When using collapse = True and intervals in lifelines.utils.survival_table_from_events, a 'Cannot convert non-finite values (NA or inf) to integer' ValueError is thrown when any of the intervals exceed the maximum time of the observations. This makes it difficult to use uniform intervals for multiple populations. I'm also seeing this error happen occasionally even when the intervals don't exceed the bounds of the observations.
I'm aiming to use the survival_table_from_events to get a table with one row per year up to a set year, where values are zero beyond the maximum year of the population.
Using lifelines version 0.27.7
from lifelines.datasets import load_waltons from lifelines.utils import survival_table_from_events import numpy as npwaltons = load_waltons() survival_table_from_events(waltons['T'], waltons['E'], collapse = True, intervals = np.arange(10,100).tolist())
When using collapse = True and intervals in lifelines.utils.survival_table_from_events, a 'Cannot convert non-finite values (NA or inf) to integer' ValueError is thrown when any of the intervals exceed the maximum time of the observations. This makes it difficult to use uniform intervals for multiple populations. I'm also seeing this error happen occasionally even when the intervals don't exceed the bounds of the observations. I'm aiming to use the survival_table_from_events to get a table with one row per year up to a set year, where values are zero beyond the maximum year of the population. Using lifelines version 0.27.7
from lifelines.datasets import load_waltonsfrom lifelines.utils import survival_table_from_eventsimport numpy as npwaltons = load_waltons()survival_table_from_events(waltons['T'], waltons['E'], collapse = True, intervals = np.arange(10,100).tolist())