Extracting Boundaries Of Dense Regions Of 1s In A Huge List Of 1s And 0s

I'm not sure how to word my problem. But here it is... I have a huge list of 1s and 0s [Total length = 53820]. Example of how the list looks like - [0,1,1,1,1,1,1,1,1,0,0,0,1,1,0,

Solution 1:

OK, you need an answer...

First, the imports (we are going to use LineCollections)

import numpy as np ; import matplotlib.pyplotas plt ;                           
from matplotlib.collectionsimportLineCollection

Next, definition of constants

N = 1001; np.random.seed(20190515)                                              

and generation of fake data

x = np.linspace(0,1, 1001)                                                       
prob = np.where(x<0.4, 0.02, np.where(x<0.7, 0.95, 0.02))                        
y = np.where(np.random.rand(1001)<prob, 1, 0)                                    

here we create the line collection, sticks is a N×2×2 array containing the start and end points of our vertical lines

sticks = np.array(list(zip(zip(x, np.zeros(N)), zip(x, y))))                                  
lc = LineCollection(sticks)                                                      

finally, the cumulated sum, here normalized to have the same scale as the vertical lines

cs = (y-0.5).cumsum()                                                            
csmin, csmax = min(cs), max(cs)                                                  
cs = (cs-csmin)/(csmax-csmin) # normalized to 0 ÷ 1                              

We have just to plot our results

f, a = plt.subplots()                                                            
a.add_collection(lc)                                                             
a.plot(x, cs, color='red')                                                       
a.grid()                                                                         
a.autoscale()                                                                    

Here it is the plot

and here a detail of the stop zone.

You can smooth the cs data and use something from scipy.optimize to spot the position of extremes. Should you have a problem in this last step please ask another question.