Custom model: Cannot find binding of given name: data

[brking]

We trained our own model and are trying to load it. I just put the onnx and label files in a folder under my home folder, and try to run detectnet using the paths to the onnx and label files:

detectnet rtsp://…/live1s2.sdp --input-codec=h264 --model=/home/brad/eoi-models/mosaic-parkinglot-ssd/vgg16-ssd-Epoch29-Loss-1.05.onnx --labels=/home/brad/eoi-models/mosaic-parkinglot-ssd/labels.txt

[TRT] INVALID_ARGUMENT: Cannot find binding of given name: data
[TRT] failed to find requested input layer data in network
[TRT] device GPU, failed to create resources for CUDA engine
[TRT] failed to create TensorRT engine for /home/brad/eoi-models/mosaic-parkinglot-ssd/vgg16-ssd-Epoch29-Loss-1.05.onnx, device GPU
[TRT] detectNet – failed to initialize.
detectnet: failed to load detectNet model

[dusty_nv]

Hi @brking, you need to specify the input/output layer names like is shown here:

• https://github.com/dusty-nv/jetson-inference/blob/master/docs/pytorch-ssd.md#processing-images-with-tensorrt

If you are using train_ssd.py normally this would be --input-blob=input_0 --output-cvg=scores --output-bbox=boxes

[brking]

@dusty_nv

–output-bbox=boxes

thanks!