darknet_for_colab

darknet_for_colab copied to clipboard

While started training ``` !./darknet detector train data/yolov4.data cfg/yolov4_custom_train.cfg yolov4.conv.137 -dont_show -map ``` I got this error. ``` CUDA status Error: file: ./src/dark_cuda.c : () : line: 454 : build...

HariTarz

Fix colab makefile error (nvcc fatal : Unsupported gpu architecture 'compute_30')

1

runesc

tsuhadaniel

 i want to save all th chart.png even i re-train my model thank you 

amirabenrabah

while training ``` !./darknet detector train data/yolov4.data cfg/yolov4_custom_train.cfg yolov4.conv.137 -dont_show -map``` I get this. Any thoughts on why this is happening? ``` CUDA allocate done! Loaded: 5.100706 seconds CUDA status...

gandharvsuri

How can I make the darknet_video.py produce higher quality video? What should I modify?

lenard26

Make failed saying *nvcc fatal : Unsupported gpu architecture 'compute_30'*

6

Support for 'compute_30' is removed from CUDA version 10.2. So it is unable to compile for this architecture.

nagasanjay

RabiaCodes

Hi, I trained YOLOv4 with my custom dataset. Then I want to test it, and It creates the predictions.png file correctly with the box in every object but without label...

alessioBelli

This solves the error while compiling #2 ...

nagasanjay