Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU) GPU • DeepStream Version 6.2 • JetPack Version (valid for Jetson only) • TensorRT Version 8.5.2.2 • NVIDIA GPU Driver Version (valid for GPU only) 525.85.12 • Issue Type( questions, new requirements, bugs) question • How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing) • Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
I am taking multiple streams of input, so I am using the streammux element to form a batched buffer. I understand that the source connected to the Sink_N pad will have pad_indexN in NvDsBatchMeta .
How do I take this pad_index and pass it to the Triton model as the input, together with the frame data? This is so that my Triton model knows which stream the frame comes from.
you might use pad_index of NvDsFrameMeta, it means “Holds the pad or port index of the Gst-streammux plugin for the frame in the batch.” , please refer to DeepStream SDK’s nvdsmeta.h.
But how do I access this pad_index and pass it as input for the triton model? Attached is my infer_config.txt for the nvdsinferserver. Ideally I want another input layer, say ‘PAD_INDEX’ along with my ‘INPUT’ to pass to the triton model as inputs.
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks
you can use nvinferserver’s IInferCustomProcessor interface, extraInputProcess can generate extra tensor input data, please refer to DeepStream SDK sample: opt\nvidia\deepstream\deepstream\sources\TritonOnnxYolo\nvdsinferserver_custom_impl_yolo\nvdsinferserver_custom_process_yolo.cpp
BTW, nvinferserver is opensource from 6.2, you might add logs to debug.