guohaoyuan

请问如何将DOTA数据的8点坐标转换成符合本代码运行规范的五点坐标呢？

2

首先很感谢作者的贡献！ 因为之前看到的角度都是利用cv2.minAreaRect得到的，范围是（-90,0），位于第一象限。 但是本算法需要的角度是labelImg2得到的，范围（0， 180）,位于第一、第四象限，不知道如何转换. 有知道的小伙伴没？

train error, RuntimeError: CuDNN error: CUDNN_STATUS_MAPPING_ERROR

4

loading all datasets... using 4 threads loading from cache file: cache/coco_trainval2014.pkl No cache file found... loading annotations into memory... Done (t=7.66s) creating index... index created! 82783it [00:23, 3500.57it/s] loading annotations...

RuntimeError: CUDNN_STATUS_MAPPING_ERROR

1

python36 cuda80 cudnn7 torch 0.4.0 I've tried to reduce BS and add CUDA () But it didn't work

RuntimeError: cuda runtime error (38) : no CUDA-capable device is detected at /opt/conda/conda-bld/pytorch_1532584813488/work/aten/src/THC/THCGeneral.cpp:74

8

Thank you for this code, it's amazing! I've been learning about your project recently, but this error occurred at the beginning of the training. I thought the graphics card didn't...

ship R:0.254256272401 P:0.958479943702 ap:0.245356778084 F:0.401900321638 ship R:0.248312385783 P:0.907084468665 ap:0.228256838226 F:0.389892542383 horizontal standard R:0.254256272401 P:0.958479943702 mAP:0.245356778084 F:0.401900321638 rotate standard R:0.248312385783 P:0.907084468665 mAP:0.228256838226 F:0.389892542383 请问，有人做过二分类实验吗？

dota数据集中的船只head_x, head_y文件没有找到

1

你好，感谢你的代码，让我学到了许多。 有个问题就是关于图像的head_x,head_y数据的标签，没有找到，需要我们自己标吗？