CuPyã®ElementwiseKernelã§æ¥½ã«GPUã®æ©æµã‚’å—ã‘ã‚‹ - QiitaDeleted articles cannot be recovered. Draft of this article would be also deleted. Are you sure you want to delete this article? Powered by DLHacks ã¯ã˜ã‚ã¾ã—ã¦ã€ãƒžãƒ³ãƒœã‚¦ã§ã™ã€‚普段ã¯ä¿¡å·å‡¦ç†ã‚„ç”»åƒè§£æžã‚’ã‚„ã£ã¦ã„ã¾ã™ã€‚ å”çªã§ã™ãŒã€Pythonã‹ã‚‰ãƒ”ンãƒã‚¤ãƒ³ãƒˆã«CUDAã®æ©Ÿèƒ½ã‚’利用ã§ãã‚‹CuPyã®ElementwiseKernelを紹介ã—ã¾ã™ã€‚ CUDAã¨è¨€ã£ã¦ã‚‚æれるã“ã¨ã¯ãªãã€ã€ŒC++ã®è¨˜æ³•ã‚’ãªã‚“ã¨ãªãç†è§£ã—ã¦ã„ã‚‹ã€ãƒ¬ãƒ™ãƒ«ã®äººã§ã‚‚ç°¡å˜ã«åˆ©ç”¨ã§ãã¾ã™ã€‚ ã¯ã˜ã‚ã« Pythonã¯æ›¸ãã‚„ã™ãèªã¿ã‚„ã™ã„ã®ã§ç§ã¯å¥½ãã§ã™ã€‚ ã—ã‹ã—ã€è†¨å¤§ãªä¿¡å·ã‚„大é‡ã®ç”»åƒã‚’Pythonã§æ‰±ã£ã¦ã„ã‚‹ã¨ã€ã©ã†ã—ã¦ã‚‚速度ãŒæ°—ã«ãªã‚Šã ã—ã¾ã™ã€‚ ã‹ã¨ã„ã£ã¦ã€å…¨ã¦ã‚’C++ã‚„CUDA
2013年度 | ＧＰＵコンピューティング - SCIENCE TOKYO OCW
{{#tags}}- {{label}}
{{/tags}}