pull container 19.02 py2 / 19.12 py3
docker pull nvcr.io/nvidia/tensorrt:19.02-py2
docker pull nvcr.io/nvidia/tensorrt:19.12-py3docker build -t trt_yolov3 .
docker run --rm -it --privileged --runtime=nvidia trt_yolov3 bash
python3 onnx_to_tensorrt.py --build --vram 8 --max_batch_size 64 -p fp16--privileged --runtime=nvidia
(19.02 for yolo2onnx , TRT 5.0.2 , onnx 1.3.0 or 1.2.3 )
docker run --rm -it --privileged --runtime=nvidia -v ~/Desktop/python:/sharefolder nvcr.io/nvidia/tensorrt:19.02-py2 bash
pip install wget onnx==1.3.0
cd tensorrt/samples/python/yolov3_onnx/ && python yolov3_to_onnx.py
cp yolov3.onnx /sharefolder/(19.12 for onnx2trt , TRT 6.0.1.8 , onnx 1.7.0)
docker run --rm -it --privileged --runtime=nvidia -v ~/Desktop/python:/sharefolder nvcr.io/nvidia/tensorrt:19.12-py3 bash
pip3 install wget onnx scipy line_profiler
git clone https://github.com/royinx/trt_yolov3.git
cp /sharefolder/yolov3.onnx trt_yolov3/ && cd trt_yolov3
python onnx_to_tensorrt.py --build --vram 8 --max_batch_size 64 -p fp16commit new image and run at a new container
install gdb apt-get update && apt-get -y install gdb python2.7-dbg
$ cd
$ gdb python(gdb) run XXX.py
(gdb) generate-core-file
(gdb) py-bt
(gdb) q
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent