DingXiaoH / RepLKNet-pytorch

Scaling Up Your Kernels to 31x31: Revisiting Large Kernel Design in CNNs (CVPR 2022)
MIT License
868 stars 86 forks source link

Instructions about how to install 19_large_depthwise_conv2d_torch_extension #57

Open liyiersan opened 8 months ago

liyiersan commented 8 months ago

Thanks for the author's nice work. I met some problems when installing large_depthwise_conv2d_torch_extension. However, I finally succeeded in the following ways. Environment: both pytorch1.12 + python3.9+cuda11.3 and pytorch1.10+python3.8+cuda11.3 on Ubuntu 20.04 work fine.

Important notes: 1, Directly use cutclass.zip in the author's repo instead of downloading it from MegEngine. 2, If you want to use the cutlass provided by MegEngine, please download this one 5f664a6 as mentioned in #39 .

The two ways both work for me.

nkzwchen commented 1 month ago

thank you very much !!!