Help : Error with TensorRT parsing with ONNX file

rdpdo2002 · June 14, 2022, 9:56am

Hello,

I got this error while converting a onnx file with tensorRT v8205. I already post on github but I dont yet be able to resolve my issue…

I compiled version 8205 from source on my jetson nano so perhpas it it a mistake in my configuration ??

My onnx file has been created by tf2onnx.convert v.1.10.0 (with opset 13) from a tensorflow v2.6.2+nv21.12 . Resenet-18 model on jetson nano.
Jetson nano 4GB
tf2onnx.convert v.1.10.0 (with opset 13)
tensorflow v2.6.2+nv21.12 (Resenet-18 model) on jetson nano.
tensorrt compiled from source

This is my onnx file after conversion from tf2 saved model with tf2onnx.convert v.1.10.0 (with opset 13 by default) :

https://drive.google.com/file/d/15MIiJjnEHqaT2vJ3ni0Q1QrxUPUyNQk9/view?usp=sharing

And this is my tf2 (v2.6.2+nv21.12) saved model directory zipped :

https://drive.google.com/file/d/1ZaeVgzY11rCbKZyk9G87VlO_RejXh9Vh/view?usp=sharing

Can someone help me please ?? Thanks a lot !!

&&&& RUNNING TensorRT.trtexec [TensorRT v8205] # TensorRT/build/trtexec --fp16 --verbose --workspace=256 --onnx=meilleur_model.onnx --saveEngine=model_collision.engine
[06/12/2022-19:05:15] [I] === Model Options ===
[06/12/2022-19:05:15] [I] Format: ONNX
[06/12/2022-19:05:15] [I] Model: meilleur_model.onnx
[06/12/2022-19:05:15] [I] Output:
[06/12/2022-19:05:15] [I] === Build Options ===
[06/12/2022-19:05:15] [I] Max batch: explicit batch
[06/12/2022-19:05:15] [I] Workspace: 256 MiB
[06/12/2022-19:05:15] [I] minTiming: 1
[06/12/2022-19:05:15] [I] avgTiming: 8
[06/12/2022-19:05:15] [I] Precision: FP32+FP16
[06/12/2022-19:05:15] [I] Calibration: 
[06/12/2022-19:05:15] [I] Refit: Disabled
[06/12/2022-19:05:15] [I] Sparsity: Disabled
[06/12/2022-19:05:15] [I] Safe mode: Disabled
[06/12/2022-19:05:15] [I] DirectIO mode: Disabled
[06/12/2022-19:05:15] [I] Restricted mode: Disabled
[06/12/2022-19:05:15] [I] Save engine: model_collision.engine
[06/12/2022-19:05:15] [I] Load engine: 
[06/12/2022-19:05:15] [I] Profiling verbosity: 0
[06/12/2022-19:05:15] [I] Tactic sources: Using default tactic sources
[06/12/2022-19:05:15] [I] timingCacheMode: local
[06/12/2022-19:05:15] [I] timingCacheFile: 
[06/12/2022-19:05:15] [I] Input(s)s format: fp32:CHW
[06/12/2022-19:05:15] [I] Output(s)s format: fp32:CHW
[06/12/2022-19:05:15] [I] Input build shapes: model
[06/12/2022-19:05:15] [I] Input calibration shapes: model
[06/12/2022-19:05:15] [I] === System Options ===
[06/12/2022-19:05:15] [I] Device: 0
[06/12/2022-19:05:15] [I] DLACore: 
[06/12/2022-19:05:15] [I] Plugins:
[06/12/2022-19:05:15] [I] === Inference Options ===
[06/12/2022-19:05:15] [I] Batch: Explicit
[06/12/2022-19:05:15] [I] Input inference shapes: model
[06/12/2022-19:05:15] [I] Iterations: 10
[06/12/2022-19:05:15] [I] Duration: 3s (+ 200ms warm up)
[06/12/2022-19:05:15] [I] Sleep time: 0ms
[06/12/2022-19:05:15] [I] Idle time: 0ms
[06/12/2022-19:05:15] [I] Streams: 1
[06/12/2022-19:05:15] [I] ExposeDMA: Disabled
[06/12/2022-19:05:15] [I] Data transfers: Enabled
[06/12/2022-19:05:15] [I] Spin-wait: Disabled
[06/12/2022-19:05:15] [I] Multithreading: Disabled
[06/12/2022-19:05:15] [I] CUDA Graph: Disabled
[06/12/2022-19:05:15] [I] Separate profiling: Disabled
[06/12/2022-19:05:15] [I] Time Deserialize: Disabled
[06/12/2022-19:05:15] [I] Time Refit: Disabled
[06/12/2022-19:05:15] [I] Skip inference: Disabled
[06/12/2022-19:05:15] [I] Inputs:
[06/12/2022-19:05:15] [I] === Reporting Options ===
[06/12/2022-19:05:15] [I] Verbose: Enabled
[06/12/2022-19:05:15] [I] Averages: 10 inferences
[06/12/2022-19:05:15] [I] Percentile: 99
[06/12/2022-19:05:15] [I] Dump refittable layers:Disabled
[06/12/2022-19:05:15] [I] Dump output: Disabled
[06/12/2022-19:05:15] [I] Profile: Disabled
[06/12/2022-19:05:15] [I] Export timing to JSON file: 
[06/12/2022-19:05:15] [I] Export output to JSON file: 
[06/12/2022-19:05:15] [I] Export profile to JSON file: 
[06/12/2022-19:05:15] [I] 
[06/12/2022-19:05:15] [I] === Device Information ===
[06/12/2022-19:05:15] [I] Selected Device: NVIDIA Tegra X1
[06/12/2022-19:05:15] [I] Compute Capability: 5.3
[06/12/2022-19:05:15] [I] SMs: 1
[06/12/2022-19:05:15] [I] Compute Clock Rate: 0.9216 GHz
[06/12/2022-19:05:15] [I] Device Global Memory: 3964 MiB
[06/12/2022-19:05:15] [I] Shared Memory per SM: 64 KiB
[06/12/2022-19:05:15] [I] Memory Bus Width: 64 bits (ECC disabled)
[06/12/2022-19:05:15] [I] Memory Clock Rate: 0.01275 GHz
[06/12/2022-19:05:15] [I] 
[06/12/2022-19:05:15] [I] TensorRT version: 8.2.5
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::BatchTilePlugin_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::BatchedNMS_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::BatchedNMSDynamic_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::CoordConvAC version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::CropAndResize version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::CropAndResizeDynamic version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::DecodeBbox3DPlugin version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::DetectionLayer_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::EfficientNMS_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::EfficientNMS_ONNX_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::EfficientNMS_Explicit_TF_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::EfficientNMS_Implicit_TF_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::FlattenConcat_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::GenerateDetection_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::GridAnchor_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::GridAnchorRect_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::InstanceNormalization_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::LReLU_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::MultilevelCropAndResize_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::MultilevelProposeROI_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::DMHA version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::NMS_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::NMSDynamic_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::Normalize_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::PillarScatterPlugin version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::PriorBox_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::ProposalLayer_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::Proposal version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::ProposalDynamic version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::PyramidROIAlign_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::Region_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::Reorg_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::ResizeNearest_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::RPROI_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::ScatterND version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::SpecialSlice_TRT version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::Split version 1
[06/12/2022-19:05:15] [V] [TRT] Registered plugin creator - ::VoxelGeneratorPlugin version 1
[06/12/2022-19:05:16] [I] [TRT] [MemUsageChange] Init CUDA: CPU +203, GPU +0, now: CPU 221, GPU 2335 (MiB)
[06/12/2022-19:05:16] [I] Start parsing network model
[06/12/2022-19:05:16] [I] [TRT] ----------------------------------------------------------------
[06/12/2022-19:05:16] [I] [TRT] Input filename:   meilleur_model.onnx
[06/12/2022-19:05:16] [I] [TRT] ONNX IR version:  0.0.7
[06/12/2022-19:05:16] [I] [TRT] Opset version:    13
[06/12/2022-19:05:16] [I] [TRT] Producer name:    tf2onnx
[06/12/2022-19:05:16] [I] [TRT] Producer version: 1.10.0 e9b6cb
[06/12/2022-19:05:16] [I] [TRT] Domain:           
[06/12/2022-19:05:16] [I] [TRT] Model version:    0
[06/12/2022-19:05:16] [I] [TRT] Doc string:       
[06/12/2022-19:05:16] [I] [TRT] ----------------------------------------------------------------
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::BatchTilePlugin_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::BatchedNMS_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::BatchedNMSDynamic_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::CoordConvAC version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::CropAndResize version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::CropAndResizeDynamic version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::DecodeBbox3DPlugin version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::DetectionLayer_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::EfficientNMS_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::EfficientNMS_ONNX_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::EfficientNMS_Explicit_TF_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::EfficientNMS_Implicit_TF_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::FlattenConcat_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::GenerateDetection_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::GridAnchor_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::GridAnchorRect_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::InstanceNormalization_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::LReLU_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::MultilevelCropAndResize_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::MultilevelProposeROI_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::DMHA version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::NMS_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::NMSDynamic_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::Normalize_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::PillarScatterPlugin version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::PriorBox_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::ProposalLayer_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::Proposal version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::ProposalDynamic version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::PyramidROIAlign_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::Region_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::Reorg_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::ResizeNearest_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::RPROI_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::ScatterND version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::SpecialSlice_TRT version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::Split version 1
[06/12/2022-19:05:16] [V] [TRT] Plugin creator already registered - ::VoxelGeneratorPlugin version 1
[06/12/2022-19:05:16] [V] [TRT] Adding network input: data with dtype: float32, dimensions: (1, 224, 224, 3)
[06/12/2022-19:05:16] [V] [TRT] Registering tensor: data for ONNX tensor: data
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: pad_const__82
[06/12/2022-19:05:16] [W] [TRT] parsers/onnx/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.
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: const_axes__203
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit2_conv2/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit2_conv1/Conv2D_weights_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit2_conv1/Conv2D_bias_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit2_bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit2_bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit2_bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit2_bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit1_sc/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit1_conv2/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit1_conv1/Conv2D_weights_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit1_conv1/Conv2D_bias_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit1_bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit1_bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit1_bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage4_unit1_bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit2_conv2/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit2_conv1/Conv2D_weights_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit2_conv1/Conv2D_bias_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit2_bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit2_bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit2_bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit2_bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit1_sc/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit1_conv2/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit1_conv1/Conv2D_weights_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit1_conv1/Conv2D_bias_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit1_bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit1_bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit1_bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage3_unit1_bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit2_conv2/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit2_conv1/Conv2D_weights_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit2_conv1/Conv2D_bias_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit2_bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit2_bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit2_bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit2_bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit1_sc/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit1_conv2/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit1_conv1/Conv2D_weights_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit1_conv1/Conv2D_bias_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit1_bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit1_bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit1_bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage2_unit1_bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit2_conv2/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit2_conv1/Conv2D_weights_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit2_conv1/Conv2D_bias_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit2_bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit2_bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit2_bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit2_bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit1_sc/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit1_conv2/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit1_conv1/Conv2D_weights_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit1_conv1/Conv2D_bias_fused_bn
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit1_bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit1_bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit1_bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/stage1_unit1_bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/dense/MatMul/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/dense/BiasAdd/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/conv0/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn_data/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn1/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn1/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn1/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn1/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn0/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn0/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Importing initializer: Func/StatefulPartitionedCall/input/_1:0
[06/12/2022-19:05:16] [V] [TRT] Parsing node: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74 [Transpose]
[06/12/2022-19:05:16] [V] [TRT] Searching for input: data
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74 [Transpose] inputs: [data -> (1, 224, 224, 3)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Registering layer: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74 for ONNX node: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74
[06/12/2022-19:05:16] [V] [TRT] Registering tensor: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74:0 for ONNX tensor: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74 [Transpose] outputs: [StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74:0 -> (1, 3, 224, 224)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Parsing node: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3 [BatchNormalization]
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: Func/StatefulPartitionedCall/input/_1:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn_data/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3 [BatchNormalization] inputs: [StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3__74:0 -> (1, 3, 224, 224)[FLOAT]], [Func/StatefulPartitionedCall/input/_1:0 -> (3)[FLOAT]], [StatefulPartitionedCall/model_1/bn_data/ReadVariableOp:0 -> (3)[FLOAT]], [StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3/ReadVariableOp:0 -> (3)[FLOAT]], [StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3/ReadVariableOp_1:0 -> (3)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Registering layer: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3 for ONNX node: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3
[06/12/2022-19:05:16] [V] [TRT] Registering tensor: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3:0 for ONNX tensor: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3 [BatchNormalization] outputs: [StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3:0 -> (1, 3, 224, 224)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Parsing node: StatefulPartitionedCall/model_1/conv0/Conv2D [Conv]
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/conv0/Conv2D/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/conv0/Conv2D [Conv] inputs: [StatefulPartitionedCall/model_1/bn_data/FusedBatchNormV3:0 -> (1, 3, 224, 224)[FLOAT]], [StatefulPartitionedCall/model_1/conv0/Conv2D/ReadVariableOp:0 -> (64, 3, 7, 7)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Convolution input dimensions: (1, 3, 224, 224)
[06/12/2022-19:05:16] [V] [TRT] Registering layer: StatefulPartitionedCall/model_1/conv0/Conv2D for ONNX node: StatefulPartitionedCall/model_1/conv0/Conv2D
[06/12/2022-19:05:16] [V] [TRT] Using kernel: (7, 7), strides: (2, 2), prepadding: (3, 3), postpadding: (3, 3), dilations: (1, 1), numOutputs: 64
[06/12/2022-19:05:16] [V] [TRT] Convolution output dimensions: (1, 64, 112, 112)
[06/12/2022-19:05:16] [V] [TRT] Registering tensor: StatefulPartitionedCall/model_1/conv0/Conv2D:0 for ONNX tensor: StatefulPartitionedCall/model_1/conv0/Conv2D:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/conv0/Conv2D [Conv] outputs: [StatefulPartitionedCall/model_1/conv0/Conv2D:0 -> (1, 64, 112, 112)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Parsing node: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3 [BatchNormalization]
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/conv0/Conv2D:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn0/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn0/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3/ReadVariableOp:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3/ReadVariableOp_1:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3 [BatchNormalization] inputs: [StatefulPartitionedCall/model_1/conv0/Conv2D:0 -> (1, 64, 112, 112)[FLOAT]], [StatefulPartitionedCall/model_1/bn0/ReadVariableOp:0 -> (64)[FLOAT]], [StatefulPartitionedCall/model_1/bn0/ReadVariableOp_1:0 -> (64)[FLOAT]], [StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3/ReadVariableOp:0 -> (64)[FLOAT]], [StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3/ReadVariableOp_1:0 -> (64)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Registering layer: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3 for ONNX node: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3
[06/12/2022-19:05:16] [V] [TRT] Registering tensor: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3:0 for ONNX tensor: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3 [BatchNormalization] outputs: [StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3:0 -> (1, 64, 112, 112)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Parsing node: StatefulPartitionedCall/model_1/relu0/Relu [Relu]
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/relu0/Relu [Relu] inputs: [StatefulPartitionedCall/model_1/bn0/FusedBatchNormV3:0 -> (1, 64, 112, 112)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Registering layer: StatefulPartitionedCall/model_1/relu0/Relu for ONNX node: StatefulPartitionedCall/model_1/relu0/Relu
[06/12/2022-19:05:16] [V] [TRT] Registering tensor: StatefulPartitionedCall/model_1/relu0/Relu:0 for ONNX tensor: StatefulPartitionedCall/model_1/relu0/Relu:0
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/relu0/Relu [Relu] outputs: [StatefulPartitionedCall/model_1/relu0/Relu:0 -> (1, 64, 112, 112)[FLOAT]], 
[06/12/2022-19:05:16] [V] [TRT] Parsing node: StatefulPartitionedCall/model_1/zero_padding2d_1/Pad [Pad]
[06/12/2022-19:05:16] [V] [TRT] Searching for input: StatefulPartitionedCall/model_1/relu0/Relu:0
[06/12/2022-19:05:16] [V] [TRT] Searching for input: pad_const__82
[06/12/2022-19:05:16] [V] [TRT] StatefulPartitionedCall/model_1/zero_padding2d_1/Pad [Pad] inputs: [StatefulPartitionedCall/model_1/relu0/Relu:0 -> (1, 64, 112, 112)[FLOAT]], [pad_const__82 -> (8)[INT32]], 
[06/12/2022-19:05:16] [E] Error[3]: [layers.cpp::setMode::3848] Error Code 3: Internal Error (Parameter check failed at: optimizer/api/layers.cpp::setMode::3848, condition: int(mode) >= 0 && int(mode) < EnumMax<SliceMode>()
)
[06/12/2022-19:05:16] [V] [TRT] Registering layer: StatefulPartitionedCall/model_1/zero_padding2d_1/Pad for ONNX node: StatefulPartitionedCall/model_1/zero_padding2d_1/Pad
[06/12/2022-19:05:16] [E] [TRT] parsers/onnx/ModelImporter.cpp:780: While parsing node number 5 [Pad -> "StatefulPartitionedCall/model_1/zero_padding2d_1/Pad:0"]:
[06/12/2022-19:05:16] [E] [TRT] parsers/onnx/ModelImporter.cpp:781: --- Begin node ---
[06/12/2022-19:05:16] [E] [TRT] parsers/onnx/ModelImporter.cpp:782: input: "StatefulPartitionedCall/model_1/relu0/Relu:0"
input: "pad_const__82"
output: "StatefulPartitionedCall/model_1/zero_padding2d_1/Pad:0"
name: "StatefulPartitionedCall/model_1/zero_padding2d_1/Pad"
op_type: "Pad"

[06/12/2022-19:05:16] [E] [TRT] parsers/onnx/ModelImporter.cpp:783: --- End node ---
[06/12/2022-19:05:16] [E] [TRT] parsers/onnx/ModelImporter.cpp:786: ERROR: parsers/onnx/ModelImporter.cpp:166 In function parseGraph:
[6] Invalid Node - StatefulPartitionedCall/model_1/zero_padding2d_1/Pad
[layers.cpp::setMode::3848] Error Code 3: Internal Error (Parameter check failed at: optimizer/api/layers.cpp::setMode::3848, condition: int(mode) >= 0 && int(mode) < EnumMax<SliceMode>()
)
[06/12/2022-19:05:16] [E] Failed to parse onnx file
[06/12/2022-19:05:16] [I] Finish parsing network model
[06/12/2022-19:05:16] [E] Parsing model failed
[06/12/2022-19:05:16] [E] Failed to create engine from model.
[06/12/2022-19:05:16] [E] Engine set up failed
&&&& FAILED TensorRT.trtexec [TensorRT v8205] # TensorRT/build/trtexec --fp16 --verbose --workspace=256 --onnx=meilleur_model.onnx --saveEngine=model_collision.engine

Here is my library file for tensorrt, does it seems correct ?

alexandre@alexandre-desktop:~$ ls -l /usr/lib/aarch64-linux-gnu/libnv*.*
lrwxrwxrwx 1 root root        21 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvcaffe_parser.a -> libnvparsers_static.a
lrwxrwxrwx 1 root root        21 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvcaffe_parser.so -> libnvparsers.so.8.0.1
-rwxr-xr-x 1 root root   5275176 juin  13 23:35 /usr/lib/aarch64-linux-gnu/libnvcaffeparser.so
lrwxrwxrwx 1 root root        21 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvcaffe_parser.so.8 -> libnvparsers.so.8.0.1
-rwxr-xr-x 1 root root   5275176 juin  13 23:35 /usr/lib/aarch64-linux-gnu/libnvcaffeparser.so.8
lrwxrwxrwx 1 root root        21 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvcaffe_parser.so.8.0.1 -> libnvparsers.so.8.0.1
-rwxr-xr-x 1 root root   5275176 juin  13 23:35 /usr/lib/aarch64-linux-gnu/libnvcaffeparser.so.8.2.5
-rw-r--r-- 1 root root   2762290 juin  13 23:35 /usr/lib/aarch64-linux-gnu/libnvcaffeparser_static.a
lrwxrwxrwx 1 root root        29 mai    3  2021 /usr/lib/aarch64-linux-gnu/libnvidia-container.so.0 -> libnvidia-container.so.0.10.0
-rwxr-xr-x 1 root root    142816 mai    3  2021 /usr/lib/aarch64-linux-gnu/libnvidia-container.so.0.10.0
lrwxrwxrwx 1 root root        22 juin  13 17:27 /usr/lib/aarch64-linux-gnu/libnvinfer_plugin.so -> libnvinfer_plugin.so.8
lrwxrwxrwx 1 root root        26 juin  13 17:27 /usr/lib/aarch64-linux-gnu/libnvinfer_plugin.so.8 -> libnvinfer_plugin.so.8.2.5
-rw-r--r-- 1 root root  18280576 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvinfer_plugin.so.8.0.1
-rwxr-xr-x 1 root root   7242296 juin  13 23:35 /usr/lib/aarch64-linux-gnu/libnvinfer_plugin.so.8.2.5
-rw-r--r-- 1 root root   9101914 juin  13 23:35 /usr/lib/aarch64-linux-gnu/libnvinfer_plugin_static.a
lrwxrwxrwx 1 root root        19 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvinfer.so -> libnvinfer.so.8.0.1
lrwxrwxrwx 1 root root        19 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvinfer.so.8 -> libnvinfer.so.8.0.1
-rw-r--r-- 1 root root 188463744 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvinfer.so.8.0.1
-rw-r--r-- 1 root root 333591900 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvinfer_static.a
lrwxrwxrwx 1 root root        20 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvonnxparser.so -> libnvonnxparser.so.8
lrwxrwxrwx 1 root root        24 juin  12 20:18 /usr/lib/aarch64-linux-gnu/libnvonnxparser.so.8 -> libnvonnxparser.so.8.2.5
-rw-r--r-- 1 root root   4876864 juin   6 17:22 /usr/lib/aarch64-linux-gnu/libnvonnxparser.so.8.0.1
-rwxr-xr-x 1 root root   4851096 juin  13 23:35 /usr/lib/aarch64-linux-gnu/libnvonnxparser.so.8.2.5
-rw-r--r-- 1 root root   1816354 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvonnxparser_static.a
lrwxrwxrwx 1 root root        21 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvparsers.so -> libnvparsers.so.8.0.1
lrwxrwxrwx 1 root root        21 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvparsers.so.8 -> libnvparsers.so.8.0.1
-rw-r--r-- 1 root root   3338536 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvparsers.so.8.0.1
-rw-r--r-- 1 root root   4640380 juin  26  2021 /usr/lib/aarch64-linux-gnu/libnvparsers_static.a
-rw-r--r-- 1 root root    420944 sept. 17  2021 /usr/lib/aarch64-linux-gnu/libnvsample_cudaprocess.so

jaybdub · June 14, 2022, 7:20pm

Hi Rdpdo2002,

Thanks for reaching out!

I haven’t yet tested on your exact configuration, but was able to parse the ONNX model you shared with a few configurations. To be specific, the following command

/usr/src/tensorrt/bin/trtexec --onnx=meilleur_model.onnx

Succeeded on the following configurations

Jetson Orin, TensorRT 8.4.0-1 (native)
Jetson Orin, TensorRT 8.4.0-1 (docker, NGC L4T-ML container, tag r34.1.1-py3)
Desktop, TensorRT 8.0.3-1 (docker, NGC PyTorch container, tag 21.09-py3)
Desktop, TensorRT 8.2.1-1 (docker NGC PyTorch container, tag 21.12-py3)

Perhaps the issue is in the source compilation as you mentioned. It may help to try parsing your model inside one of the NGC containers for Jetson to validate that it works on your system.

Please let me know if this helps or you have any questions.

Best,
John

rdpdo2002 · June 14, 2022, 7:43pm

Hello and thank you very much.

I tested my onnx file with version 8001 after uninstalling original version of tensorRT of my jetpack by issuing command

sudo apt-get remove tensorrt
sudo apt-get purge “libnvinfer*”

and reinstalling original trt : sudo apt-get install tensorrt

and it works.

Unfortunately jetpack 4.6 is the lastest available but the tensorrt version on it is too old for my needs… It is why I would like to use the lastest version.
I think that some librairies are not compiled so it do not work well…

I cannot have libnvparsers.so and libnvinfer.so to be build, I think the problem come from it… How can I make these library ? Are they normally build if I launch the make all from tensorrt forlder ? If I use apt install libnvparsers-dev I got v8001 on my jetson nano and not lastest version…

rdpdo2002 · June 15, 2022, 3:20pm

I am pretty sure that it crash because libnvparsers.so and libnvinfer.so are not builded when compiling tensorrt so it used old library version which are in /usr/lib/aarch64-linux-gnu and was installed from previous apt install tensorrt (v8001)

Pergaps you know how to build these library to get lastest version ?

I will try do copy the ones from Jet Pack 5.0 and test if it work better…

rdpdo2002 · June 15, 2022, 7:38pm

I finaly found what was wrong : The build library libnvonnxparser.so.8.2.5 do not work so I need to use the old one ibnvonnxparser.so.8.0.1 and now it works…

system · June 29, 2022, 7:38pm

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

Topic		Replies	Views
ONNX to TensorRT conversion Jetson Nano tensorrt , pytorch , onnx	4	308	July 3, 2024
Problem converting tensorflow model to TensorRT Jetson Nano tensorrt , tensorflow	5	395	March 26, 2024
Encountering error while trying to convert an ONNX model to TensorRT engine - Jetson Orin Nano TensorRT tensorrt , pytorch , cudnn , onnx , jetson	1	72	July 23, 2024
Converting onnx model to trt model (multi view CNN) Jetson Nano tensorrt , onnx	4	565	October 15, 2021
Step-wise procedure to deploy a custom tensorflow 2.4 object detection model in deepstream 5.1 DeepStream SDK tensorrt , tensorflow , jetson-inference	18	2286	September 28, 2021
Trtexec cannot convert QAT onnx model to trt model TensorRT tensorrt	7	1119	January 26, 2023
Run out of memory when creating TensorRT engine from onnx model Jetson Xavier NX tensorrt	7	2672	October 18, 2021
[onnx->TensorRT]Parser problem about padding TensorRT	5	2073	October 19, 2021
Parse onnx file failed: Parameter check failed. condition:condition: allDimsGtEq(windowSize, 1) && volume(windowSize) < MAX_KERNEL_DIMS_PRODUCT(nbSpat TensorRT tensorrt	3	942	May 17, 2022
Tensorrt8.2 convert .onnx model to .trt ,throw "two inputs (data and weights) are allowed only in explicit-quantization mode"error TensorRT	9	1933	April 6, 2022

Help : Error with TensorRT parsing with ONNX file

Related topics