[Numpy-discussion] reshape

Nils Wagner nwagner@iam.uni-stuttgart...
Thu Nov 6 09:13:59 CST 2008

```Hi all,

What can be done if
the new shape is not compatible with the original shape ?
The number of columns is fixed and should be 8.

One could split the original array C

>>> C
array([[  0.00000000e+00,   1.00000000e-01],
[  4.15000000e+01,   1.00000000e-01],
[  4.16000000e+01,   1.00000000e-01],
[  6.07500000e+01,   1.00000000e-01],
[  6.08500000e+01,   1.00000000e-01],
[  8.22550000e+01,   1.00000000e-01],
[  8.23550000e+01,   1.00000000e-01],
[  9.42550000e+01,   1.00000000e-01],
[  9.43550000e+01,   1.00000000e-01],
[  9.99100000e+01,   1.00000000e-01],
[  1.00010000e+02,   1.00000000e-01],
[  1.07660000e+02,   1.00000000e-01],
[  1.07760000e+02,   1.00000000e-01],
[  1.28000000e+02,   1.00000000e-01],
[  1.28100000e+02,   1.00000000e-01]])
>>> shape(C)
(15, 2)

>>> A = C[:8,:]
>>> A
array([[  0.   ,   0.1  ],
[ 41.5  ,   0.1  ],
[ 41.6  ,   0.1  ],
[ 60.75 ,   0.1  ],
[ 60.85 ,   0.1  ],
[ 82.255,   0.1  ],
[ 82.355,   0.1  ],
[ 94.255,   0.1  ]])
>>> A = reshape(A,(2,8))
>>> A
array([[  0.   ,   0.1  ,  41.5  ,   0.1  ,  41.6  ,   0.1
,  60.75 ,
0.1  ],
[ 60.85 ,   0.1  ,  82.255,   0.1  ,  82.355,
0.1  ,  94.255,
0.1  ]])
>>> B = C[8:,:]
>>> B
array([[  9.43550000e+01,   1.00000000e-01],
[  9.99100000e+01,   1.00000000e-01],
[  1.00010000e+02,   1.00000000e-01],
[  1.07660000e+02,   1.00000000e-01],
[  1.07760000e+02,   1.00000000e-01],
[  1.28000000e+02,   1.00000000e-01],
[  1.28100000e+02,   1.00000000e-01]])
>>> B = reshape(B,(2,7))
>>> B
array([[  9.43550000e+01,   1.00000000e-01,
9.99100000e+01,
1.00000000e-01,   1.00010000e+02,
1.00000000e-01,
1.07660000e+02],
[  1.00000000e-01,   1.07760000e+02,
1.00000000e-01,
1.28000000e+02,   1.00000000e-01,
1.28100000e+02,
1.00000000e-01]])

Nils
```