i'm trying to train yolov4 on google colab here's the code i used to change makefile
%cd darknet
!sed -i 's/OPENCV=0/OPENCV=1/' Makefile
!sed -i 's/GPU=0/GPU=1/' Makefile
!sed -i 's/CUDNN=0/CUDNN=1/' Makefile
!sed -i 's/CUDNN_HALF=0/CUDNN_HALF=1/' Makefile
!sed -i 's/LIBSO=0/LIBSO=1/' Makefile
!make
but when i train it still trains on cpu it displays this while training
(next mAP calculation at 1000 iterations)
25: 1158.892578, 1137.819946 avg loss, 0.000000 rate, 3.695665 seconds, 1600 images, 214.912276 hours left
Loaded: 3.765689 seconds - performance bottleneck on CPU or Disk HDD/SSD
214 hours is literally too much time to train so my assumption that my model train using cpu not the gpu so how can i make train using the gpu?