search

AutoViML / Auto_ViML

Automatically Build Multiple ML Models with a Single Line of Code. Created by Ram Seshadri. Collaborators Welcome. Permission Granted upon Request.
Apache License 2.0
526 stars 102 forks source link

For sklearn above 1.0, the model. estimator didn't work, while 'base_estimator' worked. #36

Closed JulesLiu closed 11 months ago

JulesLiu commented 1 year ago

Well, maybe this is an issue that I should post in scikit-learn project? I installed scikit-learn of 1.1: my modification for the codes: image results and errors: Screenshot 2023-11-24 175528

For 1.3 of scikit-learn, the result is exactly same. btw, my numpy is 1.26, I failed dozens of times to build the wheel for the ~1.19 version suggested in requirments.txt, but I doubt that this may affect sklearn CalibratedClassifierCV object's estimator attribute.

JulesLiu commented 1 year ago

The sklearn document says .estimator is new in version 1.2. Considered the results, the sklearn.version 1.0.2 is the real version, which does not have this new feature, the pip-show is not the real version? image

AutoViML commented 1 year ago

Hi @JulesLiu THANKS for trying out AutoViML 👍 I have made a few minor changes to the code and the requirements.txt and setup.py files. Can you please try one of these two install methods and see if it works:

pip install git+https://github.com/AutoViML/Auto_ViML.git
python3 -m pip install git+https://github.com/AutoViML/Auto_ViML.git

thanks again Auto Vimal