一个好的学习算法应该能够从有限的例子中推广到未见过的数据。
"The act of writing is an act of hope."
You can't trust code that you did not totally create yourself.
当一个人做出彻底的蠢事,那通常是出于最高尚的动机。