LiWentomng/OrientedRepPoints

Issues

• 如何训练UCAS-AOD数据集

  #54 opened a year ago
  1

• 怎么复现出论文精度，除了使用设备不一致以外，其他实验设置一直，但是跑出来的DOTA数据集精度相差论文里给出的3-4个点（R50-FPN)

  #53 opened a year ago
  0

• 作者你好，我想请问这个里面的定位得分是啥？

  #52 opened a year ago
  0

• view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.

  #51 opened a year ago
  0

• Question about the implementation of ConvexGIoULoss

  #50 opened a year ago
  0

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

  #49 opened a year ago
  3

• install.md 的指南是谁写的？conda 安装 Pytorch，然后用 pip 安装依赖项是吧？

  #48 opened a year ago
  0

• importerror

  #47 opened a year ago
  0

• 数据集

  #46 opened a year ago
  0

• NearestGTCorner装换函数

  #45 opened a year ago
  0

• how to add reppoints to output result

  #44 opened a year ago
  1

• 请问这个num_points的作用是什么呢？为什么设定为9.谢谢

  #43 opened a year ago
  0

• 为什么执行CUDA_VISIBLE_DEVICES=0,1,2,3 tools/dist_train.sh configs/dota/orientedrepoints_r101_demo.py 4 不报错，执行CUDA_VISIBLE_DEVICES=0,1,2,3 tools/dist_train.sh configs/dota/orientedrepoints_r50_demo.py 4就报下边的错误？

  #42 opened a year ago
  0

• Questions about OrientedRepPointsHead.loss()

  #41 opened a year ago
  0

• 请问test没有label信息，怎么进行裁剪

  #40 opened a year ago
  1

• 请问一下作者大大， OrientedRepPoints_DOTA检测器的输出格式是？

  #39 opened 2 years ago
  0

• result

  #38 opened 2 years ago
  0

• 3090环境配置问题

  #37 opened 2 years ago
  6

• 如何将边界框坐标（x1,y1,x2,y2,x3,y3,x4,y4）转换成(x,y,w,h,theta)的形式。

  #36 opened 2 years ago
  3

• map问题

  #35 opened 2 years ago
  4

• 关于oriented conversion function

  #34 opened 2 years ago
  1

• 模型速度

  #33 opened 2 years ago
  1

• error: command 'C:\\Program Files\\Microsoft Visual Studio\\2022\\Professional\\VC\\Tools\\MSVC\\14.34.31933\\bin\\HostX86\\x64\\link.exe' failed with exit code 1120

  #32 opened 2 years ago
  0

• 训练时报错

  #31 opened 2 years ago
  0

• 请问加入没有标签的图片进行训练会不会有效果？

  #30 opened 2 years ago
  0

• The environment cannot suit for swin-tiny training.

  #29 opened 2 years ago
  2

• RuntimeError: cuda runtime error (700) : an illegal memory access was encountered at mmdet/ops/iou/src/convex_giou_kernel.cu:853

  #28 opened 2 years ago
  2

• 训练HRSC数据集

  #27 opened 2 years ago
  6

• Implementation details on the DIOR-R dataset

  #26 opened 2 years ago
  2

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

  #25 opened 2 years ago
  3

• 预训练模型

  #24 opened 2 years ago
  2

• 预训练模型不匹配

  #23 opened 2 years ago
  1

• 时效性

  #22 opened 2 years ago
  1

• Result on DOTA-v1.5 dataset

  #21 opened 2 years ago
  0

• 训练时在计算giou时报错

  #20 opened 2 years ago
  2

• 你好，这个模型是否能在windows环境训练及验证？

  #19 opened 2 years ago
  5

• There are some problems with parse_pkl_mege_results_for_dota_evaluation.py

  #18 opened 2 years ago
  1

• 报错

  #17 opened 2 years ago
  2

• 转换函数

  #16 opened 2 years ago
  1

• 数据

  #15 opened 3 years ago
  12

• 你好 能分享swintransformer的模型文件吗

  #14 opened 3 years ago
  4

• quality assessment measure

  #13 opened 3 years ago
  5

• 训练报错

  #12 opened 3 years ago
  5

• 版本

  #11 opened 3 years ago
  2

• 测试命令出错。

  #10 opened 3 years ago
  0

• 数据集处理

  #9 opened 3 years ago
  1

• About “Sample Selection for Point Set Representation” and “The differentiability of convex IoU”

  #8 opened 3 years ago
  3

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

  #6 opened 3 years ago
  2

• Train with dota-train-dataset（1024，14384files），the mAP on dota-val-dataset is 70.84

  #4 opened 3 years ago
  5

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

  #3 opened 3 years ago
  6