trasfomer 疑问及答案汇总
关于transformer的所有疑问点都在这些文章中回答了,记录一下。
位置编码综述文章:https://kexue.fm/archives/8130
多头数量和效果的讨论:https://blog.csdn.net/abcdefg90876/article/details/105236500?spm=1001.2101.3001.6650.1&utm_medium=distribute.pc_relevant.none-task-blog-2~default~CTRLIST~default-1.no_search_link&depth_1-utm_source=distribute.pc_relevant.none-task-blog-2~default~CTRLIST~default-1.no_search_link
k,q,v的讨论:https://www.zhihu.com/question/325839123