bug in forecasting splits computation when a datetime cutoff is used and the index is shuffled

[lo-zed]

When using the function get_forecasting_splits with the argument datetime_col and the index of the input data frame is shuffled, the returned indices are wrong.

See minimal example with a proposed solution attached.

bug_report_index.txt

Edit: changed the zip file to a txt file

[oguiza]

Hi @lo-zed, Could you please share your proposed solution in a different way (gist for example)? I never open zip files I have not created myself.

[lo-zed]

Hi, sure. I wanted to leave a plain text file, but github wouldn't let me. I'll just change the extension from .py to .txt