gcg0210
gcg0210
您好，我也有同样的问题，请问您是怎么解决的，谢谢您。
hello! my dataset is little special .some data has no foreground ,and some data has no background. I don't konw if it would work when facing these dataset.thank you .
> First: > ImportError: /home/liangyong/anaconda3/envs/solov2/lib/python3.7/site-packages/mmdet-1.0.0+0b0dfb3-py3.7-linux-x86_64.egg/mmdet/ops/sigmoid_focal_loss/sigmoid_focal_loss_cuda.cpython-37m-x86_64-linux-gnu.so: undefined symbol: _ZN6caffe26detail37_typeMetaDataInstance_preallocated_32E > > My environment: cuda10.1 pytorch1.4 torchvision0.5 > > second: > Traceback (most recent call last): > File "/home/liangyong/SOLO/mmdet/ops/sigmoid_focal_loss/sigmoid_focal_loss.py", line 5, in...
> use pytorch==1.4 , !rm -r build !python setup.py develop > […](#) > On Sun, Dec 20, 2020 at 11:22 AM gcg0210 ***@***.***> wrote: First: ImportError: /home/liangyong/anaconda3/envs/solov2/lib/python3.7/site-packages/mmdet-1.0.0+0b0dfb3-py3.7-linux-x86_64.egg/mmdet/ops/sigmoid_focal_loss/ sigmoid_focal_loss_cuda.cpython-37m-x86_64-linux-gnu.so: undefined symbol:...
hello! my dataset is little special .some data has no foreground ,and some data has no background. I don't konw if it would work when facing these dataset.thank you .
请问，您搞明白这个了吗？手动裁剪和命令裁剪的区别
hi same problem.did you solve this problem?.if so ,how do you solve it ?.thank you .
你们的pytorch是哪个版本的
I met the same problem. Have you solved it? can you provided some suggestion? thank you very much.
> 针对LEVIR_CD数据集，我直接缩放到256*256之后，精度在49左右，是什么原因拿。。 resize 对图像影响较大，举个极端例子，本来变化区域是50个像素点构成的房子，经过你这种操作后，变化区域只有1个像素点，相当于增加了检测难度