给出两个2D numpy数组和一个1D numpy数组：
F = np.array[[-3764.9303402755891, -3764.9303494098212, -3764.9304257856452, -3764.9306329129422], [-3764.9338022566421, -3764.9338129752682, -3764.9338970928361, -3764.9341184863633]]
T = np.array[[ 10., 30.1, 50.2, 70.3],
[ 10., 30.1, 50.2, 70.3]]
V = np.array[ 226.331804, 228.817957]
我想生成4个文件：
# F_10.0K.dat:
226.331804 -3764.9303402755891
228.817957 -3764.9338022566421
# F_30.1K.dat:
226.331804 -3764.9303494098212
228.817957 -3764.9338129752682
# F_50.2K.dat:
226.331804 -3764.9304257856452
228.817957 -3764.9338970928361
# F_70.3K.dat:
226.331804 -3764.9306329129422
228.817957 -3764.9341184863633
我注意到这两个切片操作：
print ' F_all[:,0] = ', F_all[:, 0]
print ' F_all[:,1] = ', F_all[:, 1]
返回：
F_all[:,0] = [-3764.93034028 -3764.93380226]
F_all[:,1] = [-3764.93034941 -3764.93381298]
这是前两个文件的第2列：F_10.0K.dat和F_30.1K.dat。
所以，我可以循环：
F_all_each_V_at_cte_T = []
for indx in range(0, cols):
aux = F_all[:,indx]
print ' F_all[:, indx] = ', F_all[:,indx]
F_all_each_V_at_cte_T.append(aux)
print 'F_all_each_V_at_cte_T = ', F_all_each_V_at_cte_T
output_array = np.vstack((VOLUME_EACH, F_all_each_V_at_cte_T)).T
np.savetxt('F_vs_V_10.0K.dat', output_array, header="Volume F at 10.0K", fmt="%0.13f")
结果是：
F_all_each_V_at_cte_T = [array([-3764.93034028, -3764.93380226]), array([-3764.93034941, -3764.93381298]), array([-3764.93042579, -3764.93389709]), array([-3764.93063291, -3764.93411849])]
# Volume F at 10.0K
226.3318040000000 -3764.9303402755891 -3764.9303494098212 -3764.9304257856452 -3764.9306329129422
228.8179570000000 -3764.9338022566421 -3764.9338129752682 -3764.9338970928361 -3764.9341184863633
这几乎可以实现解决方案，但是，所有列都已打印出来。如何生成上述文件？
答案 0 :(得分：1)
With your F and V (I added () to make valid Python statements):
In [147]: F = np.array([[-3764.9303402755891, -3764.9303494098212, -3764.930425
...: 7856452, -3764.9306329129422], [-3764.9338022566421, -3764.93381297526
...: 82, -3764.9338970928361, -3764.9341184863633]])
In [148]: V = np.array([ 226.331804, 228.817957])
In [149]: F.shape
Out[149]: (2, 4)
In [150]: V.shape
Out[150]: (2,)
Since your F values differ on the small decimals it's a little tricky to figure out how you want to pair up values. But this appears to be what you want:
In [152]: for i in range(4):
...: print(np.column_stack((V,F[:,i])))
[[ 226.331804 -3764.93034028]
[ 228.817957 -3764.93380226]]
[[ 226.331804 -3764.93034941]
[ 228.817957 -3764.93381298]]
[[ 226.331804 -3764.93042579]
[ 228.817957 -3764.93389709]]
[[ 226.331804 -3764.93063291]
[ 228.817957 -3764.93411849]]
This isn't showing all the decimal values, but that's just a default numpy display issue. I'm using column_stack as a convenient way of concatenating two 1d arrays as columns. I could just as well have used stack or concatenate with the appropriate dimensions adjustment.
Doing the same with savetxt:
In [153]: for i in range(4):
...: np.savetxt('foo%s'%i,(np.column_stack((V,F[:,i]))))
...:
In [154]: cat foo0
2.263318040000000053e+02 -3.764930340275589060e+03
2.288179570000000069e+02 -3.764933802256642139e+03
In [155]: cat foo1
2.263318040000000053e+02 -3.764930349409821247e+03
2.288179570000000069e+02 -3.764933812975268211e+03
This uses the default savetxt formatting. You could use what ever you prefer.
You could use T values to create the file names instead. foo%s%T[0,i]`?
答案 1 :(得分：0)
对于效果更好的矩阵，您可以使用 pprint ，例如以下代码：
from pprint import pprint
pprint('your-matrix')
答案 2 :(得分：0)
基于@hpaulj的回答：
F = np.array([[-3764.9303402755891, -3764.9303494098212, -3764.9304257856452, -3764.9306329129422], [-3764.9338022566421, -3764.9338129752682, -3764.9338970928361, -3764.9341184863633]])
V = np.array([ 226.331804, 228.817957])
T = np.array([[ 10., 30.1, 50.2, 70.3], [ 10., 30.1, 50.2, 70.3]])
for i in range(len(V)):
np.savetxt('foo%s'%i,(np.column_stack((V, F[:,i]))))
我认为这也可以通过以下循环实现：
rows = F_all.shape[0]
cols = F_all.shape[1]
cols_T = T.shape[1]
rows_T = T.shape[0]
for indx, t in zip(range(0, cols), range(0, cols_T) ):
aux_T = T[:,t]
aux_F = F_all[:,indx]
output_array = np.vstack((VOLUME_EACH, aux_F)).T
np.savetxt('F_vs_V_%0.2fK.dat' %aux_T[0], output_array, header="Volume F at %0.2fK" %aux_T[0], fmt="%0.13f")
如此处所示，生成的文件是相同的：
cat foo0：
2.263318040000000053e+02 -3.764930340275589060e+03
2.288179570000000069e+02 -3.764933802256642139e+03
cat foo1：
2.263318040000000053e+02 -3.764930349409821247e+03
2.288179570000000069e+02 -3.764933812975268211e+03
cat F_vs_V_10.00K.dat：
# Volume F at 10.00K
226.3318040000000 -3764.9303402755891
228.8179570000000 -3764.9338022566421
cat F_vs_V_30.10K.dat：
# Volume F at 30.10K
226.3318040000000 -3764.9303494098212
228.8179570000000 -3764.9338129752682