如果是异步完成,网络会遍历失真的图案,在每一个节点n,它会询问n的值应该设置为- 1还是+ 1。
To do this asynchronously, the net traverses the distorted pattern, and at each node, n, it asks whether n's value should be set to -1 or + 1.
当网络访问过每一个图案后,它将一个权重对象的值设置为这个和。
When the net has visited each pattern, it sets the value of a weight object to this sum.
我还设置了一个备用的解锁功能,即在屏幕上滑动一个图案。
I also set up a back-up unlocking option: tracing a pattern on the screen.
应用推荐