Error with converting onnx model

gff1038m · June 12, 2022, 6:38pm

• Jetson AGX Xavier
• Deepstream 6.0
• JetPack 4.6
• TensorRT 8.0.1
• NVIDIA GPU Driver 32.6.1

Hi i’m trying to convert this model: face-recognition-resnet100-arcface-onnx — OpenVINO™ documentation

with this config file:

[property]
gpu-id=0
net-scale-factor=0.0039215697906911373
labelfile-path=/home/cv/Desktop/Mask-Detection/Deepstream-app/labels.txt
model-engine-file=/home/cv/Desktop/Mask-Detection/Deepstream-app/arcfaceresnet100-8.onnx_b1_gpu0_fp32.engine
onnx-file=/home/cv/Desktop/Mask-Detection/Deepstream-app/arcfaceresnet100-8.onnx
infer-dims=3;112;112
uff-input-order=0
uff-input-blob-name=input_1
output-tensor-meta=1
batch-size=1
process-mode=1
model-color-format=0
## 0=FP32, 1=INT8, 2=FP16 mode
network-mode=0
network-type=1
num-detected-classes=1
interval=0
gie-unique-id=1
output-blob-names=output_cov/Sigmoid;output_cov/Sigmoid

But got this error

Now playing...
0:00:00.170113226 20826   0x558420f0c0 INFO                 nvinfer gstnvinfer.cpp:638:gst_nvinfer_logger:<secondary1-nvinference-engine> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::buildModel() <nvdsinfer_context_impl.cpp:1914> [UID = 1]: Trying to create engine from model files
WARNING: [TRT]: onnx2trt_utils.cpp:364: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
WARNING: [TRT]: Detected invalid timing cache, setup a local cache instead
0:00:55.668094488 20826   0x558420f0c0 INFO                 nvinfer gstnvinfer.cpp:638:gst_nvinfer_logger:<secondary1-nvinference-engine> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::buildModel() <nvdsinfer_context_impl.cpp:1947> [UID = 1]: serialize cuda engine to file: /home/cv/Desktop/Mask-Detection/Deepstream-app/arcfaceresnet100-8.onnx_b1_gpu0_fp32.engine successfully
INFO: [Implicit Engine Info]: layers num: 2
0   INPUT  kFLOAT data            3x112x112       
1   OUTPUT kFLOAT fc1             512             

ERROR: [TRT]: Cannot find binding of given name: output_cov/Sigmoid
0:00:55.790537079 20826   0x558420f0c0 WARN                 nvinfer gstnvinfer.cpp:635:gst_nvinfer_logger:<secondary1-nvinference-engine> NvDsInferContext[UID 1]: Warning from NvDsInferContextImpl::checkBackendParams() <nvdsinfer_context_impl.cpp:1868> [UID = 1]: Could not find output layer 'output_cov/Sigmoid' in engine
ERROR: [TRT]: Cannot find binding of given name: output_cov/Sigmoid
0:00:55.790584633 20826   0x558420f0c0 WARN                 nvinfer gstnvinfer.cpp:635:gst_nvinfer_logger:<secondary1-nvinference-engine> NvDsInferContext[UID 1]: Warning from NvDsInferContextImpl::checkBackendParams() <nvdsinfer_context_impl.cpp:1868> [UID = 1]: Could not find output layer 'output_cov/Sigmoid' in engine
0:00:55.803115932 20826   0x558420f0c0 INFO                 nvinfer gstnvinfer_impl.cpp:313:notifyLoadModelStatus:<secondary1-nvinference-engine> [UID 1]: Load new model:/opt/nvidia/deepstream/deepstream-6.0/sources/apps/sample_apps/deepstream-infer-tensor-meta-test/dstest2_sgie1_config.txt sucessfully
Deserialize yoloLayer plugin: yolo_93
Deserialize yoloLayer plugin: yolo_96
Deserialize yoloLayer plugin: yolo_99
0:00:55.946087226 20826   0x558420f0c0 INFO                 nvinfer gstnvinfer.cpp:638:gst_nvinfer_logger:<primary-nvinference-engine> NvDsInferContext[UID 40001]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1900> [UID = 40001]: deserialized trt engine from :/opt/nvidia/deepstream/deepstream-6.0/sources/apps/sample_apps/deepstream-infer-tensor-meta-test/model_b1_gpu0_fp16.engine
INFO: [Implicit Engine Info]: layers num: 4
0   INPUT  kFLOAT data            3x640x640       
1   OUTPUT kFLOAT yolo_93         24x80x80        
2   OUTPUT kFLOAT yolo_96         24x40x40        
3   OUTPUT kFLOAT yolo_99         24x20x20        

0:00:55.946237441 20826   0x558420f0c0 INFO                 nvinfer gstnvinfer.cpp:638:gst_nvinfer_logger:<primary-nvinference-engine> NvDsInferContext[UID 40001]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:2004> [UID = 40001]: Use deserialized engine model: /opt/nvidia/deepstream/deepstream-6.0/sources/apps/sample_apps/deepstream-infer-tensor-meta-test/model_b1_gpu0_fp16.engine
0:00:55.952427215 20826   0x558420f0c0 INFO                 nvinfer gstnvinfer_impl.cpp:313:notifyLoadModelStatus:<primary-nvinference-engine> [UID 40001]: Load new model:/opt/nvidia/deepstream/deepstream-6.0/sources/apps/sample_apps/deepstream-infer-tensor-meta-test/dstest2_pgie_config.txt sucessfully
Decodebin child added: source
Decodebin child added: decodebin0
Running...
Decodebin child added: qtdemux0
Decodebin child added: multiqueue0
Decodebin child added: aacparse0
Decodebin child added: avdec_aac0
Decodebin child added: h264parse0
Decodebin child added: capsfilter0
Decodebin child added: nvv4l2decoder0
Opening in BLOCKING MODE 
NvMMLiteOpen : Block : BlockType = 261 
NVMEDIA: Reading vendor.tegra.display-size : status: 6 
NvMMLiteBlockCreate : Block : BlockType = 261 
In cb_newpad
In cb_newpad
1
File opened
Segmentation fault (core dumped)

And after loading this model (it converted sucessfully) info->inferDims.numElements equals 0

So how to convert this model to .engine?

fanzh · June 13, 2022, 3:01pm

did you still need support?
1 how did you convert model?
2 which deepstream app are you testing?
3 did you modify the code? please provide the diff.

gff1038m · June 13, 2022, 3:23pm

Yes, I’m still trying to make face recognition pipeline and trying different models. I’m just writed the dir where onnx model is. I using the same app deepstream_infer_tensor_meta_test

deepstream_infer_tensor_meta_test.cpp (35.4 KB)
dstest2_pgie_config.txt (3.2 KB)
Makefile (2.6 KB)

gff1038m · June 13, 2022, 3:25pm

I tried to make it with facenet (and I got the same output for different people (idk, maybe there is a problem with cropping the face?), tried with landmarks (you seen the problem, and I don’t understand how landmarks can help me with recognition) and I tried this model

fanzh · June 16, 2022, 2:48pm

1 using your code, I can built, but can’t run, please provide a simple code to reproduce, thanks.
2 you can use deepstream-infer-tensor-meta-test to test this model, please make sure the preprocess and postprocess are right.

gff1038m · June 19, 2022, 3:57pm

Hi, I used another model and it worked well

system · July 3, 2022, 3:57pm

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