Training Face Detection model on custom dataset using caffe

, ,
1

I am following the steps from https://github.com/opencv/opencv/blob/master/samples/dnn/face_detector/how_to_train_face_detector.txt to train Face detection model on custom dataset.

I am using Ubuntu 20.04 based docker container. I was able to complete all the steps but I am getting in final training step. I am getting this error while doing it:

I0427 08:47:07.764628    26 net.cpp:283] Network initialization done.
[libprotobuf ERROR google/protobuf/text_format.cc:307] Error parsing text-format caffe.NetParameter: 1813:9: Message type "caffe.DetectionOutputParameter" has no field named "clip".
F0427 08:47:07.765632    26 upgrade_proto.cpp:88] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: test.prototxt
*** Check failure stack trace: ***
    @     0x7fc08ccf71c3  google::LogMessage::Fail()
    @     0x7fc08ccfc25b  google::LogMessage::SendToLog()
    @     0x7fc08ccf6ebf  google::LogMessage::Flush()
    @     0x7fc08ccf76ef  google::LogMessageFatal::~LogMessageFatal()
    @     0x7fc08d10751a  caffe::ReadNetParamsFromTextFileOrDie()
    @     0x7fc08cf52ede  caffe::Solver<>::InitTestNets()
    @     0x7fc08cf5501d  caffe::Solver<>::Init()
    @     0x7fc08cf5528e  caffe::Solver<>::Solver()
    @     0x7fc08d175a88  caffe::Creator_SGDSolver<>()
    @     0x55c12a8bfee0  caffe::SolverRegistry<>::CreateSolver()
    @     0x55c12a8b9098  train()
    @     0x55c12a8b5c2c  main
    @     0x7fc08c0670b3  __libc_start_main
    @     0x55c12a8b662e  _start
root@42e16ec93639:/home/caffe/FaceDetection#

I found this https://github.com/opencv/opencv/pull/14938 in opencv but I am not sure what to do. Any help is appreciated. Thanks.

2

opencv version ? do you use your own protobuf or the 3rdparty one ?

neither net.cpp nor upgrade_proto.cpp are part of opencv,
so where does it come from ?