Conditional indexing with Numpy ndarray

Question

I have a Numpy ndarray matrix of float values and I need to select spesific rows where certain columns have values satisfying certain criteria. For example lets say I have the following numpy matrix:

matrix = np.ndarray([4, 5])
matrix[0,:] = range(1,6)
matrix[1,:] = range(6,11)
matrix[2,:] = range(11,16)
matrix[3,:] = range(16,21)

Lets say I want to select rows from the matrix where the first column's value is between 1 and 6 and the value of second column is between 2-7.

How can I get the row-indexes of the matrix where these conditions are satisfied? What about if I want to delete the rows that satisfy the conditional criterion?

Answer 1

For a numpy based solution, you can use numpy.where and then get the row indexes from it and then use it for indexing you matrix. Example -

matrix[np.where((1 <= matrix[:,0]) & (matrix[:,0] <= 6)
       & (2 <= matrix[:,1]) & (matrix[:,1] <= 7))]

Demo -

In [169]: matrix
Out[169]:
array([[  1.,   2.,   3.,   4.,   5.],
       [  6.,   7.,   8.,   9.,  10.],
       [ 11.,  12.,  13.,  14.,  15.],
       [ 16.,  17.,  18.,  19.,  20.]])

In [170]: matrix[np.where((1 <= matrix[:,0]) & (matrix[:,0] <= 6)
   .....:        & (2 <= matrix[:,1]) & (matrix[:,1] <= 7))]
Out[170]:
array([[  1.,   2.,   3.,   4.,   5.],
       [  6.,   7.,   8.,   9.,  10.]])

---

Another method , as indicated in the comments would be to use boolean masks, Example -

mask = ((1 <= matrix[:,0]) & (matrix[:,0] <= 6)
           & (2 <= matrix[:,1]) & (matrix[:,1] <= 7))

matrix[mask,:]

Demo -

In [41]: matrix
Out[41]:
array([[  1.,   2.,   3.,   4.,   5.],
       [  6.,   7.,   8.,   9.,  10.],
       [ 11.,  12.,  13.,  14.,  15.],
       [ 16.,  17.,  18.,  19.,  20.]])

In [42]: mask = ((1 <= matrix[:,0]) & (matrix[:,0] <= 6)
   ....:            & (2 <= matrix[:,1]) & (matrix[:,1] <= 7))

In [43]:

In [43]: matrix[mask,:]
Out[43]:
array([[  1.,   2.,   3.,   4.,   5.],
       [  6.,   7.,   8.,   9.,  10.]])