How can we handle few missing time indices in featuretools? Even one missing timestamp in the time index leads to all time based features being null

Many times it so happens that you do not have the time index of a data entry and a lot of other features are present. However, the time index is still present in a lot of the cases and thus the time based features could and should be computed.

I am not finding a way to create such features using featuretools. All the created features are NaN

Example:

from datetime import timedelta
import featuretools as ft
import pandas as pd
import numpy as np
from featuretools import Timedelta

tw = Timedelta(3, unit = "d")
ts = pd.to_datetime("01-01-2020 01:00:00")
time_index = list(pd.date_range(ts, ts + timedelta(hours = 168), periods = 8, inclusive = 'both'))
time_index[4] = np.nan#np.datetime64("NaT")
df1 = pd.DataFrame({'ind': [1], 'time': ts})
df2 = pd.DataFrame({'ind':[1,2,3,4,5,6,7,8], 'id':[1,1,1,1,1,1,1,1], 'time': time_index, 'feat': [np.nan,1,2,4,8,16,48,144]})
es = ft.EntitySet('es')
es.add_dataframe(df1, index = 'ind', dataframe_name = 'base', time_index = 'time')
es.add_dataframe(df2, index = 'ind', dataframe_name = 'data', time_index = 'time')
es.add_relationship('base','ind','data','id')
ct = pd.DataFrame({"instance_id":[1,1,1], "time":[time_index[-1], time_index[-2], time_index[-3]]})
es.add_last_time_indexes()
ft.dfs(entityset = es,  target_dataframe_name = 'base', agg_primitives=['trend'], 
       trans_primitives=[], cutoff_time=ct, cutoff_time_in_index=True, training_window = tw)[0].sort_index()

The output is all Null features. However, if I comment out the one line setting one of the time indices to 0 (time_index[4] = np.nan#np.datetime64("NaT")) then, we can see that the features are getting generated just fine.

How can we handle missing time indices in featuretools?

Output of featuretools.show_info()

Featuretools version: 1.30.0 Featuretools installation directory: /home/nitin/miniconda3/envs/dst/lib/python3.10/site-packages/featuretools

SYSTEM INFO

python: 3.10.13.final.0 python-bits: 64 OS: Linux OS-release: 6.5.0-26-generic machine: x86_64 processor: x86_64 byteorder: little LC_ALL: None LANG: en_IN LOCALE: en_IN.ISO8859-1

INSTALLED VERSIONS

numpy: 1.26.4 pandas: 2.2.1 tqdm: 4.65.0 cloudpickle: 2.2.1 dask: 2023.10.1 psutil: 5.9.8 pip: 23.3.1 setuptools: 68.2.2