Closed CheungZeeCn closed 1 year ago
https://github.com/gaohongkui/GlobalPointer_pytorch/blob/d32f84b423c787d07ad4092ad0a922dc594987fb/models/GlobalPointer.py#L208
谢谢
好像是 dot scaled attn 缘故 ?
是的,参考的 Self-Attention 计算方式
谢谢~
https://github.com/gaohongkui/GlobalPointer_pytorch/blob/d32f84b423c787d07ad4092ad0a922dc594987fb/models/GlobalPointer.py#L208
谢谢