frspring

About training a genelized model

2

In the readme I found the line `set batch_size to 1 when you are trining a genelized model, you can enlarge it when fine-tuning.` and since the batch_size is already...

ubuntu cuda 输出全0

1

# 平台(如果交叉编译请再附上交叉编译目标平台): # Platform(Include target platform as well if cross-compiling): Ubuntu20.04 # Github版本: # Github Version: 2.0.1 # 编译方式: # Compiling Method 本地编译 ``` 请在这里粘贴cmake参数或使用的cmake脚本路径以及完整输出 -- Use Threadpool, forbid openmp...

I notice that negative sharpness occured. Is this expected? And what does those negetive value means? It seems that sharpness is not equal to response.