darknet: ./src/utils.c:325: error: Assertion `0' failed (YOLO-v4) Google colab

Question

I try to use Google Colab for training YOLO_v4 on my own data set and met the next error:

CUDA status Error: file: ./src/blas_kernels.cu : () : line: 841 : build time: Oct  9 2020 - 12:04:13 

 CUDA Error: no kernel image is available for execution on the device
CUDA Error: no kernel image is available for execution on the device: File exists
darknet: ./src/utils.c:325: error: Assertion `0' failed.
CPU times: user 58 ms, sys: 12.4 ms, total: 70.4 ms
Wall time: 15.4 s

I've checked my code but can't find mistake. Then I ran the Google Colab notebook with example I used(https://colab.research.google.com/drive/1cbzkKcACQPzLXuBYH7KPOh73Ko8xdUPn#scrollTo=6miYFbvExqMd). When I execute the cell:

%%time
!./darknet detector train data/obj.data cfg/custom-yolov4-detector.cfg -dont_show -map #./backup/custom-yolov4-detector_last.weights
#If you get CUDA out of memory adjust subdivisions above!
#adjust max batches down for shorter training above

this error appears sometimes do not. I'm so confused... I think that it is somehow related to availability or type of GPU that I am connected. Any suggestions?

Answer 1

check this link and download darknet like what is hereenter link description here

then create a yolov4.conf file and config it base on your number of class labels. you can find an example of this configuration here:enter link description here