warp.float16# class warp.float16(x=0)[source]# 16-bit half-precision floating-point scalar type. Parameters: x (int | float)
