可以參考
LightGBM原生/sk接口的常用參數(shù)
LightGBM使用

lightGBM調(diào)參

所有的參數(shù)含義，參考：http://lightgbm.apachecn.org/cn/latest/Parameters.html

調(diào)參過程：

1. num_leaves

   LightGBM使用的是leaf-wise的算法，因此在調(diào)節(jié)樹的復(fù)雜程度時(shí)，使用的是num_leaves而不是max_depth。

2. 樣本分布非平衡數(shù)據(jù)集：可以param[‘is_unbalance’]=’true’；

3. Bagging參數(shù)：bagging_fraction+bagging_freq（必須同時(shí)設(shè)置）、feature_fraction。bagging_fraction可以使bagging的更快的運(yùn)行出結(jié)果，feature_fraction設(shè)置在每次迭代中使用特征的比例；

4. min_data_in_leaf、min_sum_hessian_in_leaf：調(diào)大它的值可以防止過擬合，它的值通常設(shè)置的比較大。

sklearn接口形式的LightGBM示例

這里主要以sklearn的使用形式來使用lightgbm算法，包含建模，訓(xùn)練，預(yù)測，網(wǎng)格參數(shù)優(yōu)化。

import lightgbm as lgb
import pandas as pd
from sklearn.metrics import mean_squared_error
from sklearn.model_selection import GridSearchCV
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.datasets import  make_classification
# 加載數(shù)據(jù)
print('Load data...')

iris = load_iris()
data=iris.data
target = iris.target
X_train,X_test,y_train,y_test =train_test_split(data,target,test_size=0.2)

# df_train = pd.read_csv('../regression/regression.train', header=None, sep='\t')
# df_test = pd.read_csv('../regression/regression.test', header=None, sep='\t')
# y_train = df_train[0].values
# y_test = df_test[0].values
# X_train = df_train.drop(0, axis=1).values
# X_test = df_test.drop(0, axis=1).values

print('Start training...')
# 創(chuàng)建模型，訓(xùn)練模型
gbm = lgb.LGBMRegressor(objective='regression',num_leaves=31,learning_rate=0.05,n_estimators=20)
gbm.fit(X_train, y_train,eval_set=[(X_test, y_test)],eval_metric='l1',early_stopping_rounds=5)

print('Start predicting...')
# 測試機(jī)預(yù)測
y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration_)
# 模型評估
print('The rmse of prediction is:', mean_squared_error(y_test, y_pred) ** 0.5)

# feature importances
print('Feature importances:', list(gbm.feature_importances_))

# 網(wǎng)格搜索，參數(shù)優(yōu)化
estimator = lgb.LGBMRegressor(num_leaves=31)

param_grid = {
    'learning_rate': [0.01, 0.1, 1],
    'n_estimators': [20, 40]
}

gbm = GridSearchCV(estimator, param_grid)

gbm.fit(X_train, y_train)

print('Best parameters found by grid search are:', gbm.best_params_)

原生形式使用lightgbm

# coding: utf-8
# pylint: disable = invalid-name, C0111
import json
import lightgbm as lgb
import pandas as pd
from sklearn.metrics import mean_squared_error
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn.datasets import  make_classification

iris = load_iris()
data=iris.data
target = iris.target
X_train,X_test,y_train,y_test =train_test_split(data,target,test_size=0.2)


# 加載你的數(shù)據(jù)
# print('Load data...')
# df_train = pd.read_csv('../regression/regression.train', header=None, sep='\t')
# df_test = pd.read_csv('../regression/regression.test', header=None, sep='\t')
#
# y_train = df_train[0].values
# y_test = df_test[0].values
# X_train = df_train.drop(0, axis=1).values
# X_test = df_test.drop(0, axis=1).values

# 創(chuàng)建成lgb特征的數(shù)據(jù)集格式
lgb_train = lgb.Dataset(X_train, y_train)
lgb_eval = lgb.Dataset(X_test, y_test, reference=lgb_train)

# 將參數(shù)寫成字典下形式
params = {
    'task': 'train',
    'boosting_type': 'gbdt',  # 設(shè)置提升類型
    'objective': 'regression', # 目標(biāo)函數(shù)
    'metric': {'l2', 'auc'},  # 評估函數(shù)
    'num_leaves': 31,   # 葉子節(jié)點(diǎn)數(shù)
    'learning_rate': 0.05,  # 學(xué)習(xí)速率
    'feature_fraction': 0.9, # 建樹的特征選擇比例
    'bagging_fraction': 0.8, # 建樹的樣本采樣比例
    'bagging_freq': 5,  # k 意味著每 k 次迭代執(zhí)行bagging
    'verbose': 1 # <0 顯示致命的, =0 顯示錯(cuò)誤 (警告), >0 顯示信息
}

print('Start training...')
# 訓(xùn)練 cv and train
gbm = lgb.train(params,lgb_train,num_boost_round=20,valid_sets=lgb_eval,early_stopping_rounds=5)

print('Save model...')
# 保存模型到文件
gbm.save_model('model.txt')

print('Start predicting...')
# 預(yù)測數(shù)據(jù)集
y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration)
# 評估模型
print('The rmse of prediction is:', mean_squared_error(y_test, y_pred) ** 0.5)