search

AlexeyAB / darknet

YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )
http://pjreddie.com/darknet/
Other
21.65k stars 7.95k forks source link

Segmentaion Error #642

Closed veer345 closed 3 years ago

veer345 commented 6 years ago

I am using linux 16.04 intel i3 error- Not used AVX Segmentation fault (core dumped) what should i do ... I am really tired now i have been trying from so many days ..... Please help me how should i proceed..

AlexeyAB commented 6 years ago
  1. Can you show screenshot of the error?
  2. What parameters in the Makefile do you use? https://github.com/AlexeyAB/darknet/blob/17d242bd506f1b94501599d68864ba36eebd2837/Makefile#L1-L6
  3. What command line do you use for runing darknet?
  4. Replace file src/gemm.c to this one: https://raw.githubusercontent.com/AlexeyAB/darknet/f1ad9a0487dc60c01e56bacb2a67de1e1006b4b7/src/gemm.c

then do

make clean
make -j8
veer345 commented 6 years ago

7 and what about gemm.h file

veer345 commented 6 years ago

same i am not making any changes in make file

veer345 commented 6 years ago

what process i follow :: Step1-Downloaded ur darknet Step 2- running make command in darknet folder(not in build/64x/) step3-creating images and txt files step4- creating yolo-obj-cfg file step4-test and train text file step5- darknet19_448.conv.23 copying to darknet directory step 6 running the train command

veer345 commented 6 years ago

after replaceing gemm file its running let's now hope for good results. THANK YOU so much :+1:

AlexeyAB commented 6 years ago

@veer345

I just fixed this bug in the last commit.

Now it can successfully detect used AVX or not used AVX on CPU on Linux without Segmentaion Error: image

veer345 commented 6 years ago

Couldn't open file: obj.names now this error comes ... after 2700 iteration.. Please help me out

AlexeyAB commented 6 years ago

@veer345 Update your code from GitHub.