If you got enough GPU memory, ain't no tensor large enough. Anyway, to answer your question, you can do convolution operation on 500 3, ...
確定! 回上一頁