keck91
keck91
I was able to compile. The runtest failed but training and detection are working fine
I had the same issue. After a while I found out that **you have to zoom in and out first**. Then the boxes can be placed.
I used a script based on this [Medium article](https://medium.com/@timothycarlen/understanding-the-map-evaluation-metric-for-object-detection-a07fe6962cf3). You can find it here on [github](https://gist.github.com/tarlen5/008809c3decf19313de216b9208f3734).
Where did you find this in the code?
According to a Report of the Department of Computer Science, Virginia Tech in 2018, DRBox requires 11GB GPU memory. [Source](https://vtechworks.lib.vt.edu/bitstream/handle/10919/83200/SatelliteImageFinderReport.pdf?sequence=19&isAllowed=y) (page 29)
@makefile > @liulei01 My training loss of ship task doesn't decrease and keeps 8.0 ~ 9.0 between iterations of 5000 ~ 20000. I do as the README instruction and used...
> BTW, I am trying to install your modified caffe with CUDNN and OpenCV. > Could you please let me know which version you used? CUDA 8.0, openCV 2.4.9 &...
Is https://wetransfer.com/ fine for you?