I have a numpy array A and B.
A = [ 1, 2, 5, 9.8, 55, 3]
B = [ 3, 4]
Now, how to remove A[3] & A[4] that is whatever indices array B is having and then put them at the start of array A. So, I want my output to be
A = [9.8, 55, 1, 2, 5, 3]
Note : Both A and B are numpy arrays.
Any help is highly appreciated.
One approach with boolean-indexing would be -
mask = np.in1d(np.arange(A.size),B)
out = np.r_[A[mask], A[~mask]]
Sample run -
In [26]: A = np.array([ 1, 2, 5, 9.8, 55, 3])
In [27]: B = np.array([ 3, 4])
In [28]: mask = np.in1d(np.arange(A.size),B)
In [29]: np.r_[A[mask], A[~mask]]
Out[29]: array([ 9.8, 55. , 1. , 2. , 5. , 3. ])
Another with integer-indexing -
idx = np.setdiff1d(np.arange(A.size),B)
out = np.r_[A[B], A[idx]]
Sample run -
In [36]: idx = np.setdiff1d(np.arange(A.size),B)
In [37]: np.r_[A[B], A[idx]]
Out[37]: array([ 9.8, 55. , 1. , 2. , 5. , 3. ])
Non-numpy answer.
A = [ 1, 2, 5, 9.8, 55, 3]
B = [ 3, 4]
new_arr = [A[i] for i in B if i<len(A)] + [A[i] for i in range(len(A)) if i not in set(B)]
# [9.8, 55, 1, 2, 5, 3]
You may remove the extra check if i<len(A) if you are sure about it.
