OrientedRepPoints

OrientedRepPoints copied to clipboard

版本

2

只能用mmdet1.1版本么

zjutlouis1011

Parse the pkl of results Now, I have trained the model and get the results.pkl.

6

How to parse the results.pkl to the format needed for DOTA evaluation (Class AP and mAP )? Looking forward for your reply, thanks

houliping

Memory Leak--UserWarning: semaphore_tracker: There appear to be 14 leaked semaphores to clean up at shutdown len(cache))

2

DOTA数据集的规模暂时还比较小，所以目前没发生过因为内存泄露而导致的程序崩溃现象。 但当我将该项目应用在一个30W规模的数据集上的时候，随着迭代次数的增加内存的使用量也在增加，最后崩溃。 下图是每个小时自动运行一次free命令的输出情况。  mmdet和python本身应该不容易出现内存管理方面的问题，感觉可能是新增的cuda代码容易出问题，但是cuda这一块我完全不懂，作者方便定位一下问题可能出现在哪些地方吗

hukaixuan19970627

the classes num in dataset doesn't correspond to the parameter (bbox_head) "num_classes" in config file. Could you explain it :)

3

for example, i find you use "wordname_15" dataset, the num classes of this dataset is [15](https://github.com/LiWentomng/OrientedRepPoints/blob/main/DOTA_devkit/DOTA2COCO_poly.py), but in your config file, the parameter “num_classes" is [16](https://github.com/LiWentomng/OrientedRepPoints/blob/main/configs/dota/orientedrepoints_swin_tiny_demo.py). The values ​​of these...

Xiloy

预训练模型

2

您好，请问可以提供一下swin_tiny的pytorch为1.4版本的预训练模型吗，现在网上所支持的pytorch版本都太高了，找不到相应的预训练模型。谢谢！

lizifan123

如何训练UCAS-AOD数据集

1

jimuIee

 我想知道这个里面在评估APAA的作用时使用了分类得分和定位得分的相关性，那这个定位得分是啥？我在代码中没有看到这个评估

Zhaowss

搭建环境遇到坑的去看这篇文章https://blog.csdn.net/SSSlasH/article/details/125255955

3

https://blog.csdn.net/SSSlasH/article/details/125255955

hawk2333

Traceback (most recent call last): File "/home/lzk/mydata/anaconda3/envs/reppointobb/lib/python3.8/runpy.py", line 194, in _run_module_as_main return _run_code(code, main_globals, None, File "/home/lzk/mydata/anaconda3/envs/reppointobb/lib/python3.8/runpy.py", line 87, in _run_code exec(code, run_globals) File "/home/lzk/.vscode/extensions/ms-python.python-2023.14.0/pythonFiles/lib/python/debugpy/adapter/../../debugpy/launcher/../../debugpy/__main__.py", line 39, in cli.main() File...

jimuIee

Hello, I'm testing your code on Ascend NPU devices. The GIoU Loss in your reports contains several cuda ops from mmcv. What about the implementation of the pure torch-based GIoU...

Balabala-Hong