ajie220209

> CAUSAL_CONV1D_FORCE_BUILD=TRUE pip install . An error occurred while executing CAUSAL_CONV1D_FORCE_BUILD=TRUE pip install. ERROR: Could not build wheels for causal_conv1d, which is required to install pyproject.toml-based projects. What can I...

> > 根据我的经验（带有 Nvidia GPU 的 Windows 11），我没有安装 CUDA，所以必须从[这里](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=11&target_type=exe_network)安装它并获取 Pytorch CUDA 版本，这基本上解决了它 > > 好的，谢谢。我正在尝试在 Ubuntu 中创建一个新的环境，并按如下方式安装此软件包，然后它可以成功安装 causal-conv1d 和 mamba-ssm`conda install cudatoolkit==11.8 -c nvidia pip install torch==2.1.1 torchvision==0.16.1 torchaudio==2.1.1 --index-url...

> 嗨，在我们的源代码中，它是通过在验证集中获取性能最佳的权重来测试的。如果你需要在训练的每个周期结束时加载最新的权重进行测试，我们建议你根据需要修改 train.py 代码。 The main problem I'm facing is that in the training phase, a new result is obtained, which contains the best latest.pth weight file. However, in the...