Find index within full array of the argmax of an array subset
Question:
Goal: to find the index of the highest value in 1d array from index 25 to [-1]
The way I do it is wrong, I get the index of the highest value of the slice.
If there is no other way than slicing it, how do I then get the correct index of the original array?
Answers:
If you know you’re indexing the array from index 25 on, you could add 25 to the result:
argmax = close[25:].argmax() + 25
But if you’re using an arbitrary slice of your array, another way to do this would be to create a similar array of indices, then slice that the same way:
indices = np.arange(len(close))
# some complicated slicing
sliced = close[2:-7:4]
# apply the same slicing:
sliced_indices = indices[2:-7:4]
# now, get the index in the original array of the argmax from the subset
argmax = sliced_indices[sliced.argmax()]
Goal: to find the index of the highest value in 1d array from index 25 to [-1]
The way I do it is wrong, I get the index of the highest value of the slice.
If there is no other way than slicing it, how do I then get the correct index of the original array?
If you know you’re indexing the array from index 25 on, you could add 25 to the result:
argmax = close[25:].argmax() + 25
But if you’re using an arbitrary slice of your array, another way to do this would be to create a similar array of indices, then slice that the same way:
indices = np.arange(len(close))
# some complicated slicing
sliced = close[2:-7:4]
# apply the same slicing:
sliced_indices = indices[2:-7:4]
# now, get the index in the original array of the argmax from the subset
argmax = sliced_indices[sliced.argmax()]