Python: How to get local maxima values from 1D-array or list

Question

I am pretty new in Python and I have a problem which I could not solve with found answers... hopefully someone can help: I need to get a list of all local maxima of a data-set which is imported from a csv-file. Values range from 0 to 0.5 or so.

I just need to get a list of those local maxima of one data row ("Werte", array or "N", list) to do statistics on them.

This is what I have got:

import numpy as np
from numpy import *

N = []   

file = open('C:/Auswertung/PEE/PEE_L_1_O_130702-1.1.csv', 'r') 
Probe = file.readline()         # lese Inhalt zeilenweise in Listen
Header = file.readline()
data = file.readlines()

for row in data:
    columns = row.split(";")       # Trenne Zeilen bei ';' 
    N.append(float(columns[1]))

Werte = np.array([N])

# one try here: only gives me a set of 1s...    
c = (diff(sign(diff(Werte))) < 0).nonzero()[0] + 1 # local max
print(c)

is there anyone who could help me find the right way to do it? Thank you a lot!

If you have [2 4 4 4 0], should 4 be considered as the max?

CT Zhu
– CT Zhu

2013-09-19 16:02:49 +00:00
Commented Sep 19, 2013 at 16:02 — CT Zhu
– CT Zhu, Commented Sep 19, 2013 at 16:02

gg349 · Accepted Answer · 2013-09-19 17:44:38Z

6

I think you are looking for argrelmax, from scipy.signal. It gives you the indices of the relative maxima of a 1d array.

from scipy.signal import argrelmax
t=linspace(-4,40,1000)
y=sin(t)
argrelmax(y)[0]

with result

[126 269 412 554 697 840 982]

to get the values, use

y[argrelmax(y)[0]]

EDIT:

notice that it does not count local maxima at the extreme of your domain.

answered Sep 19, 2013 at 17:44

gg349

22.8k5 gold badges58 silver badges65 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

CT Zhu · Accepted Answer · 2013-09-20 00:22:53Z

2

You are on the right track. The only extra thing you need is to slice the Werke array. but I think, finding the local max can be simplified to:

 werte[1:-1][(diff(werte)[:-1]>0)*(diff(werte)[1:]<0)]

@Jamine was quite right, & instead of * makes it reads better.

edited Sep 20, 2013 at 0:22

answered Sep 19, 2013 at 15:58

CT Zhu

54.6k18 gold badges125 silver badges136 bronze badges

1 Comment

Jaime Over a year ago

Since you are combining boolean arrays, I'd say it is more revealing of your intention to & your two arrays, not to * them.

Collectives™ on Stack Overflow

Python: How to get local maxima values from 1D-array or list

2 Answers 2

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related