NumPy append vs concatenate
NumPy append vs concatenate
NumPy的`append`和`concatenate`有什么区别?我的观察是`concatenate`稍微快一点,而`append`如果未指定轴,则会将数组展平。
观察结果如下:
数组a:
[[1 2]
[3 4]
[5 6]
[5 6]
[1 2]
[3 4]
[5 6]
[5 6]
[1 2]
[3 4]
[5 6]
[5 6]
[5 6]]
数组b:
[[1 2]
[3 4]
[5 6]
[5 6]
[1 2]
[3 4]
[5 6]
[5 6]
[5 6]]
使用`np.concatenate((a, b))`的运行时间为2.05微秒。
使用`np.append(a, b, axis=0)`的运行时间为2.41微秒。
`np.concatenate((a, b))`的输出结果为:
[[1, 2],
[3, 4],
[5, 6],
[5, 6],
[1, 2],
[3, 4],
[5, 6],
[5, 6],
[1, 2],
[3, 4],
[5, 6],
[5, 6],
[5, 6],
[1, 2],
[3, 4],
[5, 6],
[5, 6],
[1, 2],
[3, 4],
[5, 6],
[5, 6],
[5, 6]]
`np.append(a, b, axis=0)`的输出结果为:
[[1, 2],
[3, 4],
[5, 6],
[5, 6],
[1, 2],
[3, 4],
[5, 6],
[5, 6],
[1, 2],
[3, 4],
[5, 6],
[5, 6],
[5, 6],
[1, 2],
[3, 4],
[5, 6],
[5, 6],
[1, 2],
[3, 4],
[5, 6],
[5, 6],
[5, 6]]
`np.append(a, b)`的输出结果为:
[1, 2, 3, 4, 5, 6, 5, 6, 1, 2, 3, 4, 5, 6, 5, 6, 1, 2, 3, 4, 5, 6, 5, 6, 5, 6, 1, 2, 3, 4, 5, 6, 5, 6, 1, 2, 3, 4, 5, 6, 5, 6, 5, 6]