Output-tensor-meta Access RAW model output with batch dimension

[johannesrhvw]

Done here:

Poor performance due to INMSLayers followup nms_layer_output(1)[DevicetoShapeHostCopy] and trainstation

Description I am using this repository as a reference to include output tensor parsing and clustering in the yolo segmentation model using onnx graphsurgeon https://github.com/marcoslucianops/DeepStream-Yolo-Seg/tree/master . Now i am facing large latencies on a call after NonMaximumSuppression layer, called nms_layer_output(1)[DevicetoShapeHostCopy] and an added trainstation syncing with Host. As stated in https://forums.developer.nvidia.com/t/inmslayer-cuda-graph-invalidation-devicetoshapeho…