Which .prototxt while generating caffemodel?

[lorenzobattelli]

Dear all, I'm not sure about which prototxt should I use when i attempt to convert the darknet model to a caffemodel. Since I've cloned the repo, I possess 2 prototxts into the 2 directories:

dpu_yolov4/dpu_yolov4.prototxt
dpu_yolov4_voc/dpu_yolov4_voc.prototxt

Should I consider one on these 2 files, or should I generate my prototxt from my weights file and cfg? in the second case; how to generate it?

Many thanks ☺️

[huanghuan1]

can you solve this problem? please share your solutions

[lorenzobattelli]

next I noticed that the script darknet_convert.sh actually generates a very long prototxt, so I used that one to continue of with the tutorial, and proceeding with the conversion and quantization

[huanghuan1]

next I noticed that the script darknet_convert.sh actually generates a very long prototxt, so I used that one to continue of with the tutorial, and proceeding with the conversion and quantization Thank you for your reply. I'll try it, too

[cq182]

when I use the command 'bash scripts/darknet_convert.sh' ,why did I meet this problem?

if I add sudo before that command,which becomes..

I'm really confused Thanks for your help!

[code-locker]

Hi @cq182, I think you need to install caffe framework manually inside Vitis-AI environment. For the same I request you to go through Vitis-AI userguide for installation procedure. Thanks.

[bhargavin1872008]

when running the requirements.txt of keras-yolov3-modelset -i 'm getting error for coremltools.it is showing like "couldn't find a version that satisfies the requirement tensorflow<=1.14 and tensorflow >=1.5(from tfcoremltools -r requirements.txt).(from version :2.2.0,2.2..1, 2.2.2, ...2.7.0rc0,2.7.0.rc1............) like this .can someone help me regarding this. Also ,i have a doubt .can we use ubuntu 20.04 ,cuda 11.7 ,cudnn 8.4.0 for this project. or have to use ubuntu 18.04,cuda 10.0 only which only works.please help me regarding this,i have less time in my hand.