Thank you very much Oliver,
> It did not occurred to me that this can be done so simple with size
>of original array itself!
>Thanks a lot.
>with best regards,
>Sudheer
>From: Olivier Delalleau <[email protected]>
>To: Discussion of Numerical Python <[email protected]>
>Sent: Monday, 27 May 2013 3:22 AM
>Subject: Re: [Numpy-discussion] array manupulation
>
>
>
>Your array doesn't seem strange, it looks like a perfectly normal (11 x 5)
>matrix of dtype float64.
>
>>>> x = np.load('csum.npy')
>>>> np.vstack((np.zeros((1, x.shape[1])), x))
>array([[ 0. , 0. , 0. , 0. , 0.
> ],
> [ 31.82571459, 29.0629995 , 27.74400711, 26.6248159 ,
> 25.73787976],
> [ 59.82231014, 54.27656749, 51.87813602, 50.00937323,
> 48.51771275],
> [ 80.03460893, 73.46862838, 70.55710765, 68.412796 ,
> 66.64323907],
> [ 91.12613011, 85.96434025, 83.34633829, 81.36538282,
> 79.70197141],
> [ 96.11498624, 93.00049572, 91.13864656, 89.61535722,
> 88.27247424],
> [ 98.22403322, 96.55379518, 95.43277035, 94.39550817,
> 93.42804 ],
> [ 99.14200421, 98.27546395, 97.64792507, 97.00438205,
> 96.3689249 ],
> [ 99.55954577, 99.10418687, 98.76971791, 98.39724171,
> 98.00386825],
> [ 99.76081882, 99.51702755, 99.33960611, 99.13057243,
> 98.9007987 ],
> [ 99.8617198 , 99.72882047, 99.63273748, 99.51539561,
> 99.38460995],
> [ 100. , 100. , 100. , 100. , 100.
> ]])
>
>-=- Olivier
>
>
>
>
>2013/5/26 Sudheer Joseph <[email protected]>
>
>Thank you Aronne for the helping hand,
>> I tried the transpose as a check when I
>>could not get it correct other way. I could do it with test arrays, but it
>>appears some thing strange happens when I do the cumsum. So I am attaching
>>here the csum as csum.npy array, where I face problem if your time permits
>>please see what happens with this strange array.!
>>
>>
>>In [1]: csum=np.load('csum.npy') should get the array to you.
>>
>>This array is obtained by doing a
>>csum=np.cumsum(prcnt), which apparently doing some thing which I am not able
>>to visualize.
>>
>>with best regards,
>>Sudheer.
>>
>>>From:Aronne Merrelli <[email protected]>
>>
>>>To:Discussion of Numerical Python <[email protected]>
>>>Sent:Sunday, 26 May 2013 2:13 PM
>>
>>>Subject:Re: [Numpy-discussion] array manupulation
>>>
>>>
>>>
>>>
>>>
>>
>>>On Sun, May 26, 2013 at 4:30 AM, Sudheer Joseph <[email protected]>
>>>wrote:
>>>
>>>Dear Brian,
>>>> I even tried below but no luck!
>>>>In [138]: xx=np.zeros(11)
>>>>In [139]: xx
>>>>Out[139]: array([ 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])
>>>>
>>>>In [147]: xx.shape
>>>>Out[147]: (11,)
>>>>In [140]: xx=np.array(xx)[np.newaxis]
>>>>In [141]: xx.shape
>>>>Out[141]: (1, 11)
>>>>In [142]: xx=xx.T
>>>>In [143]: xx.shape
>>>>Out[143]: (11, 1)
>>>>In [144]: csum.shape
>>>>Out[144]: (11, 5)
>>>>In [145]: np.vstack((xx,csum))
>>>>
>>>>---------------------------------------------------------------------------
>>>>ValueError Traceback (most recent call last)
>>>>/media/SJOITB/SST_VAL/<ipython-input-145-2a0a60f68737> in <module>()
>>>>----> 1 np.vstack((xx,csum))
>>>>
>>>>
>>>>/usr/local/lib/python2.7/dist-packages/numpy-1.7.0-py2.7-linux-x86_64.egg/numpy/core/shape_base.pyc
>>>> in vstack(tup)
>>>> 224
>>>> 225 """
>>>>--> 226 return _nx.concatenate(map(atleast_2d,tup),0)
>>>> 227
>>>> 228 def hstack(tup):
>>>>
>>>>ValueError: all the input array dimensions except for the concatenation
>>>>axis must match exactly
>>>>
>>>>
>>>>
>>>
>>>
>>>You've transposed the arrays, so now you need to stack the other way. So,
>>>you need to use hstack to concatenate arrays with the same column length
>>>(first axis), or vstack to concatenate arrays with the same row length
>>>(second axis). For example:
>>>
>>>
>>>In [110]: xx1 = np.zeros((1,7)); cc1 = np.ones((3,7))
>>>
>>>
>>>In [111]: xx2 = np.zeros((7,1)); cc2 = np.ones((7,3))
>>>
>>>
>>>In [112]: np.vstack((xx1, cc1))
>>>Out[112]:
>>>array([[ 0., 0., 0., 0., 0., 0., 0.],
>>> [ 1., 1., 1., 1., 1., 1., 1.],
>>> [ 1., 1., 1., 1., 1., 1., 1.],
>>> [ 1., 1., 1., 1., 1., 1., 1.]])
>>>
>>>
>>>In [113]: np.hstack((xx2, cc2))
>>>Out[113]:
>>>array([[ 0., 1., 1., 1.],
>>> [ 0., 1., 1., 1.],
>>> [ 0., 1., 1., 1.],
>>> [ 0., 1., 1., 1.],
>>> [ 0., 1., 1., 1.],
>>> [ 0., 1., 1., 1.],
>>> [ 0., 1., 1., 1.]])
>>>
>>>
>>>
>>>
>>>Also, I would highly recommend studying the NumPy for MATLAB users guide:
>>>
>>>
>>>http://www.scipy.org/NumPy_for_Matlab_Users
>>>
>>>
>>>
>>>These issues (any many more) are discussed there.
>>>
>>>
>>>
>>>
>>>Cheers,
>>>Aronne
>>
>>>_______________________________________________
>>>NumPy-Discussion mailing list
>>>[email protected]
>>>http://mail.scipy.org/mailman/listinfo/numpy-discussion
>>>
>>>
>>>
>>_______________________________________________
>>NumPy-Discussion mailing list
>>[email protected]
>>http://mail.scipy.org/mailman/listinfo/numpy-discussion
>>
>>
>
>_______________________________________________
>NumPy-Discussion mailing list
>[email protected]
>http://mail.scipy.org/mailman/listinfo/numpy-discussion
>
>
>_______________________________________________
NumPy-Discussion mailing list
[email protected]
http://mail.scipy.org/mailman/listinfo/numpy-discussion
