Retrieving Max Lifespan from Rips

[aflytle]

I'm relatively new to TDA, so forgive some uninformed sounding wording.

In the Ripser.py implementation, I see that we are using the C++ structs for the arrays of points.
Is there any way to (still in Python) retrieve the maximum lifespan of some points in the persistence diagram/barcodes? I.E., is there a way to find the maximum lifespan of a point in a point cloud, when running the rips complex persistence homology on it?

I need to find the two longest lifespans in the dataset.

Respectfully,
Aidan Lytle

[ctralie]

Hello,
Thank you for your patience on our response. Here's an example of how one might do this:

import numpy as np
from ripser import ripser
X = np.random.randn(100, 20) # Make a random point cloud with 100 points in 20 dimensions
dgms = ripser(X)['dgms'] # Pull out the diagrams (by default up to H1)
H1 = dgms[1] # Look at H1
H1 = H1[np.argsort(H1[:, 0] - H1[:, 1]), :] # Sort the points in descending order of persistence
lifespans = H1[:, 1] - H1[:, 0] # Now we have the lifespans in decreasing order