【内含baseline】Kaggle机器学习新赛指南！文末送书~

train             = pd.read_csv('/kaggle/input/icr-identify-age-related-conditions/train.csv')test              = pd.read_csv('/kaggle/input/icr-identify-age-related-conditions/test.csv')greeks            = pd.read_csv('/kaggle/input/icr-identify-age-related-conditions/greeks.csv')sample_submission = pd.read_csv('/kaggle/input/icr-identify-age-related-conditions/sample_submission.csv')

from sklearn.impute import SimpleImputerfrom sklearn.preprocessing import MinMaxScaler, OneHotEncoder
# Combine numeric and categorical featuresFEATURES = num_cols + cat_cols
# Fill missing values with mean for numeric variablesimputer = SimpleImputer(strategy='mean')numeric_df = pd.DataFrame(imputer.fit_transform(train[num_cols]), columns=num_cols)
# Scale numeric variables using min-max scalingscaler = MinMaxScaler()scaled_numeric_df = pd.DataFrame(scaler.fit_transform(numeric_df), columns=num_cols)
# Encode categorical variables using one-hot encodingencoder = OneHotEncoder(sparse=False, handle_unknown='ignore')encoded_cat_df = pd.DataFrame(encoder.fit_transform(train[cat_cols]), columns=encoder.get_feature_names_out(cat_cols))
# Concatenate the scaled numeric and encoded categorical variablesprocessed_df = pd.concat([scaled_numeric_df, encoded_cat_df], axis=1)

from sklearn.utils import class_weight
FOLDS = 10SEED = 1004xgb_models = []xgb_oof = []f_imp = []
counter = 1X = processed_dfy = train['Class']
# Calculate the sample weightsweights = class_weight.compute_sample_weight('balanced', y)
skf = StratifiedKFold(n_splits=FOLDS, shuffle=True, random_state=SEED）for fold, (train_idx, val_idx) in enumerate(skf.split(X, y)):    if (fold + 1)%5 == 0 or (fold + 1) == 1:        print(f'{"#"*24} Training FOLD {fold+1} {"#"*24}')            X_train, y_train = X.iloc[train_idx], y.iloc[train_idx]      X_valid, y_valid = X.iloc[val_idx], y.iloc[val_idx]    watchlist = [(X_train, y_train), (X_valid, y_valid)]        # Apply weights in the XGBClassifier    model = XGBClassifier(n_estimators=1000, n_jobs=-1, max_depth=4, eta=0.2, colsample_bytree=0.67)    model.fit(X_train, y_train, sample_weight=weights[train_idx], eval_set=watchlist, early_stopping_rounds=300, verbose=0)        val_preds = model.predict_proba(X_valid)[:, 1]        # Apply weights in the log_loss    val_score = log_loss(y_valid, val_preds, sample_weight=weights[val_idx])    best_iter = model.best_iteration        idx_pred_target = np.vstack([val_idx,  val_preds, y_valid]).T    f_imp.append({i: j for i, j in zip(X.columns, model.feature_importances_)})    print(f'{" "*20} Log-loss: {val_score:.5f} {" "*6} best iteration: {best_iter}')                  xgb_oof.append(idx_pred_target)       xgb_models.append(model)      print('*'*45)print(f'Mean Log-loss: {np.mean([log_loss(item[:, 2], item[:, 1], sample_weight=weights[item[:, 0].astype(int)]) for item in xgb_oof]):.5f}')               

# Confusion Matrix for the last foldcm = confusion_matrix(y_valid, model.predict(X_valid))
# Feature Importance for the last modelfeature_imp = pd.DataFrame({'Value':xgb_models[-1].feature_importances_, 'Feature':X.columns})feature_imp = feature_imp.sort_values(by="Value", ascending=False)feature_imp_top20 = feature_imp.iloc[:20]
fig, ax = plt.subplots(1, 2, figsize=(14, 4))
# Subplot 1: Confusion Matrixsns.heatmap(cm, annot=True, fmt='d', ax=ax[0], cmap='YlOrRd')ax[0].set_title('Confusion Matrix')ax[0].set_xlabel('Predicted')ax[0].set_ylabel('True')
# Subplot 2: Feature Importancesns.barplot(x="Value", y="Feature", data=feature_imp_top20, ax=ax[1], palette='YlOrRd_r')ax[1].set_title('Feature Importance')
plt.tight_layout()plt.show()