py2ls 0.2.4.15__py3-none-any.whl → 0.2.4.16__py3-none-any.whl

py2ls/ml2ls copy.py

@@ -0,0 +1,2906 @@
+from sklearn.ensemble import (
+    RandomForestClassifier,
+    GradientBoostingClassifier,
+    AdaBoostClassifier,
+    BaggingClassifier,
+)
+from sklearn.svm import SVC, SVR
+from sklearn.calibration import CalibratedClassifierCV
+from sklearn.model_selection import GridSearchCV, StratifiedKFold
+from sklearn.linear_model import (
+    LassoCV,
+    LogisticRegression,
+    LinearRegression,
+    Lasso,
+    Ridge,
+    RidgeClassifierCV,
+    ElasticNet,
+)
+from sklearn.feature_selection import RFE
+from sklearn.naive_bayes import GaussianNB
+from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
+import xgboost as xgb  # Make sure you have xgboost installed
+
+from sklearn.model_selection import train_test_split, cross_val_score
+from sklearn.metrics import (
+    accuracy_score,
+    precision_score,
+    recall_score,
+    f1_score,
+    roc_auc_score,
+    confusion_matrix,
+    matthews_corrcoef,
+    roc_curve,
+    auc,
+    balanced_accuracy_score,
+    precision_recall_curve,
+    average_precision_score,
+)
+from imblearn.over_sampling import SMOTE
+from sklearn.pipeline import Pipeline
+from collections import defaultdict
+from sklearn.preprocessing import StandardScaler, OneHotEncoder
+from typing import Dict, Any, Optional, List, Union
+import numpy as np
+import pandas as pd
+from . import ips
+from . import plot
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+plt.style.use(str(ips.get_cwd()) + "/data/styles/stylelib/paper.mplstyle")
+import logging
+import warnings
+
+logging.basicConfig(
+    level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s"
+)
+logger = logging.getLogger()
+
+# Ignore specific warnings (UserWarning in this case)
+warnings.filterwarnings("ignore", category=UserWarning)
+from sklearn.tree import DecisionTreeClassifier
+from sklearn.neighbors import KNeighborsClassifier
+
+
+def features_knn(
+    x_train: pd.DataFrame, y_train: pd.Series, knn_params: dict
+) -> pd.DataFrame:
+    """
+    A distance-based classifier that assigns labels based on the majority label of nearest neighbors.
+    when to use:
+        Effective for small to medium datasets with a low number of features.
+        It does not directly provide feature importances but can be assessed through feature permutation or similar methods.
+    Recommended Use: Effective for datasets with low feature dimensionality and well-separated clusters.
+
+    Fits KNeighborsClassifier and approximates feature influence using permutation importance.
+    """
+    knn = KNeighborsClassifier(**knn_params)
+    knn.fit(x_train, y_train)
+    importances = permutation_importance(
+        knn, x_train, y_train, n_repeats=30, random_state=1, scoring="accuracy"
+    )
+    return pd.DataFrame(
+        {"feature": x_train.columns, "importance": importances.importances_mean}
+    ).sort_values(by="importance", ascending=False)
+
+
+#! 1. Linear and Regularized Regression Methods
+# 1.1 Lasso
+def features_lasso(
+    x_train: pd.DataFrame, y_train: pd.Series, lasso_params: dict
+) -> np.ndarray:
+    """
+    Lasso (Least Absolute Shrinkage and Selection Operator):
+    A regularized linear regression method that uses L1 penalty to shrink coefficients, effectively
+    performing feature selection by zeroing out less important ones.
+    """
+    lasso = LassoCV(**lasso_params)
+    lasso.fit(x_train, y_train)
+    # Get non-zero coefficients and their corresponding features
+    coefficients = lasso.coef_
+    importance_df = pd.DataFrame(
+        {"feature": x_train.columns, "importance": np.abs(coefficients)}
+    )
+    return importance_df[importance_df["importance"] > 0].sort_values(
+        by="importance", ascending=False
+    )
+
+
+# 1.2 Ridge regression
+def features_ridge(
+    x_train: pd.DataFrame, y_train: pd.Series, ridge_params: dict
+) -> np.ndarray:
+    """
+    Ridge Regression: A linear regression technique that applies L2 regularization, reducing coefficient
+    magnitudes to avoid overfitting, especially with multicollinearity among features.
+    """
+    from sklearn.linear_model import RidgeCV
+
+    ridge = RidgeCV(**ridge_params)
+    ridge.fit(x_train, y_train)
+
+    # Get the coefficients
+    coefficients = ridge.coef_
+
+    # Create a DataFrame to hold feature importance
+    importance_df = pd.DataFrame(
+        {"feature": x_train.columns, "importance": np.abs(coefficients)}
+    )
+    return importance_df[importance_df["importance"] > 0].sort_values(
+        by="importance", ascending=False
+    )
+
+
+# 1.3 Elastic Net(Enet)
+def features_enet(
+    x_train: pd.DataFrame, y_train: pd.Series, enet_params: dict
+) -> np.ndarray:
+    """
+    Elastic Net (Enet): Combines L1 and L2 penalties (lasso and ridge) in a linear model, beneficial
+    when features are highly correlated or for datasets with more features than samples.
+    """
+    from sklearn.linear_model import ElasticNetCV
+
+    enet = ElasticNetCV(**enet_params)
+    enet.fit(x_train, y_train)
+    # Get the coefficients
+    coefficients = enet.coef_
+    # Create a DataFrame to hold feature importance
+    importance_df = pd.DataFrame(
+        {"feature": x_train.columns, "importance": np.abs(coefficients)}
+    )
+    return importance_df[importance_df["importance"] > 0].sort_values(
+        by="importance", ascending=False
+    )
+
+
+# 1.4 Partial Least Squares Regression for Generalized Linear Models (plsRglm): Combines regression and
+# feature reduction, useful for high-dimensional data with correlated features, such as genomics.
+
+#! 2.Generalized Linear Models and Extensions
+# 2.1
+
+
+#!3.Tree-Based and Ensemble Methods
+# 3.1 Random Forest(RF)
+def features_rf(
+    x_train: pd.DataFrame, y_train: pd.Series, rf_params: dict
+) -> np.ndarray:
+    """
+    An ensemble of decision trees that combines predictions from multiple trees for classification or
+    regression, effective with high-dimensional, complex datasets.
+    when to use:
+        Handles high-dimensional data well.
+        Robust to overfitting due to averaging of multiple trees.
+        Provides feature importance, which can help in understanding the influence of different genes.
+    Fit Random Forest and return sorted feature importances.
+    Recommended Use: Great for classification problems, especially when you have many features (genes).
+    """
+    rf = RandomForestClassifier(**rf_params)
+    rf.fit(x_train, y_train)
+    return pd.DataFrame(
+        {"feature": x_train.columns, "importance": rf.featuress_}
+    ).sort_values(by="importance", ascending=False)
+
+
+# 3.2 Gradient Boosting Trees
+def features_gradient_boosting(
+    x_train: pd.DataFrame, y_train: pd.Series, gb_params: dict
+) -> pd.DataFrame:
+    """
+    An ensemble of decision trees that combines predictions from multiple trees for classification or regression, effective with
+    high-dimensional, complex datasets.
+    Gradient Boosting
+    Strengths:
+        High predictive accuracy and works well for both classification and regression.
+        Can handle a mixture of numerical and categorical features.
+    Recommended Use:
+        Effective for complex relationships and when you need a powerful predictive model.
+    Fit Gradient Boosting classifier and return sorted feature importances.
+    Recommended Use: Effective for complex datasets with many features (genes).
+    """
+    gb = GradientBoostingClassifier(**gb_params)
+    gb.fit(x_train, y_train)
+    return pd.DataFrame(
+        {"feature": x_train.columns, "importance": gb.feature_importances_}
+    ).sort_values(by="importance", ascending=False)
+
+
+# 3.3 XGBoost
+def features_xgb(
+    x_train: pd.DataFrame, y_train: pd.Series, xgb_params: dict
+) -> pd.DataFrame:
+    """
+    XGBoost: An advanced gradient boosting technique, faster and more efficient than GBM, with excellent predictive performance on structured data.
+    """
+    import xgboost as xgb
+
+    xgb_model = xgb.XGBClassifier(**xgb_params)
+    xgb_model.fit(x_train, y_train)
+    return pd.DataFrame(
+        {"feature": x_train.columns, "importance": xgb_model.feature_importances_}
+    ).sort_values(by="importance", ascending=False)
+
+
+# 3.4.decision tree
+def features_decision_tree(
+    x_train: pd.DataFrame, y_train: pd.Series, dt_params: dict
+) -> pd.DataFrame:
+    """
+    A single decision tree classifier effective for identifying key decision boundaries in data.
+    when to use:
+        Good for capturing non-linear patterns.
+        Provides feature importance scores for each feature, though it may overfit on small datasets.
+        Efficient for low to medium-sized datasets, where interpretability of decisions is key.
+    Recommended Use: Useful for interpretable feature importance analysis in smaller or balanced datasets.
+
+    Fits DecisionTreeClassifier and returns sorted feature importances.
+    """
+    dt = DecisionTreeClassifier(**dt_params)
+    dt.fit(x_train, y_train)
+    return pd.DataFrame(
+        {"feature": x_train.columns, "importance": dt.feature_importances_}
+    ).sort_values(by="importance", ascending=False)
+
+
+# 3.5 bagging
+def features_bagging(
+    x_train: pd.DataFrame, y_train: pd.Series, bagging_params: dict
+) -> pd.DataFrame:
+    """
+    A bagging ensemble of models, often used with weak learners like decision trees, to reduce variance.
+    when to use:
+        Helps reduce overfitting, especially on high-variance models.
+        Effective when the dataset has numerous features and may benefit from ensemble stability.
+    Recommended Use: Beneficial for high-dimensional or noisy datasets needing ensemble stability.
+
+    Fits BaggingClassifier and returns averaged feature importances from underlying estimators if available.
+    """
+    bagging = BaggingClassifier(**bagging_params)
+    bagging.fit(x_train, y_train)
+
+    # Calculate feature importance by averaging importances across estimators, if feature_importances_ is available.
+    if hasattr(bagging.estimators_[0], "feature_importances_"):
+        importances = np.mean(
+            [estimator.feature_importances_ for estimator in bagging.estimators_],
+            axis=0,
+        )
+        return pd.DataFrame(
+            {"feature": x_train.columns, "importance": importances}
+        ).sort_values(by="importance", ascending=False)
+    else:
+        # If the base estimator does not support feature importances, fallback to permutation importance.
+        importances = permutation_importance(
+            bagging, x_train, y_train, n_repeats=30, random_state=1, scoring="accuracy"
+        )
+        return pd.DataFrame(
+            {"feature": x_train.columns, "importance": importances.importances_mean}
+        ).sort_values(by="importance", ascending=False)
+
+
+#! 4.Support Vector Machines
+def features_svm(
+    x_train: pd.DataFrame, y_train: pd.Series, rfe_params: dict
+) -> np.ndarray:
+    """
+    Suitable for classification tasks where the number of features is much larger than the number of samples.
+        1. Effective in high-dimensional spaces and with clear margin of separation.
+        2. Works well for both linear and non-linear classification (using kernel functions).
+    Select features using RFE with SVM.When combined with SVM, RFE selects features that are most critical for the decision boundary,
+        helping reduce the dataset to a more manageable size without losing much predictive power.
+    SVM (Support Vector Machines),supports various kernels (linear, rbf, poly, and sigmoid), is good at handling high-dimensional
+        data and finding an optimal decision boundary between classes, especially when using the right kernel.
+    kernel: ["linear", "rbf", "poly", "sigmoid"]
+        'linear': simplest kernel that attempts to separate data by drawing a straight line (or hyperplane) between classes. It is effective
+            when the data is linearly separable, meaning the classes can be well divided by a straight boundary.
+                Advantages:
+                    - Computationally efficient for large datasets.
+                    - Works well when the number of features is high, which is common in genomic data where you may have thousands of genes
+                        as features.
+        'rbf':  a nonlinear kernel that maps the input data into a higher-dimensional space to find a decision boundary. It works well for
+            data that is not linearly separable in its original space.
+                Advantages:
+                    - Handles nonlinear relationships between features and classes
+                    - Often better than a linear kernel when there is no clear linear decision boundary in the data.
+        'poly': Polynomial Kernel: computes similarity between data points based on polynomial functions of the input features. It can model
+            interactions between features to a certain degree, depending on the polynomial degree chosen.
+                Advantages:
+                    - Allows modeling of feature interactions.
+                    - Can fit more complex relationships compared to linear models.
+        'sigmoid':  similar to the activation function in neural networks, and it works well when the data follows an S-shaped decision boundary.
+                Advantages:
+                - Can approximate the behavior of neural networks.
+                - Use case: It’s not as widely used as the RBF or linear kernel but can be explored when there is some evidence of non-linear
+                    S-shaped relationships.
+    """
+    # SVM (Support Vector Machines)
+    svc = SVC(kernel=rfe_params["kernel"])  # ["linear", "rbf", "poly", "sigmoid"]
+    # RFE(Recursive Feature Elimination)
+    selector = RFE(svc, n_features_to_select=rfe_params["n_features_to_select"])
+    selector.fit(x_train, y_train)
+    return x_train.columns[selector.support_]
+
+
+#! 5.Bayesian and Probabilistic Methods
+def features_naive_bayes(x_train: pd.DataFrame, y_train: pd.Series) -> list:
+    """
+    Naive Bayes: A probabilistic classifier based on Bayes' theorem, assuming independence between features, simple and fast, especially
+    effective for text classification and other high-dimensional data.
+    """
+    from sklearn.naive_bayes import GaussianNB
+
+    nb = GaussianNB()
+    nb.fit(x_train, y_train)
+    probabilities = nb.predict_proba(x_train)
+    # Limit the number of features safely, choosing the lesser of half the features or all columns
+    n_features = min(x_train.shape[1] // 2, len(x_train.columns))
+
+    # Sort probabilities, then map to valid column indices
+    sorted_indices = np.argsort(probabilities.max(axis=1))[:n_features]
+
+    # Ensure indices are within the column bounds of x_train
+    valid_indices = sorted_indices[sorted_indices < len(x_train.columns)]
+
+    return x_train.columns[valid_indices]
+
+
+#! 6.Linear Discriminant Analysis (LDA)
+def features_lda(x_train: pd.DataFrame, y_train: pd.Series) -> list:
+    """
+    Linear Discriminant Analysis (LDA): Projects data onto a lower-dimensional space to maximize class separability, often used as a dimensionality
+    reduction technique before classification on high-dimensional data.
+    """
+    from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
+
+    lda = LinearDiscriminantAnalysis()
+    lda.fit(x_train, y_train)
+    coef = lda.coef_.flatten()
+    # Create a DataFrame to hold feature importance
+    importance_df = pd.DataFrame(
+        {"feature": x_train.columns, "importance": np.abs(coef)}
+    )
+
+    return importance_df[importance_df["importance"] > 0].sort_values(
+        by="importance", ascending=False
+    )
+
+
+def features_adaboost(
+    x_train: pd.DataFrame, y_train: pd.Series, adaboost_params: dict
+) -> pd.DataFrame:
+    """
+    AdaBoost
+    Strengths:
+        Combines multiple weak learners to create a strong classifier.
+        Focuses on examples that are hard to classify, improving overall performance.
+    Recommended Use:
+        Can be effective for boosting weak models in a genomics context.
+    Fit AdaBoost classifier and return sorted feature importances.
+    Recommended Use: Great for classification problems with a large number of features (genes).
+    """
+    ada = AdaBoostClassifier(**adaboost_params)
+    ada.fit(x_train, y_train)
+    return pd.DataFrame(
+        {"feature": x_train.columns, "importance": ada.feature_importances_}
+    ).sort_values(by="importance", ascending=False)
+
+
+import torch
+import torch.nn as nn
+import torch.optim as optim
+from torch.utils.data import DataLoader, TensorDataset
+from skorch import NeuralNetClassifier  # sklearn compatible
+
+
+class DNNClassifier(nn.Module):
+    def __init__(self, input_dim, hidden_dim=128, output_dim=2, dropout_rate=0.5):
+        super(DNNClassifier, self).__init__()
+
+        self.hidden_layer1 = nn.Sequential(
+            nn.Linear(input_dim, hidden_dim),
+            nn.ReLU(),
+            nn.Dropout(dropout_rate),
+            nn.Linear(hidden_dim, hidden_dim),
+            nn.ReLU(),
+        )
+
+        self.hidden_layer2 = nn.Sequential(
+            nn.Linear(hidden_dim, hidden_dim), nn.ReLU(), nn.Dropout(dropout_rate)
+        )
+
+        # Adding a residual connection between hidden layers
+        self.residual = nn.Linear(input_dim, hidden_dim)
+
+        self.output_layer = nn.Sequential(
+            nn.Linear(hidden_dim, output_dim), nn.Softmax(dim=1)
+        )
+
+    def forward(self, x):
+        residual = self.residual(x)
+        x = self.hidden_layer1(x)
+        x = x + residual  # Residual connection
+        x = self.hidden_layer2(x)
+        x = self.output_layer(x)
+        return x
+
+
+def validate_classifier(
+    clf,
+    x_train: pd.DataFrame,
+    y_train: pd.Series,
+    x_test: pd.DataFrame,
+    y_test: pd.Series,
+    metrics: list = ["accuracy", "precision", "recall", "f1", "roc_auc"],
+    cv_folds: int = 5,
+) -> dict:
+    """
+    Perform cross-validation for a given classifier and return average scores for specified metrics on training data.
+    Then fit the best model on the full training data and evaluate it on the test set.
+
+    Parameters:
+    - clf: The classifier to be validated.
+    - x_train: Training features.
+    - y_train: Training labels.
+    - x_test: Test features.
+    - y_test: Test labels.
+    - metrics: List of metrics to evaluate (e.g., ['accuracy', 'roc_auc']).
+    - cv_folds: Number of cross-validation folds.
+
+    Returns:
+    - results: Dictionary containing average cv_train_scores and cv_test_scores.
+    """
+    cv_train_scores = {metric: [] for metric in metrics}
+    skf = StratifiedKFold(n_splits=cv_folds)
+    # Perform cross-validation
+    for metric in metrics:
+        try:
+            if metric == "roc_auc" and len(set(y_train)) == 2:
+                scores = cross_val_score(
+                    clf, x_train, y_train, cv=skf, scoring="roc_auc"
+                )
+                cv_train_scores[metric] = (
+                    np.nanmean(scores) if not np.isnan(scores).all() else float("nan")
+                )
+            else:
+                score = cross_val_score(clf, x_train, y_train, cv=skf, scoring=metric)
+                cv_train_scores[metric] = score.mean()
+        except Exception as e:
+            cv_train_scores[metric] = float("nan")
+    clf.fit(x_train, y_train)
+
+    # Evaluate on the test set
+    cv_test_scores = {}
+    for metric in metrics:
+        if metric == "roc_auc" and len(set(y_test)) == 2:
+            try:
+                y_prob = clf.predict_proba(x_test)[:, 1]
+                cv_test_scores[metric] = roc_auc_score(y_test, y_prob)
+            except AttributeError:
+                cv_test_scores[metric] = float("nan")
+        else:
+            score_func = globals().get(
+                f"{metric}_score"
+            )  # Fetching the appropriate scoring function
+            if score_func:
+                try:
+                    y_pred = clf.predict(x_test)
+                    cv_test_scores[metric] = score_func(y_test, y_pred)
+                except Exception as e:
+                    cv_test_scores[metric] = float("nan")
+
+    # Combine results
+    results = {"cv_train_scores": cv_train_scores, "cv_test_scores": cv_test_scores}
+    return results
+
+
+def get_models(
+    random_state=1,
+    cls=[
+        "lasso",
+        "ridge",
+        "Elastic Net(Enet)",
+        "gradient Boosting",
+        "Random forest (rf)",
+        "XGBoost (xgb)",
+        "Support Vector Machine(svm)",
+        "naive bayes",
+        "Linear Discriminant Analysis (lda)",
+        "adaboost",
+        "DecisionTree",
+        "KNeighbors",
+        "Bagging",
+    ],
+):
+    from sklearn.ensemble import (
+        RandomForestClassifier,
+        GradientBoostingClassifier,
+        AdaBoostClassifier,
+        BaggingClassifier,
+    )
+    from sklearn.svm import SVC
+    from sklearn.linear_model import (
+        LogisticRegression,
+        Lasso,
+        RidgeClassifierCV,
+        ElasticNet,
+    )
+    from sklearn.naive_bayes import GaussianNB
+    from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
+    import xgboost as xgb
+    from sklearn.tree import DecisionTreeClassifier
+    from sklearn.neighbors import KNeighborsClassifier
+
+    res_cls = {}
+    model_all = {
+        "Lasso": LogisticRegression(
+            penalty="l1", solver="saga", random_state=random_state
+        ),
+        "Ridge": RidgeClassifierCV(),
+        "Elastic Net (Enet)": ElasticNet(random_state=random_state),
+        "Gradient Boosting": GradientBoostingClassifier(random_state=random_state),
+        "Random Forest (RF)": RandomForestClassifier(random_state=random_state),
+        "XGBoost (XGB)": xgb.XGBClassifier(random_state=random_state),
+        "Support Vector Machine (SVM)": SVC(kernel="rbf", probability=True),
+        "Naive Bayes": GaussianNB(),
+        "Linear Discriminant Analysis (LDA)": LinearDiscriminantAnalysis(),
+        "AdaBoost": AdaBoostClassifier(random_state=random_state, algorithm="SAMME"),
+        "DecisionTree": DecisionTreeClassifier(),
+        "KNeighbors": KNeighborsClassifier(n_neighbors=5),
+        "Bagging": BaggingClassifier(),
+    }
+    print("Using default models:")
+    for cls_name in cls:
+        cls_name = ips.strcmp(cls_name, list(model_all.keys()))[0]
+        res_cls[cls_name] = model_all[cls_name]
+        print(f"- {cls_name}")
+    return res_cls
+
+
+def get_features(
+    X: Union[pd.DataFrame, np.ndarray],  # n_samples X n_features
+    y: Union[pd.Series, np.ndarray, list],  # n_samples X n_features
+    test_size: float = 0.2,
+    random_state: int = 1,
+    n_features: int = 10,
+    fill_missing=True,
+    rf_params: Optional[Dict] = None,
+    rfe_params: Optional[Dict] = None,
+    lasso_params: Optional[Dict] = None,
+    ridge_params: Optional[Dict] = None,
+    enet_params: Optional[Dict] = None,
+    gb_params: Optional[Dict] = None,
+    adaboost_params: Optional[Dict] = None,
+    xgb_params: Optional[Dict] = None,
+    dt_params: Optional[Dict] = None,
+    bagging_params: Optional[Dict] = None,
+    knn_params: Optional[Dict] = None,
+    cls: list = [
+        "lasso",
+        "ridge",
+        "Elastic Net(Enet)",
+        "gradient Boosting",
+        "Random forest (rf)",
+        "XGBoost (xgb)",
+        "Support Vector Machine(svm)",
+        "naive bayes",
+        "Linear Discriminant Analysis (lda)",
+        "adaboost",
+        "DecisionTree",
+        "KNeighbors",
+        "Bagging",
+    ],
+    metrics: Optional[List[str]] = None,
+    cv_folds: int = 5,
+    strict: bool = False,
+    n_shared: int = 2,  # 只要有两个方法有重合,就纳入common genes
+    use_selected_features: bool = True,
+    plot_: bool = True,
+    dir_save: str = "./",
+) -> dict:
+    """
+    Master function to perform feature selection and validate models.
+    """
+    from sklearn.compose import ColumnTransformer
+    from sklearn.preprocessing import StandardScaler, OneHotEncoder
+
+    # Ensure X and y are DataFrames/Series for consistency
+    if isinstance(X, np.ndarray):
+        X = pd.DataFrame(X)
+    if isinstance(y, (np.ndarray, list)):
+        y = pd.Series(y)
+
+    # fill na
+    if fill_missing:
+        ips.df_fillna(data=X, method="knn", inplace=True, axis=0)
+    if isinstance(y, str) and y in X.columns:
+        y_col_name = y
+        y = X[y]
+        y = ips.df_encoder(pd.DataFrame(y), method="dummy")
+        X = X.drop(y_col_name, axis=1)
+    else:
+        y = ips.df_encoder(pd.DataFrame(y), method="dummy").values.ravel()
+    y = y.loc[X.index]  # Align y with X after dropping rows with missing values in X
+    y = y.ravel() if isinstance(y, np.ndarray) else y.values.ravel()
+
+    if X.shape[0] != len(y):
+        raise ValueError("X and y must have the same number of samples (rows).")
+
+    # #! # Check for non-numeric columns in X and apply one-hot encoding if needed
+    # Check if any column in X is non-numeric
+    if any(not np.issubdtype(dtype, np.number) for dtype in X.dtypes):
+        X = pd.get_dummies(X, drop_first=True)
+    print(X.shape)
+
+    # #!alternative:  # Identify categorical and numerical columns
+    # categorical_cols = X.select_dtypes(include=["object", "category"]).columns
+    # numerical_cols = X.select_dtypes(include=["number"]).columns
+
+    # # Define preprocessing pipeline
+    # preprocessor = ColumnTransformer(
+    #     transformers=[
+    #         ("num", StandardScaler(), numerical_cols),
+    #         ("cat", OneHotEncoder(drop="first", handle_unknown="ignore"), categorical_cols),
+    #     ]
+    # )
+    # # Preprocess the data
+    # X = preprocessor.fit_transform(X)
+
+    # Split data into training and test sets
+    x_train, x_test, y_train, y_test = train_test_split(
+        X, y, test_size=test_size, random_state=random_state
+    )
+    # Standardize features
+    scaler = StandardScaler()
+    x_train_scaled = scaler.fit_transform(x_train)
+    x_test_scaled = scaler.transform(x_test)
+
+    # Convert back to DataFrame for consistency
+    x_train = pd.DataFrame(x_train_scaled, columns=x_train.columns)
+    x_test = pd.DataFrame(x_test_scaled, columns=x_test.columns)
+
+    rf_defaults = {"n_estimators": 100, "random_state": random_state}
+    rfe_defaults = {"kernel": "linear", "n_features_to_select": n_features}
+    lasso_defaults = {"alphas": np.logspace(-4, 4, 100), "cv": 10}
+    ridge_defaults = {"alphas": np.logspace(-4, 4, 100), "cv": 10}
+    enet_defaults = {"alphas": np.logspace(-4, 4, 100), "cv": 10}
+    xgb_defaults = {
+        "n_estimators": 100,
+        "use_label_encoder": False,
+        "eval_metric": "logloss",
+        "random_state": random_state,
+    }
+    gb_defaults = {"n_estimators": 100, "random_state": random_state}
+    adaboost_defaults = {"n_estimators": 50, "random_state": random_state}
+    dt_defaults = {"max_depth": None, "random_state": random_state}
+    bagging_defaults = {"n_estimators": 50, "random_state": random_state}
+    knn_defaults = {"n_neighbors": 5}
+    rf_params, rfe_params = rf_params or rf_defaults, rfe_params or rfe_defaults
+    lasso_params, ridge_params = (
+        lasso_params or lasso_defaults,
+        ridge_params or ridge_defaults,
+    )
+    enet_params, xgb_params = enet_params or enet_defaults, xgb_params or xgb_defaults
+    gb_params, adaboost_params = (
+        gb_params or gb_defaults,
+        adaboost_params or adaboost_defaults,
+    )
+    dt_params = dt_params or dt_defaults
+    bagging_params = bagging_params or bagging_defaults
+    knn_params = knn_params or knn_defaults
+
+    cls_ = [
+        "lasso",
+        "ridge",
+        "Elastic Net(Enet)",
+        "Gradient Boosting",
+        "Random Forest (rf)",
+        "XGBoost (xgb)",
+        "Support Vector Machine(svm)",
+        "Naive Bayes",
+        "Linear Discriminant Analysis (lda)",
+        "adaboost",
+    ]
+    cls = [ips.strcmp(i, cls_)[0] for i in cls]
+
+    # Lasso Feature Selection
+    lasso_importances = (
+        features_lasso(x_train, y_train, lasso_params)
+        if "lasso" in cls
+        else pd.DataFrame()
+    )
+    lasso_selected_features = (
+        lasso_importances.head(n_features)["feature"].values if "lasso" in cls else []
+    )
+    # Ridge
+    ridge_importances = (
+        features_ridge(x_train, y_train, ridge_params)
+        if "ridge" in cls
+        else pd.DataFrame()
+    )
+    selected_ridge_features = (
+        ridge_importances.head(n_features)["feature"].values if "ridge" in cls else []
+    )
+    # Elastic Net
+    enet_importances = (
+        features_enet(x_train, y_train, enet_params)
+        if "Enet" in cls
+        else pd.DataFrame()
+    )
+    selected_enet_features = (
+        enet_importances.head(n_features)["feature"].values if "Enet" in cls else []
+    )
+    # Random Forest Feature Importance
+    rf_importances = (
+        features_rf(x_train, y_train, rf_params)
+        if "Random Forest" in cls
+        else pd.DataFrame()
+    )
+    top_rf_features = (
+        rf_importances.head(n_features)["feature"].values
+        if "Random Forest" in cls
+        else []
+    )
+    # Gradient Boosting Feature Importance
+    gb_importances = (
+        features_gradient_boosting(x_train, y_train, gb_params)
+        if "Gradient Boosting" in cls
+        else pd.DataFrame()
+    )
+    top_gb_features = (
+        gb_importances.head(n_features)["feature"].values
+        if "Gradient Boosting" in cls
+        else []
+    )
+    # xgb
+    xgb_importances = (
+        features_xgb(x_train, y_train, xgb_params) if "xgb" in cls else pd.DataFrame()
+    )
+    top_xgb_features = (
+        xgb_importances.head(n_features)["feature"].values if "xgb" in cls else []
+    )
+
+    # SVM with RFE
+    selected_svm_features = (
+        features_svm(x_train, y_train, rfe_params) if "svm" in cls else []
+    )
+    # Naive Bayes
+    selected_naive_bayes_features = (
+        features_naive_bayes(x_train, y_train) if "Naive Bayes" in cls else []
+    )
+    # lda: linear discriminant analysis
+    lda_importances = features_lda(x_train, y_train) if "lda" in cls else pd.DataFrame()
+    selected_lda_features = (
+        lda_importances.head(n_features)["feature"].values if "lda" in cls else []
+    )
+    # AdaBoost Feature Importance
+    adaboost_importances = (
+        features_adaboost(x_train, y_train, adaboost_params)
+        if "AdaBoost" in cls
+        else pd.DataFrame()
+    )
+    top_adaboost_features = (
+        adaboost_importances.head(n_features)["feature"].values
+        if "AdaBoost" in cls
+        else []
+    )
+    # Decision Tree Feature Importance
+    dt_importances = (
+        features_decision_tree(x_train, y_train, dt_params)
+        if "Decision Tree" in cls
+        else pd.DataFrame()
+    )
+    top_dt_features = (
+        dt_importances.head(n_features)["feature"].values
+        if "Decision Tree" in cls
+        else []
+    )
+    # Bagging Feature Importance
+    bagging_importances = (
+        features_bagging(x_train, y_train, bagging_params)
+        if "Bagging" in cls
+        else pd.DataFrame()
+    )
+    top_bagging_features = (
+        bagging_importances.head(n_features)["feature"].values
+        if "Bagging" in cls
+        else []
+    )
+    # KNN Feature Importance via Permutation
+    knn_importances = (
+        features_knn(x_train, y_train, knn_params) if "KNN" in cls else pd.DataFrame()
+    )
+    top_knn_features = (
+        knn_importances.head(n_features)["feature"].values if "KNN" in cls else []
+    )
+
+    #! Find common features
+    common_features = ips.shared(
+        lasso_selected_features,
+        selected_ridge_features,
+        selected_enet_features,
+        top_rf_features,
+        top_gb_features,
+        top_xgb_features,
+        selected_svm_features,
+        selected_naive_bayes_features,
+        selected_lda_features,
+        top_adaboost_features,
+        top_dt_features,
+        top_bagging_features,
+        top_knn_features,
+        strict=strict,
+        n_shared=n_shared,
+        verbose=False,
+    )
+
+    # Use selected features or all features for model validation
+    x_train_selected = (
+        x_train[list(common_features)] if use_selected_features else x_train
+    )
+    x_test_selected = x_test[list(common_features)] if use_selected_features else x_test
+
+    if metrics is None:
+        metrics = ["accuracy", "precision", "recall", "f1", "roc_auc"]
+
+    # Prepare results DataFrame for selected features
+    features_df = pd.DataFrame(
+        {
+            "type": ["Lasso"] * len(lasso_selected_features)
+            + ["Ridge"] * len(selected_ridge_features)
+            + ["Random Forest"] * len(top_rf_features)
+            + ["Gradient Boosting"] * len(top_gb_features)
+            + ["Enet"] * len(selected_enet_features)
+            + ["xgb"] * len(top_xgb_features)
+            + ["SVM"] * len(selected_svm_features)
+            + ["Naive Bayes"] * len(selected_naive_bayes_features)
+            + ["Linear Discriminant Analysis"] * len(selected_lda_features)
+            + ["AdaBoost"] * len(top_adaboost_features)
+            + ["Decision Tree"] * len(top_dt_features)
+            + ["Bagging"] * len(top_bagging_features)
+            + ["KNN"] * len(top_knn_features),
+            "feature": np.concatenate(
+                [
+                    lasso_selected_features,
+                    selected_ridge_features,
+                    top_rf_features,
+                    top_gb_features,
+                    selected_enet_features,
+                    top_xgb_features,
+                    selected_svm_features,
+                    selected_naive_bayes_features,
+                    selected_lda_features,
+                    top_adaboost_features,
+                    top_dt_features,
+                    top_bagging_features,
+                    top_knn_features,
+                ]
+            ),
+        }
+    )
+
+    #! Validate trained each classifier
+    models = get_models(random_state=random_state, cls=cls)
+    cv_train_results, cv_test_results = [], []
+    for name, clf in models.items():
+        if not x_train_selected.empty:
+            cv_scores = validate_classifier(
+                clf,
+                x_train_selected,
+                y_train,
+                x_test_selected,
+                y_test,
+                metrics=metrics,
+                cv_folds=cv_folds,
+            )
+
+            cv_train_score_df = pd.DataFrame(cv_scores["cv_train_scores"], index=[name])
+            cv_test_score_df = pd.DataFrame(cv_scores["cv_test_scores"], index=[name])
+            cv_train_results.append(cv_train_score_df)
+            cv_test_results.append(cv_test_score_df)
+    if all([cv_train_results, cv_test_results]):
+        cv_train_results_df = (
+            pd.concat(cv_train_results)
+            .reset_index()
+            .rename(columns={"index": "Classifier"})
+        )
+        cv_test_results_df = (
+            pd.concat(cv_test_results)
+            .reset_index()
+            .rename(columns={"index": "Classifier"})
+        )
+        #! Store results in the main results dictionary
+        results = {
+            "selected_features": features_df,
+            "cv_train_scores": cv_train_results_df,
+            "cv_test_scores": rank_models(cv_test_results_df, plot_=plot_),
+            "common_features": list(common_features),
+        }
+        if all([plot_, dir_save]):
+            from datetime import datetime
+
+            now_ = datetime.now().strftime("%y%m%d_%H%M%S")
+            ips.figsave(dir_save + f"features{now_}.pdf")
+    else:
+        results = {
+            "selected_features": pd.DataFrame(),
+            "cv_train_scores": pd.DataFrame(),
+            "cv_test_scores": pd.DataFrame(),
+            "common_features": [],
+        }
+        print(f"Warning: 没有找到共同的genes, when n_shared={n_shared}")
+    return results
+
+
+#! # usage:
+# # Get features and common features
+# results = get_features(X, y)
+# common_features = results["common_features"]
+def validate_features(
+    x_train: pd.DataFrame,
+    y_train: pd.Series,
+    x_true: pd.DataFrame,
+    y_true: pd.Series,
+    common_features: set = None,
+    models: Optional[Dict[str, Any]] = None,
+    metrics: Optional[list] = None,
+    random_state: int = 1,
+    smote: bool = False,
+    n_jobs: int = -1,
+    plot_: bool = True,
+    class_weight: str = "balanced",
+) -> dict:
+    """
+    Validate models using selected features on the validation dataset.
+
+    Parameters:
+    - x_train (pd.DataFrame): Training feature dataset.
+    - y_train (pd.Series): Training target variable.
+    - x_true (pd.DataFrame): Validation feature dataset.
+    - y_true (pd.Series): Validation target variable.
+    - common_features (set): Set of common features to use for validation.
+    - models (dict, optional): Dictionary of models to validate.
+    - metrics (list, optional): List of metrics to compute.
+    - random_state (int): Random state for reproducibility.
+    - plot_ (bool): Option to plot metrics (to be implemented if needed).
+    - class_weight (str or dict): Class weights to handle imbalance.
+
+    """
+    from tqdm import tqdm
+
+    # Ensure common features are selected
+    common_features = ips.shared(
+        common_features, x_train.columns, x_true.columns, strict=True, verbose=False
+    )
+
+    # Filter the training and validation datasets for the common features
+    x_train_selected = x_train[common_features]
+    x_true_selected = x_true[common_features]
+
+    if not x_true_selected.index.equals(y_true.index):
+        raise ValueError(
+            "Index mismatch between validation features and target. Ensure data alignment."
+        )
+
+    y_true = y_true.loc[x_true_selected.index]
+
+    # Handle class imbalance using SMOTE
+    if smote:
+        if (
+            y_train.value_counts(normalize=True).max() < 0.8
+        ):  # Threshold to decide if data is imbalanced
+            smote = SMOTE(random_state=random_state)
+            x_train_resampled, y_train_resampled = smote.fit_resample(
+                x_train_selected, y_train
+            )
+        else:
+            # skip SMOTE
+            x_train_resampled, y_train_resampled = x_train_selected, y_train
+    else:
+        x_train_resampled, y_train_resampled = x_train_selected, y_train
+
+    # Default models if not provided
+    if models is None:
+        models = {
+            "Random Forest": RandomForestClassifier(
+                class_weight=class_weight, random_state=random_state
+            ),
+            "SVM": SVC(probability=True, class_weight=class_weight),
+            "Logistic Regression": LogisticRegression(
+                class_weight=class_weight, random_state=random_state
+            ),
+            "Gradient Boosting": GradientBoostingClassifier(random_state=random_state),
+            "AdaBoost": AdaBoostClassifier(
+                random_state=random_state, algorithm="SAMME"
+            ),
+            "Lasso": LogisticRegression(
+                penalty="l1", solver="saga", random_state=random_state
+            ),
+            "Ridge": LogisticRegression(
+                penalty="l2", solver="saga", random_state=random_state
+            ),
+            "Elastic Net": LogisticRegression(
+                penalty="elasticnet",
+                solver="saga",
+                l1_ratio=0.5,
+                random_state=random_state,
+            ),
+            "XGBoost": xgb.XGBClassifier(eval_metric="logloss"),
+            "Naive Bayes": GaussianNB(),
+            "LDA": LinearDiscriminantAnalysis(),
+        }
+
+    # Hyperparameter grids for tuning
+    param_grids = {
+        "Random Forest": {
+            "n_estimators": [100, 200, 300, 400, 500],
+            "max_depth": [None, 3, 5, 10, 20],
+            "min_samples_split": [2, 5, 10],
+            "min_samples_leaf": [1, 2, 4],
+            "class_weight": [None, "balanced"],
+        },
+        "SVM": {
+            "C": [0.01, 0.1, 1, 10, 100, 1000],
+            "gamma": [0.001, 0.01, 0.1, "scale", "auto"],
+            "kernel": ["linear", "rbf", "poly"],
+        },
+        "Logistic Regression": {
+            "C": [0.01, 0.1, 1, 10, 100],
+            "solver": ["liblinear", "saga", "newton-cg", "lbfgs"],
+            "penalty": ["l1", "l2"],
+            "max_iter": [100, 200, 300],
+        },
+        "Gradient Boosting": {
+            "n_estimators": [100, 200, 300, 400, 500],
+            "learning_rate": np.logspace(-3, 0, 4),
+            "max_depth": [3, 5, 7, 9],
+            "min_samples_split": [2, 5, 10],
+        },
+        "AdaBoost": {
+            "n_estimators": [50, 100, 200, 300, 500],
+            "learning_rate": np.logspace(-3, 0, 4),
+        },
+        "Lasso": {"C": np.logspace(-3, 1, 10), "max_iter": [100, 200, 300]},
+        "Ridge": {"C": np.logspace(-3, 1, 10), "max_iter": [100, 200, 300]},
+        "Elastic Net": {
+            "C": np.logspace(-3, 1, 10),
+            "l1_ratio": [0.1, 0.5, 0.9],
+            "max_iter": [100, 200, 300],
+        },
+        "XGBoost": {
+            "n_estimators": [100, 200],
+            "max_depth": [3, 5, 7],
+            "learning_rate": [0.01, 0.1, 0.2],
+            "subsample": [0.8, 1.0],
+            "colsample_bytree": [0.8, 1.0],
+        },
+        "Naive Bayes": {},
+        "LDA": {"solver": ["svd", "lsqr", "eigen"]},
+    }
+    # Default metrics if not provided
+    if metrics is None:
+        metrics = [
+            "accuracy",
+            "precision",
+            "recall",
+            "f1",
+            "roc_auc",
+            "mcc",
+            "specificity",
+            "balanced_accuracy",
+            "pr_auc",
+        ]
+
+    results = {}
+
+    # Validate each classifier with GridSearchCV
+    for name, clf in tqdm(
+        models.items(),
+        desc="for metric in metrics",
+        colour="green",
+        bar_format="{l_bar}{bar} {n_fmt}/{total_fmt}",
+    ):
+        print(f"\nValidating {name} on the validation dataset:")
+
+        # Check if `predict_proba` method exists; if not, use CalibratedClassifierCV
+        # 没有predict_proba的分类器，使用 CalibratedClassifierCV 可以获得校准的概率估计。此外，为了使代码更灵活，我们可以在创建分类器
+        # 时检查 predict_proba 方法是否存在，如果不存在且用户希望计算 roc_auc 或 pr_auc，则启用 CalibratedClassifierCV
+        if not hasattr(clf, "predict_proba"):
+            print(
+                f"Using CalibratedClassifierCV for {name} due to lack of probability estimates."
+            )
+            calibrated_clf = CalibratedClassifierCV(clf, method="sigmoid", cv="prefit")
+        else:
+            calibrated_clf = clf
+        # Stratified K-Fold for cross-validation
+        skf = StratifiedKFold(n_splits=5, shuffle=True, random_state=random_state)
+
+        # Create GridSearchCV object
+        gs = GridSearchCV(
+            estimator=calibrated_clf,
+            param_grid=param_grids[name],
+            scoring="roc_auc",  # Optimize for ROC AUC
+            cv=skf,  # Stratified K-Folds cross-validation
+            n_jobs=n_jobs,
+            verbose=1,
+        )
+
+        # Fit the model using GridSearchCV
+        gs.fit(x_train_resampled, y_train_resampled)
+        # Best estimator from grid search
+        best_clf = gs.best_estimator_
+        # Make predictions on the validation set
+        y_pred = best_clf.predict(x_true_selected)
+        # Calculate probabilities for ROC AUC if possible
+        if hasattr(best_clf, "predict_proba"):
+            y_pred_proba = best_clf.predict_proba(x_true_selected)[:, 1]
+        elif hasattr(best_clf, "decision_function"):
+            # If predict_proba is not available, use decision_function (e.g., for SVM)
+            y_pred_proba = best_clf.decision_function(x_true_selected)
+            # Ensure y_pred_proba is within 0 and 1 bounds
+            y_pred_proba = (y_pred_proba - y_pred_proba.min()) / (
+                y_pred_proba.max() - y_pred_proba.min()
+            )
+        else:
+            y_pred_proba = None  # No probability output for certain models
+
+        # Calculate metrics
+        validation_scores = {}
+        for metric in metrics:
+            if metric == "accuracy":
+                validation_scores[metric] = accuracy_score(y_true, y_pred)
+            elif metric == "precision":
+                validation_scores[metric] = precision_score(
+                    y_true, y_pred, average="weighted"
+                )
+            elif metric == "recall":
+                validation_scores[metric] = recall_score(
+                    y_true, y_pred, average="weighted"
+                )
+            elif metric == "f1":
+                validation_scores[metric] = f1_score(y_true, y_pred, average="weighted")
+            elif metric == "roc_auc" and y_pred_proba is not None:
+                validation_scores[metric] = roc_auc_score(y_true, y_pred_proba)
+            elif metric == "mcc":
+                validation_scores[metric] = matthews_corrcoef(y_true, y_pred)
+            elif metric == "specificity":
+                tn, fp, fn, tp = confusion_matrix(y_true, y_pred).ravel()
+                validation_scores[metric] = tn / (tn + fp)  # Specificity calculation
+            elif metric == "balanced_accuracy":
+                validation_scores[metric] = balanced_accuracy_score(y_true, y_pred)
+            elif metric == "pr_auc" and y_pred_proba is not None:
+                precision, recall, _ = precision_recall_curve(y_true, y_pred_proba)
+                validation_scores[metric] = average_precision_score(
+                    y_true, y_pred_proba
+                )
+
+        # Calculate ROC curve
+        # https://scikit-learn.org/stable/auto_examples/model_selection/plot_roc.html
+        if y_pred_proba is not None:
+            # fpr, tpr, roc_auc = dict(), dict(), dict()
+            fpr, tpr, _ = roc_curve(y_true, y_pred_proba)
+            lower_ci, upper_ci = cal_auc_ci(y_true, y_pred_proba, verbose=False)
+            roc_auc = auc(fpr, tpr)
+            roc_info = {
+                "fpr": fpr.tolist(),
+                "tpr": tpr.tolist(),
+                "auc": roc_auc,
+                "ci95": (lower_ci, upper_ci),
+            }
+            # precision-recall curve
+            precision_, recall_, _ = precision_recall_curve(y_true, y_pred_proba)
+            avg_precision_ = average_precision_score(y_true, y_pred_proba)
+            pr_info = {
+                "precision": precision_,
+                "recall": recall_,
+                "avg_precision": avg_precision_,
+            }
+        else:
+            roc_info, pr_info = None, None
+        results[name] = {
+            "best_params": gs.best_params_,
+            "scores": validation_scores,
+            "roc_curve": roc_info,
+            "pr_curve": pr_info,
+            "confusion_matrix": confusion_matrix(y_true, y_pred),
+        }
+
+    df_results = pd.DataFrame.from_dict(results, orient="index")
+
+    return df_results
+
+
+#! usage validate_features()
+# Validate models using the validation dataset (X_val, y_val)
+# validation_results = validate_features(X, y, X_val, y_val, common_features)
+
+
+# # If you want to access validation scores
+# print(validation_results)
+def plot_validate_features(res_val):
+    """
+    plot the results of 'validate_features()'
+    """
+    colors = plot.get_color(len(ips.flatten(res_val["pr_curve"].index)))
+    if res_val.shape[0] > 5:
+        alpha = 0
+        figsize = [8, 10]
+        subplot_layout = [1, 2]
+        ncols = 2
+        bbox_to_anchor = [1.5, 0.6]
+    else:
+        alpha = 0.03
+        figsize = [10, 6]
+        subplot_layout = [1, 1]
+        ncols = 1
+        bbox_to_anchor = [1, 1]
+    nexttile = plot.subplot(figsize=figsize)
+    ax = nexttile(subplot_layout[0], subplot_layout[1])
+    for i, model_name in enumerate(ips.flatten(res_val["pr_curve"].index)):
+        fpr = res_val["roc_curve"][model_name]["fpr"]
+        tpr = res_val["roc_curve"][model_name]["tpr"]
+        (lower_ci, upper_ci) = res_val["roc_curve"][model_name]["ci95"]
+        mean_auc = res_val["roc_curve"][model_name]["auc"]
+        plot_roc_curve(
+            fpr,
+            tpr,
+            mean_auc,
+            lower_ci,
+            upper_ci,
+            model_name=model_name,
+            lw=1.5,
+            color=colors[i],
+            alpha=alpha,
+            ax=ax,
+        )
+    plot.figsets(
+        sp=2,
+        legend=dict(
+            loc="upper right",
+            ncols=ncols,
+            fontsize=8,
+            bbox_to_anchor=[1.5, 0.6],
+            markerscale=0.8,
+        ),
+    )
+    # plot.split_legend(ax,n=2, loc=["upper left", "lower left"],bbox=[[1,0.5],[1,0.5]],ncols=2,labelcolor="k",fontsize=8)
+
+    ax = nexttile(subplot_layout[0], subplot_layout[1])
+    for i, model_name in enumerate(ips.flatten(res_val["pr_curve"].index)):
+        plot_pr_curve(
+            recall=res_val["pr_curve"][model_name]["recall"],
+            precision=res_val["pr_curve"][model_name]["precision"],
+            avg_precision=res_val["pr_curve"][model_name]["avg_precision"],
+            model_name=model_name,
+            color=colors[i],
+            lw=1.5,
+            alpha=alpha,
+            ax=ax,
+        )
+    plot.figsets(
+        sp=2,
+        legend=dict(loc="upper right", ncols=1, fontsize=8, bbox_to_anchor=[1.5, 0.5]),
+    )
+    # plot.split_legend(ax,n=2, loc=["upper left", "lower left"],bbox=[[1,0.5],[1,0.5]],ncols=2,labelcolor="k",fontsize=8)
+
+
+def plot_validate_features_single(res_val, figsize=None):
+    if figsize is None:
+        nexttile = plot.subplot(len(ips.flatten(res_val["pr_curve"].index)), 3)
+    else:
+        nexttile = plot.subplot(
+            len(ips.flatten(res_val["pr_curve"].index)), 3, figsize=figsize
+        )
+    for model_name in ips.flatten(res_val["pr_curve"].index):
+        fpr = res_val["roc_curve"][model_name]["fpr"]
+        tpr = res_val["roc_curve"][model_name]["tpr"]
+        (lower_ci, upper_ci) = res_val["roc_curve"][model_name]["ci95"]
+        mean_auc = res_val["roc_curve"][model_name]["auc"]
+
+        # Plotting
+        plot_roc_curve(fpr, tpr, mean_auc, lower_ci, upper_ci,
+            model_name=model_name, ax=nexttile())
+        plot.figsets(title=model_name, sp=2)
+
+        plot_pr_binary(
+            recall=res_val["pr_curve"][model_name]["recall"],
+            precision=res_val["pr_curve"][model_name]["precision"],
+            avg_precision=res_val["pr_curve"][model_name]["avg_precision"],
+            model_name=model_name,
+            ax=nexttile(),
+        )
+        plot.figsets(title=model_name, sp=2)
+
+        # plot cm
+        plot_cm(res_val["confusion_matrix"][model_name], ax=nexttile(), normalize=False)
+        plot.figsets(title=model_name, sp=2)
+
+
+def cal_auc_ci(
+    y_true, y_pred, n_bootstraps=1000, ci=0.95, random_state=1, verbose=True
+):
+    y_true = np.asarray(y_true)
+    y_pred = np.asarray(y_pred)
+    bootstrapped_scores = []
+    if verbose:
+        print("auroc score:", roc_auc_score(y_true, y_pred))
+    rng = np.random.RandomState(random_state)
+    for i in range(n_bootstraps):
+        # bootstrap by sampling with replacement on the prediction indices
+        indices = rng.randint(0, len(y_pred), len(y_pred))
+        if len(np.unique(y_true[indices])) < 2:
+            # We need at least one positive and one negative sample for ROC AUC
+            # to be defined: reject the sample
+            continue
+        if isinstance(y_true, np.ndarray):
+            score = roc_auc_score(y_true[indices], y_pred[indices])
+        else:
+            score = roc_auc_score(y_true.iloc[indices], y_pred.iloc[indices])
+        bootstrapped_scores.append(score)
+        # print("Bootstrap #{} ROC area: {:0.3f}".format(i + 1, score))
+    sorted_scores = np.array(bootstrapped_scores)
+    sorted_scores.sort()
+
+    # Computing the lower and upper bound of the 90% confidence interval
+    # You can change the bounds percentiles to 0.025 and 0.975 to get
+    # a 95% confidence interval instead.
+    confidence_lower = sorted_scores[int((1 - ci) * len(sorted_scores))]
+    confidence_upper = sorted_scores[int(ci * len(sorted_scores))]
+    if verbose:
+        print(
+            "Confidence interval for the score: [{:0.3f} - {:0.3}]".format(
+                confidence_lower, confidence_upper
+            )
+        )
+    return confidence_lower, confidence_upper
+
+
+def plot_roc_curve(
+    fpr=None,
+    tpr=None,
+    mean_auc=None,
+    lower_ci=None,
+    upper_ci=None,
+    model_name=None,
+    color="#FF8F00",
+    lw=2,
+    alpha=0.1,
+    ci_display=True,
+    title="ROC Curve",
+    xlabel="1−Specificity",
+    ylabel="Sensitivity",
+    legend_loc="lower right",
+    diagonal_color="0.5",
+    figsize=(5, 5),
+    ax=None,
+    **kwargs,
+):
+    if ax is None:
+        fig, ax = plt.subplots(figsize=figsize)
+    if mean_auc is not None:
+        model_name = "ROC curve" if model_name is None else model_name
+        if ci_display:
+            label = f"{model_name} (AUC = {mean_auc:.3f})\n95% CI: {lower_ci:.3f} - {upper_ci:.3f}"
+        else:
+            label = f"{model_name} (AUC = {mean_auc:.3f})"
+    else:
+        label = None
+
+    # Plot ROC curve and the diagonal reference line
+    ax.fill_between(fpr, tpr, alpha=alpha, color=color)
+    ax.plot([0, 1], [0, 1], color=diagonal_color, clip_on=False, linestyle="--")
+    ax.plot(fpr, tpr, color=color, lw=lw, label=label, clip_on=False, **kwargs)
+    # Setting plot limits, labels, and title
+    ax.set_xlim([-0.01, 1.0])
+    ax.set_ylim([0.0, 1.0])
+    ax.set_xlabel(xlabel)
+    ax.set_ylabel(ylabel)
+    ax.set_title(title)
+    ax.legend(loc=legend_loc)
+    return ax
+
+
+# * usage: ml2ls.plot_roc_curve(fpr, tpr, mean_auc, lower_ci, upper_ci)
+# for model_name in flatten(validation_results["roc_curve"].keys())[2:]:
+#     fpr = validation_results["roc_curve"][model_name]["fpr"]
+#     tpr = validation_results["roc_curve"][model_name]["tpr"]
+#     (lower_ci, upper_ci) = validation_results["roc_curve"][model_name]["ci95"]
+#     mean_auc = validation_results["roc_curve"][model_name]["auc"]
+
+#     # Plotting
+#     ml2ls.plot_roc_curve(fpr, tpr, mean_auc, lower_ci, upper_ci)
+#     figsets(title=model_name)
+
+def plot_pr_curve(
+    recall=None,
+    precision=None,
+    avg_precision=None,
+    model_name=None,
+    lw=2,
+    figsize=[5, 5],
+    title="Precision-Recall Curve",
+    xlabel="Recall",
+    ylabel="Precision",
+    alpha=0.1,
+    color="#FF8F00",
+    legend_loc="lower left",
+    ax=None,
+    **kwargs,
+):
+    if ax is None:
+        fig, ax = plt.subplots(figsize=figsize)
+    model_name = "PR curve" if model_name is None else model_name
+    # Plot Precision-Recall curve
+    ax.plot(
+        recall,
+        precision,
+        lw=lw,
+        color=color,
+        label=(f"{model_name} (AP={avg_precision:.2f})"),
+        clip_on=False,
+        **kwargs,
+    )
+    # Fill area under the curve
+    ax.fill_between(recall, precision, alpha=alpha, color=color)
+
+    # Customize axes
+    ax.set_title(title)
+    ax.set_xlabel(xlabel)
+    ax.set_ylabel(ylabel)
+    ax.set_xlim([-0.01, 1.0])
+    ax.set_ylim([0.0, 1.0])
+    ax.grid(False)
+    ax.legend(loc=legend_loc)
+    return ax
+
+# * usage: ml2ls.plot_pr_curve()
+# for md_name in flatten(validation_results["pr_curve"].keys()):
+#     ml2ls.plot_pr_curve(
+#         recall=validation_results["pr_curve"][md_name]["recall"],
+#         precision=validation_results["pr_curve"][md_name]["precision"],
+#         avg_precision=validation_results["pr_curve"][md_name]["avg_precision"],
+#         model_name=md_name,
+#         lw=2,
+#         alpha=0.1,
+#         color="r",
+#     )
+
+def plot_pr_binary(
+    recall=None,
+    precision=None,
+    avg_precision=None,
+    model_name=None,
+    lw=2,
+    figsize=[5, 5],
+    title="Precision-Recall Curve",
+    xlabel="Recall",
+    ylabel="Precision",
+    alpha=0.1,
+    color="#FF8F00",
+    legend_loc="lower left",
+    ax=None,
+    show_avg_precision=False,
+    **kwargs,
+    ):
+    from scipy.interpolate import interp1d
+    if ax is None:
+        fig, ax = plt.subplots(figsize=figsize)
+    model_name = "Binary PR Curve" if model_name is None else model_name
+
+    #* use sklearn bulitin function 'PrecisionRecallDisplay'?
+    # from sklearn.metrics import PrecisionRecallDisplay
+    # disp = PrecisionRecallDisplay(precision=precision, 
+    #                               recall=recall, 
+    #                               average_precision=avg_precision,**kwargs)
+    # disp.plot(ax=ax, name=model_name, color=color)
+    
+    # Plot Precision-Recall curve
+    ax.plot(
+        recall,
+        precision,
+        lw=lw,
+        color=color,
+        label=(f"{model_name} (AP={avg_precision:.2f})"),
+        clip_on=False,
+        **kwargs,
+    )
+
+    # Fill area under the curve
+    ax.fill_between(recall, precision, alpha=alpha, color=color)
+    # Add F1 score iso-contours
+    f_scores = np.linspace(0.2, 0.8, num=4)
+    # for f_score in f_scores:
+    #     x = np.linspace(0.01, 1)
+    #     y = f_score * x / (2 * x - f_score)
+    #     plt.plot(x[y >= 0], y[y >= 0], color="gray", alpha=1)
+    #     plt.annotate(f"$f_1={f_score:0.1f}$", xy=(0.8, y[45] + 0.02))
+
+    pr_boundary = interp1d(recall, precision, kind="linear", fill_value="extrapolate")
+    for f_score in f_scores:
+        x_vals = np.linspace(0.01, 1, 10000)
+        y_vals = f_score * x_vals / (2 * x_vals - f_score)
+        y_vals_clipped = np.minimum(y_vals, pr_boundary(x_vals))
+        y_vals_clipped = np.clip(y_vals_clipped, 1e-3, None)  # Prevent going to zero
+        valid =  y_vals_clipped < pr_boundary(x_vals)
+        valid_ = y_vals_clipped > 1e-3 
+        valid = valid&valid_ 
+        x_vals = x_vals[valid] 
+        y_vals_clipped = y_vals_clipped[valid]
+        if len(x_vals) > 0:  # Ensure annotation is placed only if line segment exists
+            ax.plot(x_vals, y_vals_clipped, color="gray", alpha=1)
+            plt.annotate(f"$f_1={f_score:0.1f}$", xy=(0.8, y_vals_clipped[-int(len(y_vals_clipped)*0.35)] + 0.02))
+
+
+    # # Plot the average precision line
+    if show_avg_precision:
+        plt.axhline(
+            y=avg_precision,
+            color="red",
+            ls="--",
+            lw=lw,
+            label=f"Avg. precision={avg_precision:.2f}",
+        )
+    # Customize axes
+    ax.set_title(title)
+    ax.set_xlabel(xlabel)
+    ax.set_ylabel(ylabel)
+    ax.set_xlim([-0.01, 1.0])
+    ax.set_ylim([0.0, 1.0])
+    ax.grid(False)
+    ax.legend(loc=legend_loc)
+    return ax
+    
+def plot_cm(
+    cm,
+    labels_name=None,
+    thresh=0.8,
+    axis_labels=None,
+    cmap="Reds",
+    normalize=True,
+    xlabel="Predicted Label",
+    ylabel="Actual Label",
+    fontsize=12,
+    figsize=[5, 5],
+    ax=None,
+):
+    if ax is None:
+        fig, ax = plt.subplots(figsize=figsize)
+
+    cm_normalized = np.round(
+        cm.astype("float") / cm.sum(axis=1)[:, np.newaxis] * 100, 2
+    )
+    cm_value = cm_normalized if normalize else cm.astype("int")
+    # Plot the heatmap
+    cax = ax.imshow(cm_normalized, interpolation="nearest", cmap=cmap)
+    plt.colorbar(cax, ax=ax, fraction=0.046, pad=0.04)
+    cax.set_clim(0, 100)
+
+    # Define tick labels based on provided labels
+    num_local = np.arange(len(labels_name)) if labels_name is not None else range(2)
+    if axis_labels is None:
+        axis_labels = labels_name if labels_name is not None else ["No", "Yes"]
+    ax.set_xticks(num_local)
+    ax.set_xticklabels(axis_labels)
+    ax.set_yticks(num_local)
+    ax.set_yticklabels(axis_labels)
+    ax.set_ylabel(ylabel)
+    ax.set_xlabel(xlabel)
+
+    # Add TN, FP, FN, TP annotations specifically for binary classification (2x2 matrix)
+    if labels_name is None or len(labels_name) == 2:
+        # True Negative (TN), False Positive (FP), False Negative (FN), and True Positive (TP)
+        #                 Predicted
+        #                0   |   1
+        #             ----------------
+        #         0 |   TN   |  FP
+        # Actual      ----------------
+        #         1 |   FN   |  TP
+        tn_label = "TN"
+        fp_label = "FP"
+        fn_label = "FN"
+        tp_label = "TP"
+
+        # Adjust positions slightly for TN, FP, FN, TP labels
+        ax.text(
+            0,
+            0,
+            (
+                f"{tn_label}:{cm_normalized[0, 0]:.2f}%"
+                if normalize
+                else f"{tn_label}:{cm_value[0, 0]}"
+            ),
+            ha="center",
+            va="center",
+            color="white" if cm_normalized[0, 0] > thresh * 100 else "black",
+            fontsize=fontsize,
+        )
+        ax.text(
+            1,
+            0,
+            (
+                f"{fp_label}:{cm_normalized[0, 1]:.2f}%"
+                if normalize
+                else f"{fp_label}:{cm_value[0, 1]}"
+            ),
+            ha="center",
+            va="center",
+            color="white" if cm_normalized[0, 1] > thresh * 100 else "black",
+            fontsize=fontsize,
+        )
+        ax.text(
+            0,
+            1,
+            (
+                f"{fn_label}:{cm_normalized[1, 0]:.2f}%"
+                if normalize
+                else f"{fn_label}:{cm_value[1, 0]}"
+            ),
+            ha="center",
+            va="center",
+            color="white" if cm_normalized[1, 0] > thresh * 100 else "black",
+            fontsize=fontsize,
+        )
+        ax.text(
+            1,
+            1,
+            (
+                f"{tp_label}:{cm_normalized[1, 1]:.2f}%"
+                if normalize
+                else f"{tp_label}:{cm_value[1, 1]}"
+            ),
+            ha="center",
+            va="center",
+            color="white" if cm_normalized[1, 1] > thresh * 100 else "black",
+            fontsize=fontsize,
+        )
+    else:
+        # Annotate cells with normalized percentage values
+        for i in range(len(labels_name)):
+            for j in range(len(labels_name)):
+                val = cm_normalized[i, j]
+                color = "white" if val > thresh * 100 else "black"
+                ax.text(
+                    j,
+                    i,
+                    f"{val:.2f}%",
+                    ha="center",
+                    va="center",
+                    color=color,
+                    fontsize=fontsize,
+                )
+
+    plot.figsets(ax=ax, boxloc="none")
+    return ax
+
+
+def rank_models(
+    cv_test_scores,
+    rm_outlier=False,
+    metric_weights=None,
+    plot_=True,
+):
+    """
+    Selects the best model based on a multi-metric scoring approach, with outlier handling, optional visualization,
+    and additional performance metrics.
+
+    Parameters:
+    - cv_test_scores (pd.DataFrame): DataFrame with cross-validation results across multiple metrics.
+                                     Assumes columns are 'Classifier', 'accuracy', 'precision', 'recall', 'f1', 'roc_auc'.
+    - metric_weights (dict): Dictionary specifying weights for each metric (e.g., {'accuracy': 0.2, 'precision': 0.3, ...}).
+                             If None, default weights are applied equally across available metrics.
+                a. equal_weights(standard approch): 所有的metrics同等重要
+                    e.g., {"accuracy": 0.2, "precision": 0.2, "recall": 0.2, "f1": 0.2, "roc_auc": 0.2}
+                b. accuracy_focosed:  classification correctness (e.g., in balanced datasets), accuracy might be weighted more heavily.
+                    e.g., {"accuracy": 0.4, "precision": 0.2, "recall": 0.2, "f1": 0.1, "roc_auc": 0.1}
+                c. Precision and Recall Emphasis: In cases where false positives and false negatives are particularly important (such as
+                    in medical applications or fraud detection), precision and recall may be weighted more heavily.
+                    e.g., {"accuracy": 0.2, "precision": 0.3, "recall": 0.3, "f1": 0.1, "roc_auc": 0.1}
+                d. F1-Focused: When balance between precision and recall is crucial (e.g., in imbalanced datasets)
+                    e.g., {"accuracy": 0.2, "precision": 0.2, "recall": 0.2, "f1": 0.3, "roc_auc": 0.1}
+                e. ROC-AUC Emphasis: In some cases, ROC AUC may be prioritized, particularly in classification tasks where class imbalance
+                    is present, as ROC AUC accounts for the model's performance across all classification thresholds.
+                    e.g., {"accuracy": 0.1, "precision": 0.2, "recall": 0.2, "f1": 0.3, "roc_auc": 0.3}
+
+    - normalize (bool): Whether to normalize scores of each metric to range [0, 1].
+    - visualize (bool): If True, generates visualizations (e.g., bar plot, radar chart).
+    - outlier_threshold (float): The threshold to detect outliers using the IQR method. Default is 1.5.
+    - cv_folds (int): The number of cross-validation folds used.
+
+    Returns:
+    - best_model (str): Name of the best model based on the combined metric scores.
+    - scored_df (pd.DataFrame): DataFrame with an added 'combined_score' column used for model selection.
+    - visualizations (dict): A dictionary containing visualizations if `visualize=True`.
+    """
+    from sklearn.preprocessing import MinMaxScaler
+    import seaborn as sns
+    import matplotlib.pyplot as plt
+    from py2ls import plot
+
+    # Check for missing metrics and set default weights if not provided
+    available_metrics = cv_test_scores.columns[1:]  # Exclude 'Classifier' column
+    if metric_weights is None:
+        metric_weights = {
+            metric: 1 / len(available_metrics) for metric in available_metrics
+        }  # Equal weight if not specified
+    elif metric_weights == "a":
+        metric_weights = {
+            "accuracy": 0.2,
+            "precision": 0.2,
+            "recall": 0.2,
+            "f1": 0.2,
+            "roc_auc": 0.2,
+        }
+    elif metric_weights == "b":
+        metric_weights = {
+            "accuracy": 0.4,
+            "precision": 0.2,
+            "recall": 0.2,
+            "f1": 0.1,
+            "roc_auc": 0.1,
+        }
+    elif metric_weights == "c":
+        metric_weights = {
+            "accuracy": 0.2,
+            "precision": 0.3,
+            "recall": 0.3,
+            "f1": 0.1,
+            "roc_auc": 0.1,
+        }
+    elif metric_weights == "d":
+        metric_weights = {
+            "accuracy": 0.2,
+            "precision": 0.2,
+            "recall": 0.2,
+            "f1": 0.3,
+            "roc_auc": 0.1,
+        }
+    elif metric_weights == "e":
+        metric_weights = {
+            "accuracy": 0.1,
+            "precision": 0.2,
+            "recall": 0.2,
+            "f1": 0.3,
+            "roc_auc": 0.3,
+        }
+    else:
+        metric_weights = {
+            metric: 1 / len(available_metrics) for metric in available_metrics
+        }
+
+    # Normalize weights if they don’t sum to 1
+    total_weight = sum(metric_weights.values())
+    metric_weights = {
+        metric: weight / total_weight for metric, weight in metric_weights.items()
+    }
+    if rm_outlier:
+        cv_test_scores_ = ips.df_outlier(cv_test_scores)
+    else:
+        cv_test_scores_ = cv_test_scores
+
+    # Normalize the scores of metrics if normalize is True
+    scaler = MinMaxScaler()
+    normalized_scores = pd.DataFrame(
+        scaler.fit_transform(cv_test_scores_[available_metrics]),
+        columns=available_metrics,
+    )
+    cv_test_scores_ = pd.concat(
+        [cv_test_scores_[["Classifier"]], normalized_scores], axis=1
+    )
+
+    # Calculate weighted scores for each model
+    cv_test_scores_["combined_score"] = sum(
+        cv_test_scores_[metric] * weight for metric, weight in metric_weights.items()
+    )
+    top_models = cv_test_scores_.sort_values(by="combined_score", ascending=False)
+    cv_test_scores = cv_test_scores.loc[top_models.index]
+    top_models.reset_index(drop=True, inplace=True)
+    cv_test_scores.reset_index(drop=True, inplace=True)
+
+    if plot_:
+
+        def generate_bar_plot(ax, cv_test_scores):
+            ax = plot.plotxy(
+                y="Classifier", x="combined_score", data=cv_test_scores, kind="bar"
+            )
+            plt.title("Classifier Performance")
+            plt.tight_layout()
+            return plt
+
+        nexttile = plot.subplot(2, 2, figsize=[10, 7])
+        generate_bar_plot(nexttile(), top_models.dropna())
+        plot.radar(
+            ax=nexttile(projection="polar"),
+            data=cv_test_scores.set_index("Classifier"),
+            ylim=[0.5, 1],
+            color=plot.get_color(10),
+            alpha=0.05,
+            circular=1,
+        )
+    return cv_test_scores
+
+
+# # Example Usage:
+# metric_weights = {
+#     "accuracy": 0.2,
+#     "precision": 0.3,
+#     "recall": 0.2,
+#     "f1": 0.2,
+#     "roc_auc": 0.1,
+# }
+# cv_test_scores = res["cv_test_scores"].copy()
+# best_model = rank_models(
+#     cv_test_scores, metric_weights=metric_weights, normalize=True, plot_=True
+# )
+
+# figsave("classifier_performance.pdf")
+
+
+def predict(
+    x_train: pd.DataFrame,
+    y_train: pd.Series,
+    x_true: pd.DataFrame = None,
+    y_true: Optional[pd.Series] = None,
+    common_features: set = None,
+    purpose: str = "classification",  # 'classification' or 'regression'
+    cls: Optional[Dict[str, Any]] = None,
+    metrics: Optional[List[str]] = None,
+    random_state: int = 1,
+    smote: bool = False,
+    n_jobs: int = -1,
+    plot_: bool = True,
+    dir_save: str = "./",
+    test_size: float = 0.2,  # specific only when x_true is None
+    cv_folds: int = 5,  # more cv_folds 得更加稳定,auc可能更低
+    cv_level: str = "l",  # "s":'low',"m":'medium',"l":"high"
+    class_weight: str = "balanced",
+    verbose: bool = False,
+) -> pd.DataFrame:
+    """
+    第一种情况是内部拆分，第二种是直接预测，第三种是外部验证。
+    Usage:
+        (1). predict(x_train, y_train,...) 对 x_train 进行拆分训练/测试集，并在测试集上进行验证.
+            predict 函数会根据 test_size 参数，将 x_train 和 y_train 拆分出内部测试集。然后模型会在拆分出的训练集上进行训练，并在测试集上验证效果。
+        (2). predict(x_train, y_train, x_true,...)使用 x_train 和 y_train 训练并对 x_true 进行预测
+            由于传入了 x_true，函数会跳过 x_train 的拆分，直接使用全部的 x_train 和 y_train 进行训练。然后对 x_true 进行预测，但由于没有提供 y_true，
+            因此无法与真实值进行对比。
+        (3). predict(x_train, y_train, x_true, y_true,...)使用 x_train 和 y_train 训练，并验证 x_true 与真实标签 y_true.
+            predict 函数会在 x_train 和 y_train 上进行训练，并将 x_true 作为测试集。由于提供了 y_true，函数可以将预测结果与 y_true 进行对比，从而
+            计算验证指标，完成对 x_true 的真正验证。
+    trains and validates a variety of machine learning models for both classification and regression tasks.
+    It supports hyperparameter tuning with grid search and includes additional features like cross-validation,
+    feature scaling, and handling of class imbalance through SMOTE.
+
+    Parameters:
+        - x_train (pd.DataFrame):Training feature data, structured with each row as an observation and each column as a feature.
+        - y_train (pd.Series):Target variable for the training dataset.
+        - x_true (pd.DataFrame, optional):Test feature data. If not provided, the function splits x_train based on test_size.
+        - y_true (pd.Series, optional):Test target values. If not provided, y_train is split into training and testing sets.
+        - common_features (set, optional):Specifies a subset of features common across training and test data.
+        - purpose (str, default = "classification"):Defines whether the task is "classification" or "regression". Determines which
+            metrics and models are applied.
+        - cls (dict, optional):Dictionary to specify custom classifiers/regressors. Defaults to a set of common models if not provided.
+        - metrics (list, optional):List of evaluation metrics (like accuracy, F1 score) used for model evaluation.
+        - random_state (int, default = 1):Random seed to ensure reproducibility.
+        - smote (bool, default = False):Applies Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance if enabled.
+        - n_jobs (int, default = -1):Number of parallel jobs for computation. Set to -1 to use all available cores.
+        - plot_ (bool, default = True):If True, generates plots of the model evaluation metrics.
+        - test_size (float, default = 0.2):Test data proportion if x_true is not provided.
+        - cv_folds (int, default = 5):Number of cross-validation folds.
+        - cv_level (str, default = "l"):Sets the detail level of cross-validation. "s" for low, "m" for medium, and "l" for high.
+        - class_weight (str, default = "balanced"):Balances class weights in classification tasks.
+        - verbose (bool, default = False):If True, prints detailed output during model training.
+        - dir_save (str, default = "./"):Directory path to save plot outputs and results.
+
+    Key Steps in the Function:
+        Model Initialization: Depending on purpose, initializes either classification or regression models.
+        Feature Selection: Ensures training and test sets have matching feature columns.
+        SMOTE Application: Balances classes if smote is enabled and the task is classification.
+        Cross-Validation and Hyperparameter Tuning: Utilizes GridSearchCV for model tuning based on cv_level.
+        Evaluation and Plotting: Outputs evaluation metrics like AUC, confusion matrices, and optional plotting of performance metrics.
+    """
+    from tqdm import tqdm
+    from sklearn.ensemble import (
+        RandomForestClassifier,
+        RandomForestRegressor,
+        ExtraTreesClassifier,
+        ExtraTreesRegressor,
+        BaggingClassifier,
+        BaggingRegressor,
+        AdaBoostClassifier,
+        AdaBoostRegressor,
+    )
+    from sklearn.svm import SVC, SVR
+    from sklearn.tree import DecisionTreeRegressor
+    from sklearn.linear_model import (
+        LogisticRegression,
+        ElasticNet,
+        ElasticNetCV,
+        LinearRegression,
+        Lasso,
+        RidgeClassifierCV,
+        Perceptron,
+        SGDClassifier,
+    )
+    from sklearn.neighbors import KNeighborsClassifier, KNeighborsRegressor
+    from sklearn.naive_bayes import GaussianNB, BernoulliNB
+    from sklearn.ensemble import GradientBoostingClassifier, GradientBoostingRegressor
+    import xgboost as xgb
+    import lightgbm as lgb
+    import catboost as cb
+    from sklearn.neural_network import MLPClassifier, MLPRegressor
+    from sklearn.model_selection import GridSearchCV, StratifiedKFold, KFold
+    from sklearn.discriminant_analysis import (
+        LinearDiscriminantAnalysis,
+        QuadraticDiscriminantAnalysis,
+    )
+    from sklearn.preprocessing import PolynomialFeatures
+
+    # 拼写检查
+    purpose = ips.strcmp(purpose, ["classification", "regression"])[0]
+    print(f"{purpose} processing...")
+    # Default models or regressors if not provided
+    if purpose == "classification":
+        model_ = {
+            "Random Forest": RandomForestClassifier(
+                random_state=random_state, class_weight=class_weight
+            ),
+            # SVC (Support Vector Classification)
+            "SVM": SVC(
+                kernel="rbf",
+                probability=True,
+                class_weight=class_weight,
+                random_state=random_state,
+            ),
+            # fit the best model without enforcing sparsity, which means it does not directly perform feature selection.
+            "Logistic Regression": LogisticRegression(
+                class_weight=class_weight, random_state=random_state
+            ),
+            # Logistic Regression with L1 Regularization (Lasso)
+            "Lasso Logistic Regression": LogisticRegression(
+                penalty="l1", solver="saga", random_state=random_state
+            ),
+            "Gradient Boosting": GradientBoostingClassifier(random_state=random_state),
+            "XGBoost": xgb.XGBClassifier(
+                eval_metric="logloss",
+                random_state=random_state,
+            ),
+            "KNN": KNeighborsClassifier(n_neighbors=5),
+            "Naive Bayes": GaussianNB(),
+            "Linear Discriminant Analysis": LinearDiscriminantAnalysis(),
+            "AdaBoost": AdaBoostClassifier(
+                algorithm="SAMME", random_state=random_state
+            ),
+            # "LightGBM": lgb.LGBMClassifier(random_state=random_state, class_weight=class_weight),
+            "CatBoost": cb.CatBoostClassifier(verbose=0, random_state=random_state),
+            "Extra Trees": ExtraTreesClassifier(
+                random_state=random_state, class_weight=class_weight
+            ),
+            "Bagging": BaggingClassifier(random_state=random_state),
+            "Neural Network": MLPClassifier(max_iter=500, random_state=random_state),
+            "DecisionTree": DecisionTreeClassifier(),
+            "Quadratic Discriminant Analysis": QuadraticDiscriminantAnalysis(),
+            "Ridge": RidgeClassifierCV(
+                class_weight=class_weight, store_cv_results=True
+            ),
+            "Perceptron": Perceptron(random_state=random_state),
+            "Bernoulli Naive Bayes": BernoulliNB(),
+            "SGDClassifier": SGDClassifier(random_state=random_state),
+        }
+    elif purpose == "regression":
+        model_ = {
+            "Random Forest": RandomForestRegressor(random_state=random_state),
+            "SVM": SVR(),  # SVR (Support Vector Regression)
+            # "Lasso": Lasso(random_state=random_state), # 它和LassoCV相同(必须要提供alpha参数),
+            "LassoCV": LassoCV(
+                cv=cv_folds, random_state=random_state
+            ),  # LassoCV自动找出最适alpha,优于Lasso
+            "Gradient Boosting": GradientBoostingRegressor(random_state=random_state),
+            "XGBoost": xgb.XGBRegressor(eval_metric="rmse", random_state=random_state),
+            "Linear Regression": LinearRegression(),
+            "Lasso": Lasso(random_state=random_state),
+            "AdaBoost": AdaBoostRegressor(random_state=random_state),
+            # "LightGBM": lgb.LGBMRegressor(random_state=random_state),
+            "CatBoost": cb.CatBoostRegressor(verbose=0, random_state=random_state),
+            "Extra Trees": ExtraTreesRegressor(random_state=random_state),
+            "Bagging": BaggingRegressor(random_state=random_state),
+            "Neural Network": MLPRegressor(max_iter=500, random_state=random_state),
+            "ElasticNet": ElasticNet(random_state=random_state),
+            "Ridge": Ridge(),
+            "KNN": KNeighborsRegressor(),
+        }
+    # indicate cls:
+    if ips.run_once_within(30):  # 10 min
+        print(f"supported models: {list(model_.keys())}")
+    if cls is None:
+        models = model_
+    else:
+        if not isinstance(cls, list):
+            cls = [cls]
+        models = {}
+        for cls_ in cls:
+            cls_ = ips.strcmp(cls_, list(model_.keys()))[0]
+            models[cls_] = model_[cls_]
+    if "LightGBM" in models:
+        x_train = ips.df_special_characters_cleaner(x_train)
+        x_true = (
+            ips.df_special_characters_cleaner(x_true) if x_true is not None else None
+        )
+
+    if isinstance(y_train, str) and y_train in x_train.columns:
+        y_train_col_name = y_train
+        y_train = x_train[y_train]
+        # y_train = ips.df_encoder(pd.DataFrame(y_train), method="dummy")
+        x_train = x_train.drop(y_train_col_name, axis=1)
+    # else:
+    #     y_train = ips.df_encoder(pd.DataFrame(y_train), method="dummy").values.ravel()
+    y_train=pd.DataFrame(y_train)
+    y_train_=ips.df_encoder(y_train, method="dummy")
+    is_binary = False if y_train_.shape[1] >1 else True
+    print(is_binary)
+    if is_binary:
+        y_train = ips.df_encoder(pd.DataFrame(y_train), method="dummy").values.ravel() 
+    if x_true is None:
+        x_train, x_true, y_train, y_true = train_test_split(
+            x_train,
+            y_train,
+            test_size=test_size,
+            random_state=random_state,
+            stratify=y_train if purpose == "classification" else None,
+        )
+        if isinstance(y_train, str) and y_train in x_train.columns:
+            y_train_col_name = y_train
+            y_train = x_train[y_train]
+            y_train = ips.df_encoder(pd.DataFrame(y_train), method="dummy")
+            x_train = x_train.drop(y_train_col_name, axis=1)
+        else:
+            y_train = ips.df_encoder(
+                pd.DataFrame(y_train), method="dummy"
+            ).values.ravel() 
+        
+    if y_true is not None:
+        if isinstance(y_true, str) and y_true in x_true.columns:
+            y_true_col_name = y_true
+            y_true = x_true[y_true]
+            # y_true = ips.df_encoder(pd.DataFrame(y_true), method="dummy")
+            y_true =  pd.DataFrame(y_true)
+            x_true = x_true.drop(y_true_col_name, axis=1)
+        # else:
+        #     y_true = ips.df_encoder(pd.DataFrame(y_true), method="dummy").values.ravel()
+
+    # to convert the 2D to 1D: 2D column-vector format (like [[1], [0], [1], ...]) instead of a 1D array ([1, 0, 1, ...]
+
+    # y_train=y_train.values.ravel() if y_train is not None else None
+    # y_true=y_true.values.ravel() if y_true is not None else None
+    y_train = (
+        y_train.ravel() if isinstance(y_train, np.ndarray) else y_train.values.ravel()
+    )
+    print(len(y_train),len(y_true))
+    y_true = y_true.ravel() if isinstance(y_true, np.ndarray) else y_true.values.ravel()
+    print(len(y_train),len(y_true))
+    # Ensure common features are selected
+    if common_features is not None:
+        x_train, x_true = x_train[common_features], x_true[common_features]
+    else:
+        share_col_names = ips.shared(x_train.columns, x_true.columns, verbose=verbose)
+        x_train, x_true = x_train[share_col_names], x_true[share_col_names]
+
+    x_train, x_true = ips.df_scaler(x_train), ips.df_scaler(x_true)
+    x_train, x_true = ips.df_encoder(x_train, method="dummy"), ips.df_encoder(x_true, method="dummy") 
+    # Handle class imbalance using SMOTE (only for classification)
+    if (
+        smote
+        and purpose == "classification"
+        and y_train.value_counts(normalize=True).max() < 0.8
+    ):
+        from imblearn.over_sampling import SMOTE
+
+        smote_sampler = SMOTE(random_state=random_state)
+        x_train, y_train = smote_sampler.fit_resample(x_train, y_train)
+
+    # Hyperparameter grids for tuning
+    if cv_level in ["low", "simple", "s", "l"]:
+        param_grids = {
+            "Random Forest": (
+                {
+                    "n_estimators": [100],  # One basic option
+                    "max_depth": [None, 10],
+                    "min_samples_split": [2],
+                    "min_samples_leaf": [1],
+                    "class_weight": [None],
+                }
+                if purpose == "classification"
+                else {
+                    "n_estimators": [100],  # One basic option
+                    "max_depth": [None, 10],
+                    "min_samples_split": [2],
+                    "min_samples_leaf": [1],
+                    "max_features": [None],
+                    "bootstrap": [True],  # Only one option for simplicity
+                }
+            ),
+            "SVM": {
+                "C": [1],
+                "gamma": ["scale"],
+                "kernel": ["rbf"],
+            },
+            "Lasso": {
+                "alpha": [0.1],
+            },
+            "LassoCV": {
+                "alphas": [[0.1]],
+            },
+            "Logistic Regression": {
+                "C": [1],
+                "solver": ["lbfgs"],
+                "penalty": ["l2"],
+                "max_iter": [500],
+            },
+            "Gradient Boosting": {
+                "n_estimators": [100],
+                "learning_rate": [0.1],
+                "max_depth": [3],
+                "min_samples_split": [2],
+                "subsample": [0.8],
+            },
+            "XGBoost": {
+                "n_estimators": [100],
+                "max_depth": [3],
+                "learning_rate": [0.1],
+                "subsample": [0.8],
+                "colsample_bytree": [0.8],
+            },
+            "KNN": (
+                {
+                    "n_neighbors": [3],
+                    "weights": ["uniform"],
+                    "algorithm": ["auto"],
+                    "p": [2],
+                }
+                if purpose == "classification"
+                else {
+                    "n_neighbors": [3],
+                    "weights": ["uniform"],
+                    "metric": ["euclidean"],
+                    "leaf_size": [30],
+                    "p": [2],
+                }
+            ),
+            "Naive Bayes": {
+                "var_smoothing": [1e-9],
+            },
+            "SVR": {
+                "C": [1],
+                "gamma": ["scale"],
+                "kernel": ["rbf"],
+            },
+            "Linear Regression": {
+                "fit_intercept": [True],
+            },
+            "Extra Trees": {
+                "n_estimators": [100],
+                "max_depth": [None, 10],
+                "min_samples_split": [2],
+                "min_samples_leaf": [1],
+            },
+            "CatBoost": {
+                "iterations": [100],
+                "learning_rate": [0.1],
+                "depth": [3],
+                "l2_leaf_reg": [1],
+            },
+            "LightGBM": {
+                "n_estimators": [100],
+                "num_leaves": [31],
+                "max_depth": [10],
+                "min_data_in_leaf": [20],
+                "min_gain_to_split": [0.01],
+                "scale_pos_weight": [10],
+            },
+            "Bagging": {
+                "n_estimators": [50],
+                "max_samples": [0.7],
+                "max_features": [0.7],
+            },
+            "Neural Network": {
+                "hidden_layer_sizes": [(50,)],
+                "activation": ["relu"],
+                "solver": ["adam"],
+                "alpha": [0.0001],
+            },
+            "Decision Tree": {
+                "max_depth": [None, 10],
+                "min_samples_split": [2],
+                "min_samples_leaf": [1],
+                "criterion": ["gini"],
+            },
+            "AdaBoost": {
+                "n_estimators": [50],
+                "learning_rate": [0.5],
+            },
+            "Linear Discriminant Analysis": {
+                "solver": ["svd"],
+                "shrinkage": [None],
+            },
+            "Quadratic Discriminant Analysis": {
+                "reg_param": [0.0],
+                "priors": [None],
+                "tol": [1e-4],
+            },
+            "Ridge": (
+                {"class_weight": [None, "balanced"]}
+                if purpose == "classification"
+                else {
+                    "alpha": [0.1, 1, 10],
+                }
+            ),
+            "Perceptron": {
+                "alpha": [1e-3],
+                "penalty": ["l2"],
+                "max_iter": [1000],
+                "eta0": [1.0],
+            },
+            "Bernoulli Naive Bayes": {
+                "alpha": [0.1, 1, 10],
+                "binarize": [0.0],
+                "fit_prior": [True],
+            },
+            "SGDClassifier": {
+                "eta0": [0.01],
+                "loss": ["hinge"],
+                "penalty": ["l2"],
+                "alpha": [1e-3],
+                "max_iter": [1000],
+                "tol": [1e-3],
+                "random_state": [random_state],
+                "learning_rate": ["constant"],
+            },
+        }
+    elif cv_level in ["high", "advanced", "h"]:
+        param_grids = {
+            "Random Forest": (
+                {
+                    "n_estimators": [100, 200, 500, 700, 1000],
+                    "max_depth": [None, 3, 5, 10, 15, 20, 30],
+                    "min_samples_split": [2, 5, 10, 20],
+                    "min_samples_leaf": [1, 2, 4],
+                    "class_weight": (
+                        [None, "balanced"] if purpose == "classification" else {}
+                    ),
+                }
+                if purpose == "classification"
+                else {
+                    "n_estimators": [100, 200, 500, 700, 1000],
+                    "max_depth": [None, 3, 5, 10, 15, 20, 30],
+                    "min_samples_split": [2, 5, 10, 20],
+                    "min_samples_leaf": [1, 2, 4],
+                    "max_features": [
+                        "auto",
+                        "sqrt",
+                        "log2",
+                    ],  # Number of features to consider when looking for the best split
+                    "bootstrap": [
+                        True,
+                        False,
+                    ],  # Whether bootstrap samples are used when building trees
+                }
+            ),
+            "SVM": {
+                "C": [0.001, 0.01, 0.1, 1, 10, 100, 1000],
+                "gamma": ["scale", "auto", 0.001, 0.01, 0.1],
+                "kernel": ["linear", "rbf", "poly"],
+            },
+            "Logistic Regression": {
+                "C": [0.001, 0.01, 0.1, 1, 10, 100, 1000],
+                "solver": ["liblinear", "saga", "newton-cg", "lbfgs"],
+                "penalty": ["l1", "l2", "elasticnet"],
+                "max_iter": [100, 200, 300, 500],
+            },
+            "Lasso": {
+                "alpha": [0.0001, 0.001, 0.01, 0.1, 1.0, 10.0, 100.0],
+                "max_iter": [500, 1000, 2000, 5000],
+                "tol": [1e-4, 1e-5, 1e-6],
+                "selection": ["cyclic", "random"],
+            },
+            "LassoCV": {
+                "alphas": [[0.0001, 0.001, 0.01, 0.1, 1.0, 10.0, 100.0]],
+                "max_iter": [500, 1000, 2000, 5000],
+                "cv": [3, 5, 10],
+                "tol": [1e-4, 1e-5, 1e-6],
+            },
+            "Gradient Boosting": {
+                "n_estimators": [100, 200, 300, 400, 500, 700, 1000],
+                "learning_rate": [0.001, 0.01, 0.1, 0.2, 0.3, 0.5],
+                "max_depth": [3, 5, 7, 9, 15],
+                "min_samples_split": [2, 5, 10, 20],
+                "subsample": [0.8, 1.0],
+            },
+            "XGBoost": {
+                "n_estimators": [100, 200, 500, 700],
+                "max_depth": [3, 5, 7, 10],
+                "learning_rate": [0.01, 0.1, 0.2, 0.3],
+                "subsample": [0.8, 1.0],
+                "colsample_bytree": [0.8, 0.9, 1.0],
+            },
+            "KNN": (
+                {
+                    "n_neighbors": [1, 3, 5, 10, 15, 20],
+                    "weights": ["uniform", "distance"],
+                    "algorithm": ["auto", "ball_tree", "kd_tree", "brute"],
+                    "p": [1, 2],  # 1 for Manhattan, 2 for Euclidean distance
+                }
+                if purpose == "classification"
+                else {
+                    "n_neighbors": [3, 5, 7, 9, 11],  # Number of neighbors
+                    "weights": [
+                        "uniform",
+                        "distance",
+                    ],  # Weight function used in prediction
+                    "metric": [
+                        "euclidean",
+                        "manhattan",
+                        "minkowski",
+                    ],  # Distance metric
+                    "leaf_size": [
+                        20,
+                        30,
+                        40,
+                        50,
+                    ],  # Leaf size for KDTree or BallTree algorithms
+                    "p": [
+                        1,
+                        2,
+                    ],  # Power parameter for the Minkowski metric (1 = Manhattan, 2 = Euclidean)
+                }
+            ),
+            "Naive Bayes": {
+                "var_smoothing": [1e-10, 1e-9, 1e-8, 1e-7],
+            },
+            "AdaBoost": {
+                "n_estimators": [50, 100, 200, 300, 500],
+                "learning_rate": [0.001, 0.01, 0.1, 0.5, 1.0],
+            },
+            "SVR": {
+                "C": [0.01, 0.1, 1, 10, 100, 1000],
+                "gamma": [0.001, 0.01, 0.1, "scale", "auto"],
+                "kernel": ["linear", "rbf", "poly"],
+            },
+            "Linear Regression": {
+                "fit_intercept": [True, False],
+            },
+            "Lasso": {
+                "alpha": [0.001, 0.01, 0.1, 1.0, 10.0, 100.0],
+                "max_iter": [1000, 2000],  # Higher iteration limit for fine-tuning
+            },
+            "Extra Trees": {
+                "n_estimators": [100, 200, 500, 700, 1000],
+                "max_depth": [None, 5, 10, 15, 20, 30],
+                "min_samples_split": [2, 5, 10, 20],
+                "min_samples_leaf": [1, 2, 4],
+            },
+            "CatBoost": {
+                "iterations": [100, 200, 500],
+                "learning_rate": [0.001, 0.01, 0.1, 0.2],
+                "depth": [3, 5, 7, 10],
+                "l2_leaf_reg": [1, 3, 5, 7, 10],
+                "border_count": [32, 64, 128],
+            },
+            "LightGBM": {
+                "n_estimators": [100, 200, 500, 700, 1000],
+                "learning_rate": [0.001, 0.01, 0.1, 0.2],
+                "num_leaves": [31, 50, 100, 200],
+                "max_depth": [-1, 5, 10, 20, 30],
+                "min_child_samples": [5, 10, 20],
+                "subsample": [0.8, 1.0],
+                "colsample_bytree": [0.8, 0.9, 1.0],
+            },
+            "Neural Network": {
+                "hidden_layer_sizes": [(50,), (100,), (100, 50), (200, 100)],
+                "activation": ["relu", "tanh", "logistic"],
+                "solver": ["adam", "sgd", "lbfgs"],
+                "alpha": [0.0001, 0.001, 0.01],
+                "learning_rate": ["constant", "adaptive"],
+            },
+            "Decision Tree": {
+                "max_depth": [None, 5, 10, 20, 30],
+                "min_samples_split": [2, 5, 10, 20],
+                "min_samples_leaf": [1, 2, 5, 10],
+                "criterion": ["gini", "entropy"],
+                "splitter": ["best", "random"],
+            },
+            "Linear Discriminant Analysis": {
+                "solver": ["svd", "lsqr", "eigen"],
+                "shrinkage": [
+                    None,
+                    "auto",
+                    0.1,
+                    0.5,
+                    1.0,
+                ],  # shrinkage levels for 'lsqr' and 'eigen'
+            },
+            "Ridge": (
+                {"class_weight": [None, "balanced"]}
+                if purpose == "classification"
+                else {
+                    "alpha": [0.1, 1, 10, 100, 1000],
+                    "solver": ["auto", "svd", "cholesky", "lsqr", "lbfgs"],
+                    "fit_intercept": [
+                        True,
+                        False,
+                    ],  # Whether to calculate the intercept
+                    "normalize": [
+                        True,
+                        False,
+                    ],  # If True, the regressors X will be normalized
+                }
+            ),
+        }
+    else:  # median level
+        param_grids = {
+            "Random Forest": (
+                {
+                    "n_estimators": [100, 200, 500],
+                    "max_depth": [None, 10, 20, 30],
+                    "min_samples_split": [2, 5, 10],
+                    "min_samples_leaf": [1, 2, 4],
+                    "class_weight": [None, "balanced"],
+                }
+                if purpose == "classification"
+                else {
+                    "n_estimators": [100, 200, 500],
+                    "max_depth": [None, 10, 20, 30],
+                    "min_samples_split": [2, 5, 10],
+                    "min_samples_leaf": [1, 2, 4],
+                    "max_features": [
+                        "auto",
+                        "sqrt",
+                        "log2",
+                    ],  # Number of features to consider when looking for the best split
+                    "bootstrap": [
+                        True,
+                        False,
+                    ],  # Whether bootstrap samples are used when building trees
+                }
+            ),
+            "SVM": {
+                "C": [0.1, 1, 10, 100],  # Regularization strength
+                "gamma": ["scale", "auto"],  # Common gamma values
+                "kernel": ["rbf", "linear", "poly"],
+            },
+            "Logistic Regression": {
+                "C": [0.1, 1, 10, 100],  # Regularization strength
+                "solver": ["lbfgs", "liblinear", "saga"],  # Common solvers
+                "penalty": ["l2"],  # L2 penalty is most common
+                "max_iter": [
+                    500,
+                    1000,
+                    2000,
+                ],  # Increased max_iter for better convergence
+            },
+            "Lasso": {
+                "alpha": [0.001, 0.01, 0.1, 1.0, 10.0, 100.0],
+                "max_iter": [500, 1000, 2000],
+            },
+            "LassoCV": {
+                "alphas": [[0.001, 0.01, 0.1, 1.0, 10.0, 100.0]],
+                "max_iter": [500, 1000, 2000],
+            },
+            "Gradient Boosting": {
+                "n_estimators": [100, 200, 500],
+                "learning_rate": [0.01, 0.1, 0.2],
+                "max_depth": [3, 5, 7],
+                "min_samples_split": [2, 5, 10],
+                "subsample": [0.8, 1.0],
+            },
+            "XGBoost": {
+                "n_estimators": [100, 200, 500],
+                "max_depth": [3, 5, 7],
+                "learning_rate": [0.01, 0.1, 0.2],
+                "subsample": [0.8, 1.0],
+                "colsample_bytree": [0.8, 1.0],
+            },
+            "KNN": (
+                {
+                    "n_neighbors": [3, 5, 7, 10],
+                    "weights": ["uniform", "distance"],
+                    "algorithm": ["auto", "ball_tree", "kd_tree", "brute"],
+                    "p": [1, 2],
+                }
+                if purpose == "classification"
+                else {
+                    "n_neighbors": [3, 5, 7, 9, 11],  # Number of neighbors
+                    "weights": [
+                        "uniform",
+                        "distance",
+                    ],  # Weight function used in prediction
+                    "metric": [
+                        "euclidean",
+                        "manhattan",
+                        "minkowski",
+                    ],  # Distance metric
+                    "leaf_size": [
+                        20,
+                        30,
+                        40,
+                        50,
+                    ],  # Leaf size for KDTree or BallTree algorithms
+                    "p": [
+                        1,
+                        2,
+                    ],  # Power parameter for the Minkowski metric (1 = Manhattan, 2 = Euclidean)
+                }
+            ),
+            "Naive Bayes": {
+                "var_smoothing": [1e-9, 1e-8, 1e-7],
+            },
+            "SVR": {
+                "C": [0.1, 1, 10, 100],
+                "gamma": ["scale", "auto"],
+                "kernel": ["rbf", "linear"],
+            },
+            "Linear Regression": {
+                "fit_intercept": [True, False],
+            },
+            "Lasso": {
+                "alpha": [0.1, 1.0, 10.0],
+                "max_iter": [1000, 2000],  # Sufficient iterations for convergence
+            },
+            "Extra Trees": {
+                "n_estimators": [100, 200, 500],
+                "max_depth": [None, 10, 20, 30],
+                "min_samples_split": [2, 5, 10],
+                "min_samples_leaf": [1, 2, 4],
+            },
+            "CatBoost": {
+                "iterations": [100, 200],
+                "learning_rate": [0.01, 0.1],
+                "depth": [3, 6, 10],
+                "l2_leaf_reg": [1, 3, 5, 7],
+            },
+            "LightGBM": {
+                "n_estimators": [100, 200, 500],
+                "learning_rate": [0.01, 0.1],
+                "num_leaves": [31, 50, 100],
+                "max_depth": [-1, 10, 20],
+                "min_data_in_leaf": [20],  # Minimum samples in each leaf
+                "min_gain_to_split": [0.01],  # Minimum gain to allow a split
+                "scale_pos_weight": [10],  # Address class imbalance
+            },
+            "Bagging": {
+                "n_estimators": [10, 50, 100],
+                "max_samples": [0.5, 0.7, 1.0],
+                "max_features": [0.5, 0.7, 1.0],
+            },
+            "Neural Network": {
+                "hidden_layer_sizes": [(50,), (100,), (100, 50)],
+                "activation": ["relu", "tanh"],
+                "solver": ["adam", "sgd"],
+                "alpha": [0.0001, 0.001],
+            },
+            "Decision Tree": {
+                "max_depth": [None, 10, 20],
+                "min_samples_split": [2, 10],
+                "min_samples_leaf": [1, 4],
+                "criterion": ["gini", "entropy"],
+            },
+            "AdaBoost": {
+                "n_estimators": [50, 100],
+                "learning_rate": [0.5, 1.0],
+            },
+            "Linear Discriminant Analysis": {
+                "solver": ["svd", "lsqr", "eigen"],
+                "shrinkage": [None, "auto"],
+            },
+            "Quadratic Discriminant Analysis": {
+                "reg_param": [0.0, 0.1, 0.5, 1.0],  # Regularization parameter
+                "priors": [None, [0.5, 0.5], [0.3, 0.7]],  # Class priors
+                "tol": [
+                    1e-4,
+                    1e-3,
+                    1e-2,
+                ],  # Tolerance value for the convergence of the algorithm
+            },
+            "Perceptron": {
+                "alpha": [1e-4, 1e-3, 1e-2],  # Regularization parameter
+                "penalty": ["l2", "l1", "elasticnet"],  # Regularization penalty
+                "max_iter": [1000, 2000],  # Maximum number of iterations
+                "eta0": [1.0, 0.1],  # Learning rate for gradient descent
+                "tol": [1e-3, 1e-4, 1e-5],  # Tolerance for stopping criteria
+                "random_state": [random_state],  # Random state for reproducibility
+            },
+            "Bernoulli Naive Bayes": {
+                "alpha": [0.1, 1.0, 10.0],  # Additive (Laplace) smoothing parameter
+                "binarize": [
+                    0.0,
+                    0.5,
+                    1.0,
+                ],  # Threshold for binarizing the input features
+                "fit_prior": [
+                    True,
+                    False,
+                ],  # Whether to learn class prior probabilities
+            },
+            "SGDClassifier": {
+                "eta0": [0.01, 0.1, 1.0],
+                "loss": [
+                    "hinge",
+                    "log",
+                    "modified_huber",
+                    "squared_hinge",
+                    "perceptron",
+                ],  # Loss function
+                "penalty": ["l2", "l1", "elasticnet"],  # Regularization penalty
+                "alpha": [1e-4, 1e-3, 1e-2],  # Regularization strength
+                "l1_ratio": [0.15, 0.5, 0.85],  # L1 ratio for elasticnet penalty
+                "max_iter": [1000, 2000],  # Maximum number of iterations
+                "tol": [1e-3, 1e-4],  # Tolerance for stopping criteria
+                "random_state": [random_state],  # Random state for reproducibility
+                "learning_rate": [
+                    "constant",
+                    "optimal",
+                    "invscaling",
+                    "adaptive",
+                ],  # Learning rate schedule
+            },
+            "Ridge": (
+                {"class_weight": [None, "balanced"]}
+                if purpose == "classification"
+                else {
+                    "alpha": [0.1, 1, 10, 100],
+                    "solver": [
+                        "auto",
+                        "svd",
+                        "cholesky",
+                        "lsqr",
+                    ],  # Solver for optimization
+                }
+            ),
+        }
+
+    results = {}
+    # Use StratifiedKFold for classification and KFold for regression
+    cv = (
+        StratifiedKFold(n_splits=cv_folds, shuffle=True, random_state=random_state)
+        if purpose == "classification"
+        else KFold(n_splits=cv_folds, shuffle=True, random_state=random_state)
+    )
+
+    # Train and validate each model
+    for name, clf in tqdm(
+        models.items(),
+        desc="models",
+        colour="green",
+        bar_format="{l_bar}{bar} {n_fmt}/{total_fmt}",
+    ):
+        if verbose:
+            print(f"\nTraining and validating {name}:")
+
+        # Grid search with KFold or StratifiedKFold
+        gs = GridSearchCV(
+            clf,
+            param_grid=param_grids.get(name, {}),
+            scoring=(
+                "roc_auc" if purpose == "classification" else "neg_mean_squared_error"
+            ),
+            cv=cv,
+            n_jobs=n_jobs,
+            verbose=verbose,
+        )
+        gs.fit(x_train, y_train)
+        best_clf = gs.best_estimator_
+        # make sure x_train and x_test has the same name
+        x_true = x_true.reindex(columns=x_train.columns, fill_value=0)
+        y_pred = best_clf.predict(x_true)
+
+        # y_pred_proba
+        if hasattr(best_clf, "predict_proba"):
+            y_pred_proba = best_clf.predict_proba(x_true)[:, 1]
+        elif hasattr(best_clf, "decision_function"):
+            # If predict_proba is not available, use decision_function (e.g., for SVM)
+            y_pred_proba = best_clf.decision_function(x_true)
+            # Ensure y_pred_proba is within 0 and 1 bounds
+            y_pred_proba = (y_pred_proba - y_pred_proba.min()) / (
+                y_pred_proba.max() - y_pred_proba.min()
+            )
+        else:
+            y_pred_proba = None  # No probability output for certain models
+
+        validation_scores = {}
+        if y_true is not None:
+            validation_scores = cal_metrics(
+                y_true,
+                y_pred,
+                y_pred_proba=y_pred_proba,
+                is_binary=is_binary,
+                purpose=purpose,
+                average="weighted",
+            )
+
+            # Calculate ROC curve
+            # https://scikit-learn.org/stable/auto_examples/model_selection/plot_roc.html
+            if y_pred_proba is not None:
+                # fpr, tpr, roc_auc = dict(), dict(), dict()
+                fpr, tpr, _ = roc_curve(y_true, y_pred_proba)
+                lower_ci, upper_ci = cal_auc_ci(y_true, y_pred_proba, verbose=False)
+                roc_auc = auc(fpr, tpr)
+                roc_info = {
+                    "fpr": fpr.tolist(),
+                    "tpr": tpr.tolist(),
+                    "auc": roc_auc,
+                    "ci95": (lower_ci, upper_ci),
+                }
+                # precision-recall curve
+                precision_, recall_, _ = precision_recall_curve(y_true, y_pred_proba)
+                avg_precision_ = average_precision_score(y_true, y_pred_proba)
+                pr_info = {
+                    "precision": precision_,
+                    "recall": recall_,
+                    "avg_precision": avg_precision_,
+                }
+            else:
+                roc_info, pr_info = None, None
+            if purpose == "classification":
+                results[name] = {
+                    "best_clf": gs.best_estimator_,
+                    "best_params": gs.best_params_,
+                    "auc_indiv": [
+                        gs.cv_results_[f"split{i}_test_score"][gs.best_index_]
+                        for i in range(cv_folds)
+                    ],
+                    "scores": validation_scores,
+                    "roc_curve": roc_info,
+                    "pr_curve": pr_info,
+                    "confusion_matrix": confusion_matrix(y_true, y_pred),
+                    "predictions": y_pred.tolist(),
+                    "predictions_proba": (
+                        y_pred_proba.tolist() if y_pred_proba is not None else None
+                    ),
+                }
+            else:  # "regression"
+                results[name] = {
+                    "best_clf": gs.best_estimator_,
+                    "best_params": gs.best_params_,
+                    "scores": validation_scores,  # e.g., neg_MSE, R², etc.
+                    "predictions": y_pred.tolist(),
+                    "predictions_proba": (
+                        y_pred_proba.tolist() if y_pred_proba is not None else None
+                    ),
+                }
+
+        else:
+            results[name] = {
+                "best_clf": gs.best_estimator_,
+                "best_params": gs.best_params_,
+                "scores": validation_scores,
+                "predictions": y_pred.tolist(),
+                "predictions_proba": (
+                    y_pred_proba.tolist() if y_pred_proba is not None else None
+                ),
+            }
+
+    # Convert results to DataFrame
+    df_results = pd.DataFrame.from_dict(results, orient="index")
+
+    # sort
+    if y_true is not None and purpose == "classification":
+        df_scores = pd.DataFrame(
+            df_results["scores"].tolist(), index=df_results["scores"].index
+        ).sort_values(by="roc_auc", ascending=False)
+        df_results = df_results.loc[df_scores.index]
+
+        if plot_:
+            from datetime import datetime
+
+            now_ = datetime.now().strftime("%y%m%d_%H%M%S")
+            nexttile = plot.subplot(figsize=[12, 10])
+            plot.heatmap(df_scores, kind="direct", ax=nexttile())
+            plot.figsets(xangle=30)
+            if dir_save:
+                ips.figsave(dir_save + f"scores_sorted_heatmap{now_}.pdf")
+            if df_scores.shape[0] > 1:  # draw cluster
+                plot.heatmap(df_scores, kind="direct", cluster=True)
+                plot.figsets(xangle=30)
+                if dir_save:
+                    ips.figsave(dir_save + f"scores_clus{now_}.pdf")
+    if all([plot_, y_true is not None, purpose == "classification"]):
+        try:
+            if len(models) > 3:
+                plot_validate_features(df_results)
+            else:
+                plot_validate_features_single(df_results, figsize=(12, 4 * len(models)))
+            if dir_save:
+                ips.figsave(dir_save + f"validate_features{now_}.pdf")
+        except Exception as e:
+            print(f"Error: 在画图的过程中出现了问题:{e}")
+    return df_results
+
+
+def cal_metrics(
+    y_true, y_pred, y_pred_proba=None, is_binary=True,purpose="regression", average="weighted"
+):
+    """
+    Calculate regression or classification metrics based on the purpose.
+
+    Parameters:
+    - y_true: Array of true values.
+    - y_pred: Array of predicted labels for classification or predicted values for regression.
+    - y_pred_proba: Array of predicted probabilities for classification (optional).
+    - purpose: str, "regression" or "classification".
+    - average: str, averaging method for multi-class classification ("binary", "micro", "macro", "weighted", etc.).
+
+    Returns:
+    - validation_scores: dict of computed metrics.
+    """
+    from sklearn.metrics import (
+        mean_squared_error,
+        mean_absolute_error,
+        mean_absolute_percentage_error,
+        explained_variance_score,
+        r2_score,
+        mean_squared_log_error,
+        accuracy_score,
+        precision_score,
+        recall_score,
+        f1_score,
+        roc_auc_score,
+        matthews_corrcoef,
+        confusion_matrix,
+        balanced_accuracy_score,
+        average_precision_score,
+        precision_recall_curve,
+    )
+
+    validation_scores = {}
+
+    if purpose == "regression":
+        y_true = np.asarray(y_true)
+        y_true = y_true.ravel()
+        y_pred = np.asarray(y_pred)
+        y_pred = y_pred.ravel()
+        # Regression metrics
+        validation_scores = {
+            "mse": mean_squared_error(y_true, y_pred),
+            "rmse": np.sqrt(mean_squared_error(y_true, y_pred)),
+            "mae": mean_absolute_error(y_true, y_pred),
+            "r2": r2_score(y_true, y_pred),
+            "mape": mean_absolute_percentage_error(y_true, y_pred),
+            "explained_variance": explained_variance_score(y_true, y_pred),
+            "mbd": np.mean(y_pred - y_true),  # Mean Bias Deviation
+        }
+        # Check if MSLE can be calculated
+        if np.all(y_true >= 0) and np.all(y_pred >= 0):  # Ensure no negative values
+            validation_scores["msle"] = mean_squared_log_error(y_true, y_pred)
+        else:
+            validation_scores["msle"] = "Cannot be calculated due to negative values"
+
+    elif purpose == "classification":
+        # Classification metrics
+        validation_scores = {
+            "accuracy": accuracy_score(y_true, y_pred),
+            "precision": precision_score(y_true, y_pred, average=average),
+            "recall": recall_score(y_true, y_pred, average=average),
+            "f1": f1_score(y_true, y_pred, average=average),
+            "mcc": matthews_corrcoef(y_true, y_pred),
+            "specificity": None,
+            "balanced_accuracy": balanced_accuracy_score(y_true, y_pred),
+        }
+
+        # Confusion matrix to calculate specificity
+        if is_binary:
+            tn, fp, fn, tp = confusion_matrix(y_true, y_pred).ravel()
+        else:
+            cm=onfusion_matrix(y_true, y_pred)
+        validation_scores["specificity"] = (
+            tn / (tn + fp) if (tn + fp) > 0 else 0
+        )  # Specificity calculation
+
+        if y_pred_proba is not None:
+            # Calculate ROC-AUC
+            validation_scores["roc_auc"] = roc_auc_score(y_true, y_pred_proba)
+            # PR-AUC (Precision-Recall AUC) calculation
+            validation_scores["pr_auc"] = average_precision_score(y_true, y_pred_proba)
+    else:
+        raise ValueError(
+            "Invalid purpose specified. Choose 'regression' or 'classification'."
+        )
+
+    return validation_scores