zhenghao977/FCOS-PyTorch

Issues

• 不能使用多卡训练

  #31 opened a year ago
  1

• 与FCOS.Pytorch的对比

  #30 opened a year ago
  0

• 输入图像预处理

  #29 opened 2 years ago
  0

• 测试代码问题

  #28 opened 3 years ago
  2

• 测试mAP

  #27 opened 3 years ago
  1

• 你好，请问detect时候，为啥使用百度盘中的权重voc_77.8.pth，为啥会出现很多框？

  #26 opened 3 years ago
  10

• Error in loading resnet model

  #25 opened 3 years ago
  2

• A confusing line of code, which may have to be modified

  #24 opened 3 years ago
  1

• 学习率调整

  #23 opened 3 years ago
  1

• 执行detect出错。

  #22 opened 3 years ago
  3

• 训练自己数据集，loss不下降但是正确率很高

  #21 opened 3 years ago
  1

• 训练的模型测试只有66.9%的mAP

  #20 opened 3 years ago
  2

• 测试问题

  #19 opened 3 years ago
  1

• pre trained weight file

  #18 opened 3 years ago
  1

• 召回率recall和FPS指标

  #17 opened 3 years ago
  1

• 关于代码中的小细节

  #16 opened 4 years ago
  2

• What's meaning of 'prior' in nn.init.constant_(self.cls_logits.bias,-math.log((1 - prior) / prior)) in head.py

  #15 opened 4 years ago
  2

• 训练VOC数据集出问题

  #14 opened 4 years ago
  29

• 几处不明白的地方

  #13 opened 4 years ago
  4

• 大佬，想问一下检测头部分为什么要对检测分支的输出进行指数处理？（菜鸡提问。。）

  #12 opened 4 years ago
  1

• resnet50的权重能提供一下么

  #11 opened 4 years ago
  4

• About APs,APm,APL

  #10 opened 4 years ago
  7

• 训练结束后模型权重的保存

  #9 opened 4 years ago
  1

• voc dataset

  #8 opened 4 years ago
  1

• 感觉莫名训练COCO的时候卡住了，新手不太懂原因

  #7 opened 4 years ago
  4

• 你好 我改了原文网络 用这个数据集 map达到了0.897 请问算是很大的提升吗？

  #6 opened 4 years ago
  3

• Can you give me some advice

  #5 opened 4 years ago
  1

• 可视化结果很差

  #4 opened 4 years ago
  2

• pretrained weight

  #3 opened 4 years ago
  3

• 0it [00:00, ?it/s]

  #2 opened 4 years ago
  3

• About voc_dataset.py

  #1 opened 4 years ago
  5