tensorRT for yolo v3 optimization failed ?

I try to use tensorRT for optimization of SSD? it success.
however, when I try to apply trt for yolov3 optimization. it failed.
my configuration is:
tensorflow 1.12
tensorRT 5
cuda 10.0
cudnn 7.3.1

tesla P4 (8G)

the error is reported as follows:
“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”"
E tensorflow/contrib/tensorrt/kernels/trt_engi ne_op.cc:599] Calibration failed: Internal: Failed to build TensorRT engin e
2018-10-15 15:08:06.450968: E tensorflow/contrib/tensorrt/log/trt_logger.c c:38] DefaultLogger Tensor TensorRTInputPH_1 is uniformly zero; network ca libration failed.
terminate called after throwing an instance of ‘std::runtime_error’
what(): Could not find tensor TensorRTInputPH_1 in tensorScales.
./run_all.sh: line 15: 9957 Aborted (core dumped) python test_pb2.py --INT8 --native --num_loops 1 --topN 1 --batch_size 1 --worksp ace_size 3200 --log_file log.txt --img_size 640 --frozen_graph frozen_infe rence_graph3.pb --input_node image_tensor --output_node1 detection_scores --output_node detection_boxes --img_file dog.jpg --update_graphdef
“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”"

i also use the tensorflow 18.09 container for the int8 calibration of yolov3
the error is reported as follows:
“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”"
Traceback (most recent call last):
File “/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/importer.py”, line 418, in import _graph_def
graph._c_graph, serialized, options) # pylint: disable=protected-access
tensorflow.python.framework.errors_impl.InvalidArgumentError: Node ‘my_trt_op_28’: Unknown input node ‘my_tr t_op_14’

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File “test_pb2.py”, line 282, in
timingsint8,comp,_,mdstats=timeGraph(calibGraph,f.batch_size,f.num_loops,dummy_input)
File “test_pb2.py”, line 143, in timeGraph
return_elements=[f.output_node]
File “/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/deprecation.py”, line 454, in new_func
return func(*args, **kwargs)
File “/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/importer.py”, line 422, in import _graph_def
raise ValueError(str(e))
ValueError: Node ‘my_trt_op_28’: Unknown input node ‘my_trt_op_14’
“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”“”"

Hello, it’d help us debug if you can share a small reproduction package that contains the source, model that demonstrates the symptoms are you seeing.

Hi, you can refer this video demonstration YOLO optim. using TensorRT.
[url]06 Optimizing YOLO version 3 Model using TensorRT with 1.5x Faster Inference Time - YouTube