AutoViML / featurewiz

Use advanced feature engineering strategies and select best features from your data set with a single line of code. Created by Ram Seshadri. Collaborators welcome.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Conflict Error Among Poetry Package Dependencies: lazytransform, tqdm, featurewiz

ferrotem opened this issue · comments

Error message:
Because no versions of lazytransform match >1.2,<1.3 || >1.3,<1.4 || >1.4,<1.5 || >1.5
and lazytransform (1.2) depends on tqdm (>=4.61.1,<4.62.0), lazytransform (>=1.2,<1.3 || >1.3,<1.4 || >1.4,<1.5 || >1.5) requires tqdm (>=4.61.1,<4.62.0).
And because lazytransform (1.3) depends on tqdm (>=4.61.1,<4.62.0), lazytransform (>=1.2,<1.4 || >1.4,<1.5 || >1.5) requires tqdm (>=4.61.1,<4.62.0).
And because lazytransform (1.4) depends on tqdm (>=4.61.1,<4.62.0)
and lazytransform (1.5) depends on tqdm (>=4.61.1,<4.62.0), lazytransform (>=1.2) requires tqdm (>=4.61.1,<4.62.0).
Because no versions of featurewiz match >0.4.2,<0.5.0
and featurewiz (0.4.2) depends on lazytransform (>=1.2), featurewiz (>=0.4.2,<0.5.0) requires lazytransform (>=1.2).
Thus, featurewiz (>=0.4.2,<0.5.0) requires tqdm (>=4.61.1,<4.62.0).

Steps to Reproduce:
poetry add featurewiz

python = ">=3.8,<=3.10"

Hi @ferrotem 👍
Thanks for testing out featurewiz install using poetry. I have modified the requirements.txt and setup.py files accordingly.
Take a look.
Auto Vimal

Thank you, your update of the tqdm requirements solved problem.
However, there is similar problem with lazytransform: lazytransform (>=1.2) requires scikit-learn (>=0.24.2,<0.25.0).
Can you please remove upper-limit <0.25.0 ?

Hi @ferrotem :
Sorry I cannot remove the upper limit since scikit-learn 1.0 breaks a lot of things in lazytransform.
Auto Vimal

Hi Auto Vimal,
I am sorry, but I found out that featurewiz works well with scikit-learn 1.2.2 when we install via pip. For example, this is message during the installation: Requirement already satisfied: scikit-learn>=0.22.0 in /usr/local/lib/python3.10/dist-packages (from tsfresh) (1.2.2)
Problem occurs when I install featurewiz via poetry. Could you please explain, why there are different requirements for poetry and pip?

I have no idea about Poetry. For now, let's close this.
Auto Vimal