Find index of max value in each row of 2D array

Question

I have the following code:

amplitude=[]
for i in range (0,160):
    amplitude.append(i+1)

#print(amplitude)

#split arrays up into a line for each sample
traceno=10                  #number of traces in file
samplesno=16             #number of samples in each trace. This wont change.

amplitude_split=np.array(amplitude, dtype=np.int).reshape((traceno,samplesno))
print(amplitude_split)

#print the maximum value of array along axis=1
max_amp=np.amax(amplitude_split,1)
print(max_amp)

#print the  indices of the maximum values along axis=1
ind_max_amp=np.argmax(amplitude_split, axis=1, out=None)
print(ind_max_amp)

I would like to find the indices for the max value in each row of the array. I am currently only getting ind_max_amp = [15 15 15 15 15 15 15 15 15 15] which I presume is the column of each max value.

What I need is tuples: (0,15) (1,15) etc...

Also, could anyone help me finding the index of 90% and 10% maximum in each row?

nb this is just test code and the max of my real data won't all be in the final column

Answer 1

You can get the maximum value for each row via np.amax using axis=1 and keepdims to keep each maximum value in a separate row, and then use np.argwhere to find the index of that maximum value in each row.

This will give you a list of indices tuples you want:

ind = np.argwhere(amplitude_split==np.amax(amplitude_split,1, keepdims=True))
list(map(tuple, ind))

output:

[(0, 15), (1, 15), (2, 15), (3, 15), (4, 15), (5, 15), (6, 15), (7, 15), (8, 15), (9, 15)]