ÀÌ·± ÀÏÀ» ÇÕ´Ï´Ù.
NPU Çϵå¿þ¾î¿¡ ƯÈÇÑ Deep Learning °ü·Ã ¾Ë°í¸®ÁòÀ» °³¹ßÇÒ ¿£Áö´Ï¾î¸¦ ã½À´Ï´Ù.
NPUÀÇ Çϵå¿þ¾î ¹× ÄÄÆÄÀÏ·¯ Ư¼ºÀ» °í·ÁÇÏ¿© neural networkÀ» ÃÖÀûÈ ÇÕ´Ï´Ù.
Layerº°, functionº° computationÀ» ºÐ¼®ÇÏ°í ´Ù¾çÇÑ ¾Ë°í¸®ÁòÀ» Àû¿ëÇÏ¿© ¼º´É ¹× Àü·Â¼Ò¸ð¸¦ ÃÖÀûÈ ÇÕ´Ï´Ù.
Quantization, pruningÀ» Æ÷ÇÔÇÑ ´Ù¾çÇÑ ¹æ¹ýÀ» È°¿ëÇϸç, À½¼º, ¿µ»ó, ÀÚ¿¬¾î, Ãßõ µî ´Ù¾çÇÑ ÀÀ¿ë¿¡ Àû¿ëÇÏ°í, ¼º´É »Ó ¾Æ´Ï¶ó ¼ºñ½º°¡ ¿ä±¸Çϴ Ư¼º¿¡ ¸ÂÃç ÃÖÀûÈ °¡´ÉÇÑ ¾Ë°í¸®ÁòÀ» °³¹ßÇÕ´Ï´Ù.
[ÁÖ¿ä ¼öÇà¾÷¹« ¹× ¿ªÇÒ]
- NPU Æ¯È Neural Network ÃÖÀûÈ
- Quantization, pruning, quantization-aware training µî ÃÖÀûÈ ¹æ¹ý È°¿ë
- NPU Æ¯È ÃÖÀû Neural Network »ý¼º ¹× ÇнÀ