trellixvulnteam/ml_phw1_z6gm

Programming Homework 1

Auto ML for classificaion

FindBestModel

Goal : For each model, we find a combination of k-value, scaling, encoding, score, and parameters.
print best score and best combination for each model.

Parameter

:param df: The Wisconsin Cancer dataset preprocessed
:param target: target feature's name
:param scalers: list of scalers
        None: [StandardScaler(), MinMaxScaler(), MaxAbsScaler(), RobustScaler()]
        if you want to scale other ways, then put the sclaer in list

:param encoders: list of encoders
    None: [OrdinalEncoder(), LabelEncoder() OneHotEncoder()]
    if you want to use only one, put a encoder in list

:param models: list of models
    None: [DecisionTreeClassifier(criterion='gini'), DecisionTreeClassifier(criterion='entropy'),
    LogisticRegression(), SVC()]
    if you want to fit other ways, then put the model in the list