LPR - TensorRT build not working on DS 6.0

rmpt · February 2, 2022, 3:10am

Deepstream version: 6.0
Platform: dGPU

According to the License Plate Recognization sample the usage of tao-converter should not be required on DS 6.0:

NVIDIA-AI-IOT/deepstream_lpr_app/blob/master/deepstream-lpr-app/lpr_config_sgie_us.txt#L49-L73


      
          [property]
          gpu-id=0
          model-engine-file=../models/LP/LPR/us_lprnet_baseline18_deployable.etlt_b16_gpu0_fp16.engine
          labelfile-path=../models/LP/LPR/labels_us.txt
          tlt-encoded-model=../models/LP/LPR/us_lprnet_baseline18_deployable.etlt
          tlt-model-key=nvidia_tlt
          batch-size=16
          ## 0=FP32, 1=INT8, 2=FP16 mode
          network-mode=2
          num-detected-classes=3
          gie-unique-id=3
          output-blob-names=output_bbox/BiasAdd;output_cov/Sigmoid
          #0=Detection 1=Classifier 2=Segmentation
          network-type=1
          parse-classifier-func-name=NvDsInferParseCustomNVPlate
          custom-lib-path=../nvinfer_custom_lpr_parser/libnvdsinfer_custom_impl_lpr.so
          process-mode=2
          operate-on-gie-id=2
          net-scale-factor=0.00392156862745098
          #net-scale-factor=1.0

This file has been truncated. show original

I’m trying to load us_lprnet_baseline18_deployable.etlt directly but it results in the following error when building the TensorRT engine:

INFO                 nvinfer gstnvinfer.cpp:638:gst_nvinfer_logger:<model_inference3> NvDsInferContext[UID 3]: Info from NvDsInferContextImpl::buildModel() <nvdsinfer_context_impl.cpp:1914> [UID = 3]: Trying to create engine from model files
ERROR: [TRT]: 10: [optimizer.cpp::computeCosts::2011] Error Code 10: Internal Error (Could not find any implementation for node {ForeignNode[lstm_W...Max_reduce_min]}.)
ERROR: ../nvdsinfer/nvdsinfer_model_builder.cpp:1119 Build engine failed from config file

Is this model supported natively or not?

mchi · February 13, 2022, 1:48pm

I can’t reproduce this issue with command below on DS6.0GA+dGPU

git clone https://github.com/NVIDIA-AI-IOT/deepstream_lpr_app.git
cd deepstream_lpr_app/
./download_us.sh
cd deepstream-lpr-app/
cp dict_us.txt dict.txt
./deepstream-lpr-app 1 2 0 /opt/nvidia/deepstream/deepstream/samples/streams/sample_1080p_h264.mp4 output.264

system · March 8, 2022, 2:15am

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