search

lewes6369 / TensorRT-Yolov3

TensorRT for Yolov3
MIT License
489 stars 165 forks source link

can not find output named yolo-det #72

Closed jgqandsh closed 4 years ago

jgqandsh commented 4 years ago

my develop environmnet is ubuntu18.04, cuda10.0, cudnn 7.6 , tensorrt 6.0.1.5 when i run ./install/runYolov3 --caffemodel=./yolov3_608.caffemodel --prototxt=./yolov3_608.prototxt --input=./test.jpg --W=608 --H=608 --class=80 i get the info:

####### input args####### C=3; H=608; W=608; batchsize=1; caffemodel=./yolov3_608.caffemodel; calib=; class=80; enginefile=; evallist=; input=./test.jpg; mode=fp32; nms=0.450000; outputs=yolo-det; prototxt=./yolov3_608.prototxt; ####### end args####### init plugin proto: ./yolov3_608.prototxt caffemodel: ./yolov3_608.caffemodel Begin parsing model... End parsing model... can not find output named yolo-det Segmentation fault (core dumped)

Can you help me? Thank you

mbahaa commented 4 years ago

You need to load the tensorrt-compatible prototxt file yolov3_608_trt.prototxt not yolov3_608.prototxt

jgqandsh commented 4 years ago

@mbahaa

You need to load the tensorrt-compatible prototxt file yolov3_608_trt.prototxt not yolov3_608.prototxt

I changed the yolov3_608.prototxt to yolov3_608_trt.prototxt. It works well. Thank you very much!. I will close the issue. thank you again.

jgqandsh commented 4 years ago

@mbahaa told me to change the yolov3_608.prototx to yolov3_608_trt.prototxt, it runned well, i get a output. Thanks his answer.