xytpai / fcos Goto Github PK

Hello,
could you please upload the net.pkl model file hosted on baidu (https://pan.baidu.com/s/1MjsgE2fu86qmWZpVD7a8Aw) to another file hoster, or maybe as a github artifact? It is pretty cumbersome to download from baidu without registration or additional client software.

Thanks a lot.

Hello, I'm trying to your FCOS implementaitn on my own geospatial dataset (I guess it's similar to the spacenet dataset) to detect buildings. I've followed your instructions for setting up the training and evaluation datasets, but I get no output, (the loss seems to go down but i'm guessing it's converging on predicting nothing).

Just to confirm a few silly things. Your xmin, xmax, ymin, ymax are defined in the standard format right? So left, right, top, bottom.

The typical image i'm training on looks like this:

Do you have any ideas of what parameters I might need to adjust to get good output?

Thanks a lot!
Roger

请问c_y, c_x = h * float(stride), w * float(stride)在映射回原图的方法为什么和fcos论文提到的ws+1/2s和hs+1/2s不一样呢？

你好，请问映射回原图的操作是在代码哪里体现的？我看到之前有人提问提到过c_y, c_x = h * float(stride), w * float(stride)，请问在代码哪个部分？

indexing with dtype torch.uint8 is now deprecated, please use a dtype torch.bool instead.
how to avoid this warning when training？

Loss printed out is NAN

Hello author, I found that in the assign_box.py, the code without the assign_box_cuda is not given, can I ask for the function assign_box(), and the code that can be used without cuda.
Thanks a lot!

I find that the official code get 38.7%AP(80k, about 12 epoch)
I modify your train.json by following official training instructions(12 epoch) and then only get 30%AP

你好，请问如果修改训练时候图像的输入尺寸，1024，200,5这几个数字需要修改吗？比如要修改成256*896大小训练

net = torch.nn.DataParallel(net, device_ids=cfg['device'])
net = net.cuda(cfg['device'][0])
net.train()
这是在run_train.py中准备model时候的操作。
想请问，为什么先分配到所有的gpu上net = torch.nn.DataParallel(net, device_ids=cfg['device'])，然后再单独指定一块net = net.cuda(cfg['device'][0])？不太清楚其中意义。

希望能够获得答复，谢谢！

您好，请问你把centerness去掉后，不会引起误检测变多么？我看原文中这是一个创新点

readme 里的37的精度是 2x 的训练结果？还有4 GPU 的精度是多少，我看跑了官方的源码resnet50 4GPU训练1x 有38+的精度

代码默认的尺寸是1025*1025，而不是readme里写的700？

If using VOC to train ,how can I set the learning rate(2 GPUs)? Looking forward for your reply,thank you very much

我该怎么修改代码得到pascal上的准确率呢

运行run_train.py后，提示

use lib assign_box_cuda
use lib sigmoid_focal_loss_cuda
use lib nms_cuda
Segmentation fault (core dumped)

调试后定位到net = net.cuda(cfg['device'][0])
这是CUDA的版本问题吗？

回归的输出不是偏移量吗，iou_loss的计算是否有问题

你好，我用这个代码，训练coco2017数据集，发现focal loss不能正常收敛，完全偏向了负样本，正样本的分类准确率为0。我使用python3.7和pytorch1.0，batch size为8。请问你有什么经验和建议么