lilong-epfl

Thank you very much for your reply and correction. Now the training runs perfectly. But the evaluate get an issue:  which cause by the following code:  I believe...

@AlexeyAB Hi, it works with rtsp stream, but with 3 seconds of delay. Do you have any idea to solve this problem? Thanks in advance.

I trained the same dataset using YOLOv3-model-pruning, it worked well, but somehow it still doesn't work with this algorithm.

@Damon2019 no, I am using my own data.

@Gaondong no, I didn't solve it.

@AlexeyAB Hi, when we use resnet18 as the backbone, why we only use the first 25[0 -24] layers for pre-train, instead of the first 26[0-25] layers?

> It's run using MMLab, specifically [MMSegmentation](https://github.com/open-mmlab/mmsegmentation/tree/main). You can follow the notebook [here](https://github.com/facebookresearch/dinov2/blob/main/notebooks/semantic_segmentation.ipynb) to load the mmsegmentation config file used to run the model. You may have to modify some...

> 2 errors detected in the compilation of "E:/Py_Conda_Project/detrex/detrex/layers/csrc/DCNv3/dcnv3_cuda.cu". dcnv3_cuda.cu ninja: build stopped: subcommand failed. Traceback (most recent call last): File "C:\Users\admin01.conda\envs\detrex-main\lib\site-packages\torch\utils\cpp_extension.py", line 1814, in _run_ninja_build env=env) File "C:\Users\admin01.conda\envs\detrex-main\lib\subprocess.py", line...