yys-abu
yys-abu
File "./tools/dist_train.sh", line 3 CONFIG=$1 ^ SyntaxError: invalid syntax
我在训练过程中用自己的数据集出现了loss为NaN的情况,具体是在PointTransformerV3-main\pointcept\pointcept\models\point_transformer_v3\point_transformer_v3m1_base.py backbone网络的point = self.cpe(point)这一步出现问题 再具体就是self.cpe = PointSequential( spconv.SubMConv3d( channels, channels, kernel_size=3, bias=True, indice_key=cpe_indice_key, ), nn.Linear(channels, channels), norm_layer(channels), ) 中的nn.Linear(channels, channels)这一步出现问题 特征在经过全连接层之前是正常的,经过这个全连接层就是NaN了
Hello, I downloaded the ShapeNet55 dataset from the Google Drive link you provided and found that it only contains the shapenet_pc folder and npy files. I converted the npy files...
I download glia_trt-0.4.5-cp39-cp39-win_amd64.whl and glia-0.4.5-cp39-cp39-win_amd64.whl and install them but the error occurs:FileNotFoundError: Could not find module 'D:\Anaconda\envs\yoto\lib\site-packages\glia\bin\GLIA.dll' (or one of its dependencies). Try using the full path with constructor syntax....
Hi, thank you for your great work! I use these following instructions to configure environment but failed: cd thirdparty/DROID-SLAM python setup.py install Because the link to the third-party library is...
D:\Anaconda\envs\SymmCompletion2\Lib\site-packages\torch\utils\cpp_extension.py:1964: UserWarning: TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation. If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST']. warnings.warn( Traceback (most recent call last): File...