something went wrong while convert model

harrisonvanderbyl / rwkv-cpp-accelerated

A torchless, c++ rwkv implementation using 8bit quantization, written in cuda/hip/vulkan for maximum compatibility and minimum dependencies

MIT License

303 stars 19 forks source link

Closed innnk closed 1 year ago

innnk commented 1 year ago

hoping for a reply,thx

harrisonvanderbyl commented 1 year ago

something wrong with the compilation step for the conversion, are there errors above this?

innnk commented 1 year ago

something wrong with the compilation step for the conversion, are there errors above this?

I followed the steps above and then occurred the error.

innnk commented 1 year ago

i got it！thx for your reply @harrisonvanderbyl