pip install lightgbm

import lightgbm as lgb
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
import pandas as pd

# 加載數(shù)據(jù)
data = pd.read_csv('example_data.csv')
X = data.drop(columns=['target'])
y = data['target']

# 劃分訓(xùn)練集和測(cè)試集
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 創(chuàng)建LightGBM數(shù)據(jù)集
lgb_train = lgb.Dataset(X_train, y_train)
lgb_eval = lgb.Dataset(X_test, y_test, reference=lgb_train)

# 設(shè)置參數(shù)
params = {
    'boosting_type': 'gbdt',
    'objective': 'binary',
    'metric': 'binary_logloss',
    'num_leaves': 31,
    'learning_rate': 0.05,
    'feature_fraction': 0.9,
    'bagging_fraction': 0.8,
    'bagging_freq': 5,
    'verbose': 0
}

# 訓(xùn)練模型
gbm = lgb.train(params, lgb_train, num_boost_round=20, valid_sets=lgb_eval, early_stopping_rounds=5)

# 預(yù)測(cè)
y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration)
y_pred = [1 if x > 0.5 else 0 for x in y_pred]

# 評(píng)估模型
accuracy = accuracy_score(y_test, y_pred)
print(f'Accuracy: {accuracy * 100:.2f}%')

import pandas as pd

# 讀取數(shù)據(jù)
data = pd.read_csv('house_prices.csv')

# 查看數(shù)據(jù)基本信息
print(data.info())
print(data.describe())

# 處理缺失值
data.fillna(data.mean(), inplace=True)

# 特征選擇
features = ['area', 'bedrooms', 'bathrooms', 'garage']
X = data[features]
y = data['price']

# 劃分訓(xùn)練集和測(cè)試集
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

import lightgbm as lgb

# 創(chuàng)建LightGBM數(shù)據(jù)集
lgb_train = lgb.Dataset(X_train, y_train)
lgb_eval = lgb.Dataset(X_test, y_test, reference=lgb_train)

# 設(shè)置參數(shù)
params = {
    'boosting_type': 'gbdt',
    'objective': 'regression',
    'metric': 'rmse',
    'num_leaves': 31,
    'learning_rate': 0.05,
    'feature_fraction': 0.9,
    'bagging_fraction': 0.8,
    'bagging_freq': 5,
    'verbose': 0
}

# 訓(xùn)練模型
gbm = lgb.train(params, lgb_train, num_boost_round=200, valid_sets=lgb_eval, early_stopping_rounds=10)

from sklearn.model_selection import GridSearchCV
from lightgbm import LGBMRegressor

# 定義參數(shù)范圍
param_grid = {
    'num_leaves': [31, 50, 100],
    'learning_rate': [0.05, 0.1, 0.2]
}

# 使用GridSearchCV進(jìn)行參數(shù)搜索
model = LGBMRegressor()
grid_search = GridSearchCV(model, param_grid, cv=5, scoring='neg_mean_squared_error')
grid_search.fit(X_train, y_train)

# 輸出最佳參數(shù)
best_params = grid_search.best_params_
print(f'Best parameters: {best_params}')

# 使用最佳參數(shù)重新訓(xùn)練模型
final_model = LGBMRegressor(**best_params)
final_model.fit(X_train, y_train)

import lightgbm as lgb
import numpy as np

# 生成一些示例數(shù)據(jù)
data = np.random.rand(1000, 1)
label = np.random.randint(0, 2, size=1000)

# 轉(zhuǎn)換為L(zhǎng)ightGBM數(shù)據(jù)集
lgb_data = lgb.Dataset(data, label=label)

# 設(shè)置參數(shù)
params = {
    'boosting_type': 'gbdt',
    'objective': 'binary',
    'metric': 'binary_logloss',
    'num_leaves': 31,
    'learning_rate': 0.05,
    'verbose': -1
}

# 訓(xùn)練模型
gbm = lgb.train(params, lgb_data, num_boost_round=10)

# 獲取第一棵樹的結(jié)構(gòu)
tree_info = gbm.dump_model()['tree_info'][0]['tree_structure']
print(tree_info)

# 計(jì)算特征重要性
feature_importance = gbm.feature_importance()

# 打印特征重要性
for feature, importance in zip(features, feature_importance):
    print(f'{feature}: {importance}')

# 使用交叉驗(yàn)證
cv_results = lgb.cv(params, lgb_data, num_boost_round=100, nfold=5, stratified=False, shuffle=True, metrics='rmse', early_stopping_rounds=10)

# 打印交叉驗(yàn)證結(jié)果
print(cv_results)

import pandas as pd
import lightgbm as lgb
from sklearn.model_selection import train_test_split
from sklearn.metrics import roc_auc_score

# 加載數(shù)據(jù)
data = pd.read_csv('credit_data.csv')
X = data.drop(columns=['credit_score'])
y = data['credit_score']

# 劃分訓(xùn)練集和測(cè)試集
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 創(chuàng)建LightGBM數(shù)據(jù)集
lgb_train = lgb.Dataset(X_train, y_train)
lgb_eval = lgb.Dataset(X_test, y_test, reference=lgb_train)

# 設(shè)置參數(shù)
params = {
    'boosting_type': 'gbdt',
    'objective': 'binary',
    'metric': 'auc',
    'num_leaves': 31,
    'learning_rate': 0.05,
    'feature_fraction': 0.9,
    'bagging_fraction': 0.8,
    'bagging_freq': 5,
    'verbose': 0
}

# 訓(xùn)練模型
gbm = lgb.train(params, lgb_train, num_boost_round=200, valid_sets=lgb_eval, early_stopping_rounds=10)

# 預(yù)測(cè)
y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration)

# 評(píng)估模型
auc = roc_auc_score(y_test, y_pred)
print(f'AUC: {auc:.4f}')

import pandas as pd
import lightgbm as lgb
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# 加載數(shù)據(jù)
data = pd.read_csv('medical_data.csv')
X = data.drop(columns=['disease_label'])
y = data['disease_label']

# 劃分訓(xùn)練集和測(cè)試集
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 創(chuàng)建LightGBM數(shù)據(jù)集
lgb_train = lgb.Dataset(X_train, y_train)
lgb_eval = lgb.Dataset(X_test, y_test, reference=lgb_train)

# 設(shè)置參數(shù)
params = {
    'boosting_type': 'gbdt',
    'objective': 'multiclass',
    'metric': 'multi_logloss',
    'num_class': 3,
    'num_leaves': 31,
    'learning_rate': 0.05,
    'feature_fraction': 0.9,
    'bagging_fraction': 0.8,
    'bagging_freq': 5,
    'verbose': 0
}

# 訓(xùn)練模型
gbm = lgb.train(params, lgb_train, num_boost_round=200, valid_sets=lgb_eval, early_stopping_rounds=10)

# 預(yù)測(cè)
y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration)
y_pred = np.argmax(y_pred, axis=1)

# 評(píng)估模型
accuracy = accuracy_score(y_test, y_pred)
print(f'Accuracy: {accuracy * 100:.2f}%')