read.py

import numpy as np
from scipy.stats import rankdata

def read_data(data, t_min = None, t_max = None):

    path = data + '/' + data + '-'
    # read in the data

    nverts = np.array([int(f.rstrip('\n')) for f in open(path + 'nverts.txt')])
    times = np.array([float(f.rstrip('\n')) for f in open(path + 'times.txt')])
    simplices = np.array([int(f.rstrip('\n')) for f in open(path + 'simplices.txt')])

    times_ex = np.repeat(times, nverts)

    # time filtering

    t_ix = np.repeat(True, len(times))
    if t_min is not None:
        simplices = simplices[times_ex >= t_min]
        nverts = nverts[times >= t_min]
    if t_max is not None:
        simplices = simplices[times_ex <= t_max]
        nverts = nverts[times <= t_max]

    # relabel nodes: consecutive integers from 0 to n-1 

    unique = np.unique(simplices)
    mapper = {unique[i] : i for i in range(len(unique))}
    simplices = np.array([mapper[s] for s in simplices])

    # format as list of lists

    l = np.split(simplices, np.cumsum(nverts))
    C = [list(c) for c in l]
    return(C)