-
Notifications
You must be signed in to change notification settings - Fork 0
nn 만들때 #4
Copy link
Copy link
Open
Labels
enhancementNew feature or requestNew feature or requesthelp wantedExtra attention is neededExtra attention is needed
Metadata
Metadata
Assignees
Labels
enhancementNew feature or requestNew feature or requesthelp wantedExtra attention is neededExtra attention is needed
다음 글은 새로 추가될 수 있는 아이디어를 제시하며, 실현 가능성은 0에 발산함
TWC 같이 大고수들이 많은 경기 플레이 영상을 학습 한다. 플레이 화면이랑 홀드, 넥스트를 입력층으로 만들면 될듯. nn결과는 미노가 "어디에 갈지" 만 정하고, 해당 자리에 미노를 두는 작업은 따로 구현해야 함
문제는 이거임.
내 컴퓨터 스펙이 안타깝게도 Intel Iris Xe Graphics를 GPU로 사용하기 때문에 cuda를 쓸 수가 없음.
그럼 선택지는 2개
-> 근데 이러면 Nvidia GPU 가지는 컴에선 안돌아감
-> 간지남. 하지만 어려움.
근데 이거 작동하면 재밌긴 하겠다