z")

</div>

yes

---

<div class="post-metadata">

**Author:** ![system](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/system/32/68080_2.png) [@system](https://forums.developer.nvidia.com/u/system)\
**Post date:** [April 20, 2022, 7:24am UTC](https://forums.developer.nvidia.com/t/erorr-with-onnx-to-trt/208796/11 "2022-04-20T07:24:10Z")

</div>

Closed on April 20, 2022, 7:24am UTC

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.