Closed lhoch5813 closed 8 months ago
@lhoch5813 That import is not coming from a featuretools file. Rather, our call in our num_peaks.py
file to from scipy.signal import find_peaks
eventually reaches the import you mention, which is internal to scipy. I'm guessing there may be issues with how your dependencies are installed.
@lhoch5813 I am unable to reproduce this locally using Python3.8 with Featuretools 1.28.0 and scipy 1.10.1 (see below).
Can you try to reinstall all your dependencies in a fresh virtual environment to see if the problem persists?
As @tamargrey mentioned, the particular import that is failing is not inside the Featuretools code base. If you continue to have problems, can you try to run from scipy.signal import find_peaks
to see if that works? If that doesn't work you might either need to try a different version of scipy or report the issue to the scipy repo.
(env) ➜ tmp pip freeze | grep scipy
scipy==1.10.1
(env) ➜ tmp python
Python 3.8.10 (v3.8.10:3d8993a744, May 3 2021, 09:09:08)
[Clang 12.0.5 (clang-1205.0.22.9)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import featuretools as ft
>>> ft.__version__
'1.28.0'
Closing this issue as it doesn't seem to be a Featuretools issue. Please feel free to reopen if the problem persists.
Importing featuretools throws an import error due to trying to import
'float_factorial' from 'scipy._lib._util'.
, which I believe is private. I've included a screenshot of the traceback for the error. Occurring on featuretools 1.28 and scipy 1.10.1.