I got a big error

[Airyzf]

nvidia@tegra-ubuntu:/demo/tf_to_trt_image_classification/build$ cmake ..
-- Configuring done
-- Generating done
-- Build files have been written to: /home/nvidia/demo/tf_to_trt_image_classification/build
nvidia@tegra-ubuntu:/demo/tf_to_trt_image_classification/build$ make
[ 33%] Built target classify_image
[ 50%] Building CXX object src/CMakeFiles/uff_to_plan.dir/uff_to_plan.cpp.o
/home/nvidia/demo/tf_to_trt_image_classification/src/uff_to_plan.cpp: In function ‘int main(int, char**)’:
/home/nvidia/demo/tf_to_trt_image_classification/src/uff_to_plan.cpp:71:79: error: no matching function for call to ‘nvuffparser::IUffParser::registerInput(const char*, nvinfer1::DimsCHW)’
parser->registerInput(inputName.c_str(), DimsCHW(3, inputHeight, inputWidth));
^
In file included from /home/nvidia/demo/tf_to_trt_image_classification/src/uff_to_plan.cpp:12:0:
/usr/include/aarch64-linux-gnu/NvUffParser.h:182:18: note: candidate: virtual bool nvuffparser::IUffParser::registerInput(const char*, nvinfer1::Dims, nvuffparser::UffInputOrder)
virtual bool registerInput(const char* inputName, nvinfer1::Dims inputDims,
^
/usr/include/aarch64-linux-gnu/NvUffParser.h:182:18: note: candidate expects 3 arguments, 2 provided
src/CMakeFiles/uff_to_plan.dir/build.make:62: recipe for target 'src/CMakeFiles/uff_to_plan.dir/uff_to_plan.cpp.o' failed
make[2]: *** [src/CMakeFiles/uff_to_plan.dir/uff_to_plan.cpp.o] Error 1
CMakeFiles/Makefile2:160: recipe for target 'src/CMakeFiles/uff_to_plan.dir/all' failed
make[1]: *** [src/CMakeFiles/uff_to_plan.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2
nvidia@tegra-ubuntu:~/demo/tf_to_trt_image_classification/build$

[ghost]

Hi Airyzf.

This is likely an API discrepancy between TensorRT 3 and TensorRT 4.

Could you try compiling after checking out the branch trt_4plus?

git checkout trt_4plus

Thanks,
John

[Airyzf]

thanks, it works.

[Airyzf]

I got an other problem. I collected some data ,trained with yolov3 model and output a "yolov3.pb" file.
But the step "Convert frozen graph to TensorRT engine" need a output node name, how can i get the output node name? please help ,thanks.

[ghost]

I don't have experience with YoloV3 specifically, but it may help to visualize the frozen graph using Tensorboard. There should be plenty of documentation on this but I'm happy to provide more details if needed.

John

[ghost]

Also, note that many object detection models have layers that are not natively supported by TensorRT. For these cases it may be easiest to try using TensorRT integration inside TensorFlow.

We have examples for doing this using models from the TensorFlow object detection API here

https://github.com/NVIDIA-Jetson/tf_trt_models

John

[Airyzf]

ok,thanks

[Airyzf]

I don't have experience with YoloV3 specifically, but it may help to visualize the frozen graph using Tensorboard. There should be plenty of documentation on this but I'm happy to provide more details if needed.

John

Hi,John
When I run "Convert frozen graph to TensorRT engine" with other tensorflow frozen graph, I always got the other exception "list indices must be integers, not AttrValue".

python scripts/convert_plan.py data/frozen_graphs/test.h5.pb data/plans/test.h5.pb.plan input 416 416 output_node0 1 0 float
Using output node output_node0
Converting to UFF graph
Warning: No conversion function registered for layer: Identity yet.
Converting as custom op Identity output_node0
name: "output_node0"
op: "Identity"
input: "activation_2/Softmax"
attr {
key: "T"
value {
type: DT_FLOAT
}
}

Traceback (most recent call last):
File "scripts/convert_plan.py", line 71, in
data_type
File "scripts/convert_plan.py", line 22, in frozenToPlan
text=False,
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/conversion_helpers.py", line 103, in from_tensorflow_frozen_model
return from_tensorflow(graphdef, output_nodes, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/conversion_helpers.py", line 75, in from_tensorflow
name="main")
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py", line 64, in convert_tf2uff_graph
uff_graph, input_replacements)
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py", line 51, in convert_tf2uff_node
op, name, tf_node, inputs, uff_graph, tf_nodes=tf_nodes)
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py", line 28, in convert_layer
fields = cls.parse_tf_attrs(tf_node.attr)
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py", line 177, in parse_tf_attrs
for key, val in attrs.items()}
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py", line 177, in
for key, val in attrs.items()}
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py", line 172, in parse_tf_attr_value
return cls.convert_tf2uff_field(code, val)
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py", line 146, in convert_tf2uff_field
return TensorFlowToUFFConverter.convert_tf2numpy_dtype(val)
File "/usr/local/lib/python2.7/dist-packages/uff/converters/tensorflow/converter.py", line 74, in convert_tf2numpy_dtype
return np.dtype(dt[dtype])
TypeError: list indices must be integers, not AttrValue

[mitesh741]

I have Tensor RT 5.0.6 included in Jetpack for Xavier.
I got the same error.
Can you please ?

[AHA03926]

I also have same problem.
I have Tensor RT 5.0.6 included in Jetpack for TX2
please gime a solution for this?

===============================
make
・
・
・
-- Configuring done
-- Generating done
-- Build files have been written to: /home/nvidia/project4/tf_to_trt_image_classification/build
[ 16%] Linking CXX executable classify_image
[ 33%] Built target classify_image
[ 50%] Building CXX object src/CMakeFiles/uff_to_plan.dir/uff_to_plan.cpp.o
/home/nvidia/project4/tf_to_trt_image_classification/src/uff_to_plan.cpp: In function ‘int main(int, char**)’:
/home/nvidia/project4/tf_to_trt_image_classification/src/uff_to_plan.cpp:71:79: error: no matching function for call to ‘nvuffparser::IUffParser::registerInput(const char*, nvinfer1::DimsCHW)’
   parser->registerInput(inputName.c_str(), DimsCHW(3, inputHeight, inputWidth));
                                                                               ^
In file included from /home/nvidia/project4/tf_to_trt_image_classification/src/uff_to_plan.cpp:12:0:
/usr/include/aarch64-linux-gnu/NvUffParser.h:182:18: note: candidate: virtual bool nvuffparser::IUffParser::registerInput(const char*, nvinfer1::Dims, nvuffparser::UffInputOrder)
     virtual bool registerInput(const char* inputName, nvinfer1::Dims inputDims, UffInputOrder inputOrder) = 0;
                  ^~~~~~~~~~~~~
/usr/include/aarch64-linux-gnu/NvUffParser.h:182:18: note:   candidate expects 3 arguments, 2 provided
src/CMakeFiles/uff_to_plan.dir/build.make:62: recipe for target 'src/CMakeFiles/uff_to_plan.dir/uff_to_plan.cpp.o' failed
make[2]: *** [src/CMakeFiles/uff_to_plan.dir/uff_to_plan.cpp.o] Error 1
CMakeFiles/Makefile2:160: recipe for target 'src/CMakeFiles/uff_to_plan.dir/all' failed
make[1]: *** [src/CMakeFiles/uff_to_plan.dir/all] Error 2
Makefile:83: recipe for target 'all' failed
make: *** [all] Error 2
nvidia@jetsontx2:~/project4/tf_to_trt_image_classification/build$