DOTA_devkit
DOTA_devkit copied to clipboard
您好,请教一下,我用dotav1数据集进行了训练,然后分别用 dota_evaluation_task1和 dota_evaluation_task2计算出的map值差距很大,请问这两个evaluation有什么区别吗
can't open DOTAv1.5 test set evalutation server
您好,非常感谢您的工作!我有一个问题希望能得到您的解答:在计算task1的AP值时,计算正样本数量使用了npos = npos + sum(~difficult)这个来计算,请问为什么分类难度为difficult的样本不被计算在正样本之内呢?
As the question states, is DOTA_Devkit licensed under MIT, Simplified BSD or some other license? [Here](https://github.com/CAPTAIN-WHU/DOTA_devkit/blob/99388551054be9a6dabb01c8bb2a7eb562d57b4f/DOTA.py#L2) you say Simplified BSD but in the evaluation scripts you say MIT.
Hi,My data set is cropped images and annotations. The results I get are also converted using parse_results.py. When I use dota_evaluation_task1.py for evaluation, what file formats do detpath, annopath, and...
请问dota数据集官网没有提供test集合中的标注信息,可以从哪里获取呢?
May I ask what tools do you use for marking? I need to make my own dataset.
In my label file some file named `P1670_0`, `1650_0`, somelike that. But I cannot find corresponding image in my imagsets. the length of `labelTxt` and `Imagests` are same. Any idea...
Hi, I was playing with your scripts recently, and I noticed that you applied NMS when you merged back crops to the original image space. I was trying to use...
请问有人遇到这个问题吗 E:\数据预处理\DOTA_devkit>python setup.py build_ext --inplace running build_ext building '_polyiou' extension d:\Program Files (x86)\Microsoft Visual Studio\2017\Enterprise\VC\Tools\MSVC\14.16.27023\bin\HostX86\x64\cl.exe /c /nologo /Ox /W3 /GL /DNDEBUG /MD -ID:\python3.7\anzhaung\include -ID:\python3.7\anzhaung\include "-Id:\Program Files (x86)\Microsoft Visual Studio\2017\Enterprise\VC\Tools\MSVC\14.16.27023\ATLMFC\include" "-Id:\Program...