darknet_face_with_landmark
darknet_face_with_landmark copied to clipboard
make error
version.h fatal error
I upload the version.h, you can pull and try again!
ok,thank you ! but it can not train, also; ./darknet detector train data/face.data cfg/mbv2_yolov3_face.cfg -gpus 1,2,3,4 ,it's wrong?
@huanpig Do you have any specific error messages?
Create 64 permanent cpu-threads or min_batch = * batch = * time_steps = *, it will stop
@huanpig Try training on a single gpu,maybe you should change the batch and subdivisions
no ,it will stop again
what's the difference bwtween origin darknet and yours?
@huanpig I only change the data.c yolo_layer.c utils.c
it can run at your computer?
@huanpig yes! you can run AlexeyAB's darknet?
can run origin darknet, Alex's ,,, i do not know
error in ./darknet double free or corruption(!prev)
and i want to konw , what's the meaning of -1 in retinaface's landmark?thank you
@huanpig -1 means not mark this face, because to small or blur
how about face cut off by picture,for example , only eye is reveal, it's also -1 for all landmark?
Warning: in txt-labels class_id = -1 > = classes = 1 in cfg-file, in txt-labels class_id should be [from 0 to 0]. truth.x = -1.000 ........ when run widerface data,it will show these and stop ,is threr some errors in cfg.file?
./src/yolo_layer.c:724:21: error: ‘for’ loop initial declarations are only allowed in C99 mode for (int i = 0 ; i< l.classes + 1 ; i++){ ^ compilation terminated due to -Wfatal-errors.
?? centos7 gcc-4.8.5-36.el7_6.2.x86_64
Warning: in txt-labels class_id = -1 > = classes = 1 in cfg-file, in txt-labels class_id should be [from 0 to 0]. truth.x = -1.000 ........ when run widerface data,it will show these and stop ,is threr some errors in cfg.file?
you should modify data.c by change 5*boxes to (10+4+1)*boxes