我有一个3d数组,当以给定的方式打印时,
for i in range(5):
print("======="+str(i)+"=========")
for k in range(5):
dos2[i][k]=atom[i][k][:5]
print(dos2[i][k][:])
看起来像:
=======0=========
0[-4.052, 0.0, 0.0, 0.0, 0.0]
1[-3.995, 0.0, 0.0, 0.0, 0.0]
2[-3.938, -7.918e-34, -2.143e-34, -3.146e-35, -8.847e-36]
3[-3.88, -7.654e-30, -2.276e-30, -3.097e-31, -9.563e-32]
4[-3.823, -3.763e-26, -1.227e-26, -1.541e-27, -5.217e-28]
=======1=========
0[-4.052, 0.0, 0.0, 0.0, 0.0]
1[-3.995, 0.0, 0.0, 0.0, 0.0]
2[-3.938, -1.856e-33, -5.119e-34, -3.842e-35, -1.066e-35]
3[-3.88, -1.795e-29, -5.435e-30, -3.744e-31, -1.141e-31]
4[-3.823, -8.826e-26, -2.931e-26, -1.852e-27, -6.185e-28]
=======2=========
0[-4.052, 0.0, 0.0, 0.0, 0.0]
1[-3.995, 0.0, 0.0, 0.0, 0.0]
2[-3.938, -4.011e-34, -1.081e-34, -1.304e-35, -3.678e-36]
3[-3.88, -3.829e-30, -1.133e-30, -1.274e-31, -3.945e-32]
4[-3.823, -1.867e-26, -6.063e-27, -6.317e-28, -2.144e-28]
=======3=========
0[-4.052, 0.0, 0.0, 0.0, 0.0]
1[-3.995, 0.0, 0.0, 0.0, 0.0]
2[-3.938, -1.204e-32, -3.362e-33, -4.686e-34, -1.34e-34]
3[-3.88, -1.198e-28, -3.676e-29, -4.691e-30, -1.474e-30]
4[-3.823, -6e-25, -2.018e-25, -2.358e-26, -8.124e-27]
=======4=========
0[-4.052, 0.0, 0.0, 0.0, 0.0]
1[-3.995, 0.0, 0.0, 0.0, 0.0]
2[-3.938, -2.455e-33, -6.644e-34, -5.033e-35, -1.412e-35]
3[-3.88, -2.404e-29, -7.148e-30, -5.091e-31, -1.568e-31]
4[-3.823, -1.192e-25, -3.888e-26, -2.575e-27, -8.693e-28]
顶级循环i
表示原子,k
是二进制文件大小,二进制文件的值是[i,k,0]
,并且[i,k,[1:]]
表示为{{ 1}}。
我想要得到的是每个箱子中所有原子的p
之和,即比如说,
p
等等。
我怎么能得到它?
答案 0 :(得分:1)
这是一个猜测,但我认为你想要总结axis=0
:
我喜欢混合各种形状,因此我们更清楚:
In [558]: atom=np.arange(3*4*5).reshape(3,4,5)
In [559]: atom
Out[559]:
array([[[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]],
[[20, 21, 22, 23, 24],
[25, 26, 27, 28, 29],
[30, 31, 32, 33, 34],
[35, 36, 37, 38, 39]],
[[40, 41, 42, 43, 44],
[45, 46, 47, 48, 49],
[50, 51, 52, 53, 54],
[55, 56, 57, 58, 59]]])
In [560]: np.sum(atom, axis=0)
Out[560]:
array([[ 60, 63, 66, 69, 72],
[ 75, 78, 81, 84, 87],
[ 90, 93, 96, 99, 102],
[105, 108, 111, 114, 117]])
我从一个(3,4,5)数组开始,得到一个(4,5),将第一个(你的术语中的原子')压缩为一个值。
或者留下3d,单身第一维:
In [561]: np.sum(atom, axis=0, keepdims=True).shape
Out[561]: (1, 4, 5)
你想要切片的数组,如:
In [563]: np.sum(atom[:,1:,:], axis=0)
Out[563]:
array([[ 75, 78, 81, 84, 87],
[ 90, 93, 96, 99, 102],
[105, 108, 111, 114, 117]])