numpy 函数diff 用法初探

encoding=utf-8

# 导入numpy 模块
import numpy as np

随机生成一个4行，6列的数组

a = np.random.randn(4, 6)
print a
print ‘\n’
print np.diff(a)
print ‘test’
打印结果如下
[[-1.71410282 0.12532478 0.45723509 0.54354484 0.25882893 2.76672223]
[ 1.85177107 0.06898675 0.96400795 1.95304515 -0.75269802 0.71917141]
[ 1.3604721 0.54327318 -1.48450286 0.90393593 0.6427886 1.26583073]
[ 0.10683071 0.80828143 0.68135935 0.08793795 1.19330235 -1.23413727]]

[[ 1.8394276 0.33191031 0.08630975 -0.28471592 2.5078933 ]
[-1.78278432 0.8950212 0.9890372 -2.70574317 1.47186943]
[-0.81719892 -2.02777605 2.3884388 -0.26114733 0.62304213]
[ 0.70145072 -0.12692208 -0.5934214 1.1053644 -2.42743962]]
test

总结下就是每行的后一个值减去前一个，然后放到一个新的数组里面