BruceYu-Bit

About eval

1

想请问在评测代码里面，是否对真值进行了二值化，以及在运行评测标准时，会计算人类的标准，大概score是0.8，这个是如何计算的？还有个问题是bsds500上面的原数据是由七个人标注，然后融合取平均，那在测试集上也是拿平均后的真值去评测的吗？

闪退

1

when i open the app,it crashes,how can i do with it?And i use adb to debug,and it shows lack of brief file?What's brief files,i didn't see in your project

Can u give all the training datasets link you used?Including NYUD2,multicue and so on

For Create Edge Maps

3

run code/demoPreproc_gen_png_label.m 运行该命令，无法得到处理的边缘图，我是在windows下运行的，请问有方法解决吗？

How to run on nuscenes test dataset?

1

I download the test set and make test.pkl, but i run the eval file, it seems to see the following issues 

[紧急!!!]训练 MoE 模型建议额外安装 GroupedGEMM失败

16

按照官方文档安装GroupedGEMM，无法build。报错如下 Running setup.py clean for grouped_gemm Running command python setup.py clean /root/anaconda3/envs/xtuner/lib/python3.10/site-packages/setuptools/dist.py:759: SetuptoolsDeprecationWarning: License classifiers are deprecated. !! ******************************************************************************** Please consider removing the following classifiers in favor of a SPDX...