pycaret version checks

• I have checked that this issue has not already been reported here.

• I have confirmed this bug exists on the latest version of pycaret.

• I have confirmed this bug exists on the master branch of pycaret (pip install -U git+https://github.com/pycaret/pycaret.git@master).

Issue Description

My X_train has 2552 samples and my X_HO has 851 samples.

• In the latest version of pycaret, when i'm using just test_data=X_HO in the setup() method, my transformed test set shape has 1702 samples (the origin samples are doubled), however as stated before my X_HO has just 851 samples.
  
  This issue was not observed in the previous version of pycaret, i tested with the old version ('2.3.10') '2.3.10'
  

Reproducible Example

exp_name = setup(
    session_id=1234,
    data=X_train,
    target='target',
    preprocess=False,
    remove_outliers=False,
    test_data=X_HO,
    data_split_shuffle=True,
    ordinal_features={'column': list(X_train.column.unique())})

Expected Behavior

I'm expecting that the results be the same as the previous version of pycaret, as here in the screen.

Actual Results

The problem is already detailed in the description.

Installed Versions