By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement . We’ll occasionally send you account related emails.

Hello,

I have an issue when trying to re use my saved regression Model.
I have been able to create use and save my model within the same session.
However after i stop my session and want to load previously save model and make new prediction as per the Tutorial.
I am loading the Model succesfully but after typing the cmd "predict_model" I'm getting an an "Attribute error"

I am getting this error on the computer i have develop the model in and the sample of data i m doing the prediction have the same frame/attributes only the size changed compare the previous prediction done successfully.

Here is my pip and pd version

Thanks for your help!

Hi pycaret.,

Here is my file i give you an extract of what i use.
Dataforexample.xlsx is the data i use to train the model (an extract of it), in my setup "target = Vol."
Prediction.xlsx is the subset of data i was trying to predict the value for prediction that's why there is no "Vol." Column

Let me know if you need anything.

Prediction.xlsx
Dataforexample.xlsx

I tried to reproduce this issue but I cant. See below:

Here is output from my logs.log file which shows my environment and the major dependencies:

Let me know if this helps?

Thanks for your quick reply.

Still not working for me... I check the environment and all my library version seems updated

I m attaching the fill error message log in a "Series Object has no attribute _data.txt"
Series Object has no attribute _data.txt

As well i try load another xgboost model and got this error as well (attached the full message "XGBoostError.txt"
XGBoostError.txt

Is something went wrong when saving my model ?

@pycaret Hi, I did re train with 2.1 version, still facing the same issue . is there any hack to solve this "AttributeError: 'Empty' object has no attribute 'p_transform_target'"

target_transformer = estimator.steps[13][1].p_transform_target # make it dynamic instead of hardcoding no 13