I have a numpy matrix with 130 X 13. Say I want to select a specific set of rows meeting a condition and a subset of columns -
trainx[trainy==label,[0,6]]
The above code does not work and throws an error - IndexError: shape mismatch: indexing arrays could not be broadcast together with shapes (43,) (2,).
However if I do it in 2 steps - first subset rows and then columns, it works. Is it something weird or numpy works this way?
temp1 = trainx[trainy==label,:]
temp1 = temp1[:,[0,6]]
you can simply chain the indexing like
trainx[trainy==label][:, [0,6]]
Runable Example
arr = np.random.rand(130,13)
arr[arr[:,0]>0.5][:, [0,6]]
In [154]: x = np.arange(24).reshape(6,4)
In [155]: mask = np.array([1,0,1,0,1,0],bool)
With your two step approach:
In [156]: x[mask] # x[mask, :]
Out[156]:
array([[ 0, 1, 2, 3],
[ 8, 9, 10, 11],
[16, 17, 18, 19]])
In [157]: x[mask][:,[1,3]]
Out[157]:
array([[ 1, 3],
[ 9, 11],
[17, 19]])
Or the two indices could be combined with ix_:
In [158]: np.ix_(mask, [1,3])
Out[158]:
(array([[0],
[2],
[4]]), array([[1, 3]]))
In [159]: x[np.ix_(mask, [1,3])]
Out[159]:
array([[ 1, 3],
[ 9, 11],
[17, 19]])
Note that the first array in Out[158] is np.nonzero(mask)[0][:,None], the nonzero indices in column vector form. That (3,1) indexing array can broadcast with the (2,) column array to select a (3,2) array of elements. Or in your example a (43,2) array.
The boolean mask cannot be turned into a (6,1) array and used to mask x; that would only work if it was turned into a (6,4) mask, matching the shape of x.
So either use the 2 step indexing, or use ix_.
Related
I am trying to figure out a way to get the rows of a 2-d matrix squared.
The behaviour I would like to have is something like this:
in[1] import numpy as np
in[2] a = np.array([[1,2,3],
[4,5,6]])
in[3] some_function(a) # for each row, row.reshape(-1,1); row # row.T
out[1] array([[[ 1, 2, 3],
[ 2, 4, 6],
[ 3, 6, 9]],
[[16, 20, 24],
[20, 25, 30],
[24, 30, 36]]])
I need this to make a softmax derivative for auto diff in a manual implementation of a feed-forward neural network.
The same derivative would look like this for a point:
in[4] def softmax_derivative(x):
in[5] s = x.reshape(-1,1)
in[6] return np.diagflat(s) - np.dot(s,s.T)
Instead of np.diagflat I am using:
in[7] matrix = np.array([[1,2,3],
[4,5,6])
in[8] matrix.shape
out[2] (2,3)
in[9] Id = np.eye(matrix.shape[-1])
in[10] (matrix[...,np.newaxis] * Id).shape
out[3] (2,3,3)
The reason I want a 3-d array of the squared rows is to subtract it from the 3-d array of the diagonal rows which I get in the same way as in the above example.
While I know that I can get the same multiplication result from
in[11] def get_squared_rows(matrix):
in[12] s = matrix.reshape(-1,1)
in[13] return s # s.T
I do not know how to get it to the correct shape in a fast way. Since, yes, the correct 2-d arrays are a part of the matrix on the diagonal, I have to get them together to match the shape of the diagonal 3-d matrix I got. This means I would somehow both have to extract the correct matrices and then turn that into a 3-d array of shape (n_samples,row,row). I do not know how to do that any faster than just a simple loop through all rows of the input matrix.
Use broadcasting:
>>> a[:, None, :] * a[:, :, None]
array([[[ 1, 2, 3],
[ 2, 4, 6],
[ 3, 6, 9]],
[[16, 20, 24],
[20, 25, 30],
[24, 30, 36]]])
I am trying to convert this matlab code to python:
#T2 = (sum((log(X(1:m,:)))'));
Here is my code in python:
T2 = sum(np.log(X[0:int(m),:]).T)
where m = 103 and X is a matrix:
f1 = np.float64(135)
f2 = np.float64(351)
X = np.float64(p[:, int(f1):int(f2)])
and p is dictionary (loaded data)
The problem is python gives me the exact same value with same dimension (216x103) like matlab before applying the sum function on (np.log(X[0:int(m), :]).T). However. after applying the sum function it gives me the correct value but wrong dimension (103x1). The correct dimension is (1x103). I have tried using transpose after getting the sum but it doesnt work. Any suggestions how to get my desired dimension?
A matrix in MATLAB consists of m rows and n columns, but a matrix in NumPy is an array of arrays. Each subarray is a flat vector having 1 dimension equal to the number of its elements n. MATLAB doesn't have flat vectors at all, a row is 1xn matrix, a column is mx1 matrix, and a scalar is 1x1 matrix.
So, back to the question, when you write T2 = sum(np.log(X[0:int(m),:]).T) in Python, it's neither 103x1 nor 1x103, it's a flat 103 vector. If you specifically want a 1x103 matrix like MATLAB, just reshape(1,-1) and you don't have to transpose since you can sum over the second axis.
import numpy as np
X = np.random.rand(216,103)
m = 103
T2 = np.sum(np.log(X[:m]), axis=1).reshape(1,-1)
T2.shape
# (1, 103)
Lets make a demo 2d array:
In [19]: x = np.arange(12).reshape(3,4)
In [20]: x
Out[20]:
array([[ 0, 1, 2, 3],
[ 4, 5, 6, 7],
[ 8, 9, 10, 11]])
And apply the base Python sum function (which isn't the same as numpy's own):
In [21]: sum(x)
Out[21]: array([12, 15, 18, 21])
The result is a (4,) shape array (not 4x1). Print sum(x).shape if you don't believe me.
The numpy.sum function adds all terms if no axis is given:
In [22]: np.sum(x)
Out[22]: 66
or with axis:
In [23]: np.sum(x, axis=0)
Out[23]: array([12, 15, 18, 21])
In [24]: np.sum(x, axis=1)
Out[24]: array([ 6, 22, 38])
The Python sum treats x as a list of arrays, and adds them together
In [25]: list(x)
Out[25]: [array([0, 1, 2, 3]), array([4, 5, 6, 7]), array([ 8, 9, 10, 11])]
In [28]: x[0]+x[1]+x[2]
Out[28]: array([12, 15, 18, 21])
Transpose, without parameter, switch axes. It does not add any dimensions:
In [29]: x.T # (4,3) shape
Out[29]:
array([[ 0, 4, 8],
[ 1, 5, 9],
[ 2, 6, 10],
[ 3, 7, 11]])
In [30]: sum(x).T
Out[30]: array([12, 15, 18, 21]) # still (4,) shape
Octave
>> x=reshape(0:11,4,3)'
x =
0 1 2 3
4 5 6 7
8 9 10 11
>> sum(x)
ans =
12 15 18 21
>> sum(x,1)
ans =
12 15 18 21
>> sum(x,2)
ans =
6
22
38
edit
The np.sum function has a keepdims parmeter:
In [32]: np.sum(x, axis=0, keepdims=True)
Out[32]: array([[12, 15, 18, 21]]) # (1,4) shape
In [33]: np.sum(x, axis=1, keepdims=True)
Out[33]:
array([[ 6], # (3,1) shape
[22],
[38]])
If I reshape the array to 3d, and sum, the result is 2d - unless I keepdims:
In [34]: np.sum(x.reshape(3,2,2), axis=0).shape
Out[34]: (2, 2)
In [36]: np.sum(x.reshape(3,2,2), axis=0,keepdims=True).shape
Out[36]: (1, 2, 2)
MATLAB/Octave on the other hand keeps the dims by default:
sum(reshape(x,3,2,2)) # (1,2,2)
unless I sum on that last, 3rd:
sum(reshape(x,3,2,2),3) # (3,2)
The key is that MATLAB everything is 2d, with the option of additional trailing dimensions, which aren't handled the same way. In numpy every number of dimensions, from 0 on up is handled the same way.
import numpy as np
m = []
k = []
a = np.array([[1,2,3,4,5,6],[50,51,52,40,20,30],[60,71,82,90,45,35]])
for i in range(len(a)):
m.append(a[i, -1:])
for j in range(len(a[i])-1):
n = abs(m[i] - a[i,j])
k.append(n)
k.append(m[i])
print(k)
Expected Output in k:
[5,4,3,2,1,6],[20,21,22,10,10,30],[25,36,47,55,10,35]
which is also a numpy array.
But the output that I am getting is
[array([5]), array([4]), array([3]), array([2]), array([1]), array([6]), array([20]), array([21]), array([22]), array([10]), array([10]), array([30]), array([25]), array([36]), array([47]), array([55]), array([10]), array([35])]
How can I solve this situation?
You want to subtract the last column of each sub array from themselves. Why don't you use a vectorized approach? You can do all the subtractions at once by subtracting the last column from the rest of the items and then column_stack together with unchanged version of the last column. Also note that you need to change the dimension of the last column inorder to be subtractable from the 2D array. For that sake we can use broadcasting.
In [71]: np.column_stack((abs(a[:, :-1] - a[:, None, -1]), a[:,-1]))
Out[71]:
array([[ 5, 4, 3, 2, 1, 6],
[20, 21, 22, 10, 10, 30],
[25, 36, 47, 55, 10, 35]])
I'm trying to use individual 1-dimensional boolean arrays to slice a multi-dimension array. For some reason, this code doesn't work:
>>> a = np.ones((100, 200, 300, 2))
>>> a.shape
(100, 200, 300, 2)
>>> m1 = np.asarray([True]*200)
>>> m2 = np.asarray([True]*300)
>>> m2[-1] = False
>>> a[:,m1,m2,:]
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
IndexError: shape mismatch: indexing arrays could not be broadcast together with shapes (200,) (299,)
>>> m2 = np.asarray([True]*300) # try again with all 300 dimensions True
>>> a[:,m1,m2,:]
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
IndexError: shape mismatch: indexing arrays could not be broadcast together with shapes (200,) (300,)
But this works just fine:
>>> a = np.asarray([[[1, 2], [3, 4], [5, 6]], [[11, 12], [13, 14], [15, 16]]])
>>> a.shape
(2, 3, 2)
>>> m1 = np.asarray([True, False, True])
>>> m2 = np.asarray([True, False])
>>> a[:,m1,m2]
array([[ 1, 5],
[11, 15]])
Any idea of what I might be doing wrong in the first example?
Short answer: The number of True elements in m1 and m2 must match, unless one of them has only one True term.
Also distinguish between 'diagonal' indexing and 'rectangular' indexing. This is about indexing, not slicing. The dimensions with : are just along for the ride.
Initial ideas
I can get your first case working with:
In [137]: a=np.ones((100,200,300,2))
In [138]: m1=np.ones((200,),bool)
In [139]: m2=np.ones((300,),bool)
In [140]: m2[-1]=False
In [141]: I,J=np.ix_(m1,m2)
In [142]: a[:,I,J,:].shape
Out[142]: (100, 200, 299, 2)
np.ix_ turns the 2 boolean arrays into broadcastable index arrays
In [143]: I.shape
Out[143]: (200, 1)
In [144]: J.shape
Out[144]: (1, 299)
Note that this picks 200 'rows' in one dimension, and 299 in the other.
I'm not sure why this kind of reworking of the arrays is needed in this case, but not in the 2nd
In [154]: b=np.arange(2*3*2).reshape((2,3,2))
In [155]: n1=np.array([True,False,True])
In [156]: n2=np.array([True,False])
In [157]: b[:,n1,n2]
Out[157]:
array([[ 0, 4], # shape (2,2)
[ 6, 10]])
Taking the same ix_ strategy produces the same values but a different shape:
In [164]: b[np.ix_(np.arange(b.shape[0]),n1,n2)]
# or I,J=np.ix_(n1,n2);b[:,I,J]
Out[164]:
array([[[ 0],
[ 4]],
[[ 6],
[10]]])
In [165]: _.shape
Out[165]: (2, 2, 1)
Both cases use all rows of the 1st dimension. The ix one picks 2 'rows' of the 2nd dim, and 1 column of the last, resulting the (2,2,1) shape. The other picks b[:,0,0] and b[0,2,0] terms, resulting (2,2) shape.
(see my addenda as to why both are simply broadcasting).
These are all cases of advanced indexing, with boolean and numeric indexes. One can study the docs, or one can play around. Sometimes it's more fun to do the later. :)
(I knew that ix_ was good for adding the necessary np.newaxis to arrays so can be broadcast together, but didn't realize that worked with boolean arrays as well - it uses np.nonzero() to convert boolean to indices.)
Resolution
Underlying this is, I think, a confusion over 2 modes of indexing. which might called 'diagonal' and 'rectangular' (or element-by-element selection versus block selection). To illustrate look at a small 2d array
In [73]: M=np.arange(6).reshape(2,3)
In [74]: M
Out[74]:
array([[0, 1, 2],
[3, 4, 5]])
and 2 simple numeric indexes
In [75]: m1=np.arange(2); m2=np.arange(2)
They can be used 2 ways:
In [76]: M[m1,m2]
Out[76]: array([0, 4])
and
In [77]: M[m1[:,None],m2]
Out[77]:
array([[0, 1],
[3, 4]])
The 1st picks 2 points, the M[0,0] and M[1,1]. This kind of indexing lets us pick out the diagonals of an array.
The 2nd picks 2 rows and from that 2 columns. This is the kind of indexing the np.ix_ produces. The 1st picks 2 points, the M[0,0] and M[1,1]. This a 'rectangular' form of indexing.
Change m2 to 3 values:
In [78]: m2=np.arange(3)
In [79]: M[m1[:,None],m2] # returns a 2x3
Out[79]:
array([[0, 1, 2],
[3, 4, 5]])
In [80]: M[m1,m2] # produces an error
...
ValueError: shape mismatch: objects cannot be broadcast to a single shape
But if m2 has just one element, we don't get the broadcast error - because the size 1 dimension can be expanded during broadcasting:
In [81]: m2=np.arange(1)
In [82]: M[m1,m2]
Out[82]: array([0, 3])
Now change the index arrays to boolean, each matching the length of the respective dimensions, 2 and 3.
In [91]: m1=np.ones(2,bool); m2=np.ones(3,bool)
In [92]: M[m1,m2]
...
ValueError: shape mismatch: objects cannot be broadcast to a single shape
In [93]: m2[2]=False # m1 and m2 each have 2 True elements
In [94]: M[m1,m2]
Out[94]: array([0, 4])
In [95]: m2[0]=False # m2 has 1 True element
In [96]: M[m1,m2]
Out[96]: array([1, 4])
With 2 and 3 True terms we get an error, but with 2 and 2 or 2 and 1 it runs - just as though we'd used the indices of the True elements: np.nonzero(m2).
To apply this to your examples. In the first, m1 and m2 have 200 and 299 True elements. a[:,m1,m2,:] fails because of a mismatch in the number of True terms.
In the 2nd, they have 2 and 1 True terms, with nonzero indices of [0,2] and [0], which can be broadcast to [0,0]. So it runs.
http://docs.scipy.org/doc/numpy-1.10.0/reference/arrays.indexing.html
explains boolean array indexing in terms of nonzero and ix_.
Combining multiple Boolean indexing arrays or a Boolean with an integer indexing array can best be understood with the obj.nonzero() analogy. The function ix_ also supports boolean arrays and will work without any surprises.
Addenda
On further thought the distinction between 'diagonal' and 'block/rectangular' indexing might be more my mental construct that numpys. Underlying both is the concept of broadcasting.
Take the n1 and n2 booleans, and get their nonzero equivalents:
In [107]: n1
Out[107]: array([ True, False, True], dtype=bool)
In [108]: np.nonzero(n1)
Out[108]: (array([0, 2], dtype=int32),)
In [109]: n2
Out[109]: array([ True, False], dtype=bool)
In [110]: np.nonzero(n2)
Out[110]: (array([0], dtype=int32),)
Now try broadcasting in the 'diagonal' and 'rectangular' modes:
In [105]: np.broadcast_arrays(np.array([0,2]),np.array([0]))
Out[105]: [array([0, 2]),
array([0, 0])]
In [106]: np.broadcast_arrays(np.array([0,2])[:,None],np.array([0]))
Out[106]:
[array([[0],
[2]]),
array([[0],
[0]])]
One produces (2,) arrays, the other (2,1).
This might be a simple workaround:
a[:,m1,:,:][:,:,m2,:]
In my dataset I've close to 200 rows but for a minimal working e.g., let's assume the following array:
arr = np.array([[1,2,3,4], [5,6,7,8],
[9,10,11,12], [13,14,15,16],
[17,18,19,20], [21,22,23,24]])
I can take a random sampling of 3 of the rows as follows:
indexes = np.random.choice(np.arange(arr.shape[0]), int(arr.shape[0]/2), replace=False)
Using these indexes, I can select my test cases as follows:
testing = arr[indexes]
I want to delete the rows at these indexes and I can use the remaining elements for my training set.
From the post here, it seems that training = np.delete(arr, indexes) ought to do it. But I get 1d array instead.
I also tried the suggestion here using training = arr[indexes.astype(np.bool)] but it did not give a clean separation. I get element [5,6,7,8] in both the training and testing sets.
training = arr[indexes.astype(np.bool)]
testing
Out[101]:
array([[13, 14, 15, 16],
[ 5, 6, 7, 8],
[17, 18, 19, 20]])
training
Out[102]:
array([[ 1, 2, 3, 4],
[ 5, 6, 7, 8],
[ 9, 10, 11, 12]])
Any idea what I am doing wrong? Thanks.
To delete indexed rows from numpy array:
arr = np.delete(arr, indexes, axis=0)
One approach would be to get the remaining row indices with np.setdiff1d and then use those row indices to get the desired output -
out = arr[np.setdiff1d(np.arange(arr.shape[0]), indexes)]
Or use np.in1d to leverage boolean indexing -
out = arr[~np.in1d(np.arange(arr.shape[0]), indexes)]