xiyufeng2
xiyufeng2
why can't I install pytorch1.7.0 directly?compile torch in _https://github.com/torch/distro/_ is too complex!
I have a question about label format during trainging. data_loader.py line250: if(3==len(label_3.shape)): label = label_3[:,:,0] When the input label is RGB image, only the first dimension is taken as GT....
When I run demo.py in Treminal, it's successful. But when running it in IDE(pycharm etc.), shows the error:ImportError: cannot import name '_C' from 'detectron2' (/home/ubuntu/CenterNet2/detectron2/__init__.py) 在终端运行成功，但在IDE运行失败（line11：from detectron2.data.detection_utils import read_image）
there is no trained models in your link(only detaset).Can you provide it?
训练快结束时，neg_inds = neg_inds.unique()，这步做张量的去重时，报错transform: failed to synchronize: cudaErrorAssert: device-side assert triggered。这个张量很奇怪，无法作任何操作。
2021-12-20 17:13:30,133 - mmdet - INFO - Epoch [12][150/4867] lr: 5.000e-05, eta: 0:29:00, time: 0.400, data_time: 0.005, memory: 4601, loss_rpn_cls: nan, loss_rpn_bbox: nan, loss_cls: nan, acc: 11.5102, loss_bbox: nan, loss:...
did it support higher pytorch version like pytorch 1.7.1??
整体架构特别复杂，做了十几个项目的我还是看不懂，主要是层层嵌套太多了。作者能首先告我怎么传参数和数据进去吗？paths_catalog.py 这个看不懂
使用test_net.py文件无法验证，只能出来结果。如何可以测出监测结果的mAP值？感谢
git submodule update --init --recursive，这句命令中克隆子模块总是失败。请问有好的方法吗