‍

作者简介Introduction

苏高生，西南财经大学统计学硕士毕业，现就职于中国电信，主要负责企业存量客户大数据分析、数据建模。研究方向：机器学习，最喜欢的编程语言：R语言，没有之一。

E-mail：sugs01@outlook.com

往期回顾：

Xgboost算法——Kaggle案例

The rxfastforest algorithm case of kaggle

零、案例背景介绍与建模思路说明

1.背景介绍

本案例使用的数据为kaggle中“Santander Customer Satisfaction”比赛的数据。此案例为不平衡二分类问题，目标为最大化auc值（ROC曲线下方面积）。竞赛题目链接为：https://www.kaggle.com/c/santander-customer-satisfaction 。目前此比赛已经结束。

2.建模思路

本文档采用微软开源的lightgbm算法进行分类，运行速度极快，超过xgboost算法与rxFastForest算法。

1) 读取数据；

2) 并行运算：由于lightgbm包可以通过设置相应参数进行并行运算，因此不再调用doParallel与foreach包进行并行运算；

3) 特征选择：使用mlr包提取了99%的信息增益；

4) 调参：逐步调试lgb.cv函数的参数，并多次调试，直到满意为止；

5) 集成预测结果：在每个参数的适宜范围内随机抽取参数值构建lightgbm模型，并将多个模型进行集成，输出预测结果；本案例所用程序输出结果的ROC值为0.832023,已绝对超过Private Leaderboard排名第一的结果(0.829072)。

一、读取数据

options(java.parameters = "-Xmx8g") ## 特征选择时使用，但是需要在加载包之前设置,否则无效

library(readr)

lgb_tr1 <- read_csv("C:/Users/Administrator/Documents/kaggle/scs_lgb/train.csv")

lgb_te1 <- read_csv("C:/Users/Administrator/Documents/kaggle/scs_lgb/test.csv")

二、数据探索

1.设置并行运算

library(dplyr)

library(mlr)

library(parallelMap)

parallelStartSocket(2)

2.数据各列初步探索

summarizeColumns(lgb_tr1) %>% View()

3.处理缺失值

#impute missing values by mean and mode

imp_tr1 <- impute(

as.data.frame(lgb_tr1),

classes = list(

integer = imputeMean(),

numeric = imputeMean()

)

)

imp_te1 <- impute(

as.data.frame(lgb_te1),

classes = list(

integer = imputeMean(),

numeric = imputeMean()

)

)

## 处理缺失值后summarizeColumns(imp_tr1$data) %>% View()

4.观察训练数据类别的比例–数据类别不平衡

table(lgb_tr1$TARGET)

5.剔除数据集中的常数列

lgb_tr2 <- removeConstantFeatures(imp_tr1$data)

lgb_te2 <- removeConstantFeatures(imp_te1$data)

6.保留训练数据集与测试数据及相同的列

tr2_name <- data.frame(tr2_name = colnames(lgb_tr2))

te2_name <- data.frame(te2_name = colnames(lgb_te2))

tr2_name_inner <- tr2_name %>%

inner_join(te2_name, by = c('tr2_name' = 'te2_name'))

TARGET = data.frame(TARGET = lgb_tr2$TARGET)

lgb_tr2 <- lgb_tr2[, c(tr2_name_inner$tr2_name[2:dim(tr2_name_inner)[1]])]

lgb_te2 <- lgb_te2[, c(tr2_name_inner$tr2_name[2:dim(tr2_name_inner)[1]])]

lgb_tr2 <- cbind(lgb_tr2, TARGET)

三、特征筛选–信息增益

library(lightgbm)

library(ggplot2)

1.试算最优的weight参数

grid_search <- expand.grid(

weight = seq(1, 30, 2))

lgb_rate_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr2$TARGET * i + 1) / sum(lgb_tr2$TARGET * i + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr2[, 1:300]),

label = lgb_tr2$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc'

)

# 交叉验证

lgb_tr2_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

learning_rate = .1,

num_threads = 2,

early_stopping_rounds = 10

)

lgb_rate_1[i] <- unlist(lgb_tr2_mod$record_evals$valid$auc$eval)

[length(unlist(lgb_tr2_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- lgb_rate_1

ggplot(grid_search,aes(x = weight, y = perf)) +

geom_point()

结论：从此图可知auc值受权重影响不大,在weight=9时达到最大,weight>=11时呈负相关

2.特征选择

1)特征选择

lgb_tr2$TARGET <- factor(lgb_tr2$TARGET)

lgb.task <- makeClassifTask(data = lgb_tr2, target = 'TARGET')

lgb.task.smote <- oversample(lgb.task, rate = 9)

fv_time <- system.time(

fv <- generateFilterValuesData(

lgb.task.smote,

method = c('information.gain')

)

)

2)制图查看

library(ggvis)

plotFilterValues(fv)

plotFilterValuesGGVIS(fv)

3)提取99%的信息增益(lightgbm算法效率极高，因此可以取更多的变量)

fv_data2 <- fv$data %>%

arrange(desc(information.gain)) %>%

mutate(info_gain_cul = cumsum(information.gain) / sum(information.gain))

fv_data2_filter <- fv_data2 %>% filter(info_gain_cul <= 0.99)

dim(fv_data2_filter)

fv_feature <- fv_data2_filter$name

lgb_tr3 <- lgb_tr2[, c(fv_feature, 'TARGET')]

lgb_te3 <- lgb_te2[, fv_feature]

4)写出数据

write_csv(lgb_tr3, 'C:/users/Administrator/Documents/kaggle/scs_lgb/lgb_tr3.csv')

write_csv(lgb_te3, 'C:/users/Administrator/Documents/kaggle/scs_lgb/lgb_te3.csv')

四、算法

lgb_tr <- rxImport('C:/Users/Administrator/Documents/kaggle/scs_lgb/lgb_tr3.csv')

lgb_te <- rxImport('C:/Users/Administrator/Documents/kaggle/scs_lgb/lgb_te3.csv')

1.调试weight参数

grid_search <- expand.grid(

weight = 1:30

)

perf_weight_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * i + 1) / sum(lgb_tr$TARGET * i + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc'

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

learning_rate = .1,

num_threads = 2,

early_stopping_rounds = 10

)

perf_weight_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)

[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_weight_1

ggplot(grid_search,aes(x = weight, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在weight=2时达到最大

2.调试learning_rate参数

grid_search <- expand.grid(

learning_rate = 2 ^ (-(8:1))

)

perf_learning_rate_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_learning_rate_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)

[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_learning_rate_1

ggplot(grid_search,aes(x = learning_rate, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在learning_rate=2^(-3)时达到最大

3.调试num_leaves参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = seq(50, 1000, 50))

perf_num_leaves_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_num_leaves_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)

[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_num_leaves_1

ggplot(grid_search,aes(x = num_leaves, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在num_leaves=600时达到最大

4.调试min_data_in_leaf参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

min_data_in_leaf = 2 ^ (1:7))

perf_min_data_in_leaf_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

min_data_in_leaf = grid_search[i, 'min_data_in_leaf']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_min_data_in_leaf_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_min_data_in_leaf_1

ggplot(grid_search,aes(x = min_data_in_leaf, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值对min_data_in_leaf不敏感，因此不做调整

5.调试max_bin参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

max_bin = 2 ^ (5:10))

perf_max_bin_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

max_bin = grid_search[i, 'max_bin']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_max_bin_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)

[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_max_bin_1

ggplot(grid_search,aes(x = max_bin, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在max_bin=2^6时达到最大,需要再次微调max_bin值

6.微调max_bin参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

max_bin = 10 * (3:12)

)

perf_max_bin_2 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

max_bin = grid_search[i, 'max_bin']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_max_bin_2[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)

[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_max_bin_2

ggplot(grid_search,aes(x = max_bin, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在max_bin=30时达到最大

7.调试min_data_in_bin参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

max_bin = 30,

min_data_in_bin = 2 ^ (1:9)

)

perf_min_data_in_bin_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

max_bin = grid_search[i, 'max_bin'],

min_data_in_bin = grid_search[i, 'min_data_in_bin']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_min_data_in_bin_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_min_data_in_bin_1

ggplot(grid_search,aes(x = min_data_in_bin, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在min_data_in_bin=64时达到最大,但是变化极其细微,因此不做调整

8.调试feature_fraction参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

max_bin = 30,

min_data_in_bin = 64,

feature_fraction = seq(.5, 1, .02)

)

perf_feature_fraction_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

max_bin = grid_search[i, 'max_bin'],

min_data_in_bin = grid_search[i, 'min_data_in_bin'],

feature_fraction = grid_search[i, 'feature_fraction']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_feature_fraction_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_feature_fraction_1

ggplot(grid_search,aes(x = feature_fraction, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在feature_fraction=.64时达到最大,feature_fraction在[.62, .70]之间时，auc值保持稳定,表现较好;从.64开始呈下降趋势

9.调试min_sum_hessian参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

max_bin = 30,

min_data_in_bin = 64,

feature_fraction = .64,

min_sum_hessian = seq(0, .02, .001)

)

perf_min_sum_hessian_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

max_bin = grid_search[i, 'max_bin'],

min_data_in_bin = grid_search[i, 'min_data_in_bin'],

feature_fraction = grid_search[i, 'feature_fraction'],

min_sum_hessian = grid_search[i, 'min_sum_hessian']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_min_sum_hessian_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_min_sum_hessian_1

ggplot(grid_search,aes(x = min_sum_hessian, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在min_sum_hessian=0.04时达到最大,建议min_sum_hessian取值在[0.001, 0.005]区间,auc趋于最大

10.调试lamda参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

max_bin = 30,

min_data_in_bin = 64,

feature_fraction = .64,

min_sum_hessian = .004,

lambda_l1 = seq(0, .01, .002),

lambda_l2 = seq(0, .01, .002)

)

perf_lamda_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

max_bin = grid_search[i, 'max_bin'],

min_data_in_bin = grid_search[i, 'min_data_in_bin'],

feature_fraction = grid_search[i, 'feature_fraction'],

min_sum_hessian = grid_search[i, 'min_sum_hessian'],

lambda_l1 = grid_search[i, 'lambda_l1'],

lambda_l2 = grid_search[i, 'lambda_l2']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_lamda_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_lamda_1

ggplot(data = grid_search, aes(x = lambda_l1, y = perf)) +

geom_point() +

facet_wrap(~ lambda_l2, nrow = 5)

结论：从此图可知建议去lambda_l1 = .002, lambda_l2 = .008，即在不影响auc值的情况下，尽量增加lambda值以降低模型复杂度

11.调试drop_rate参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

max_bin = 30,

min_data_in_bin = 64,

feature_fraction = .64,

min_sum_hessian = .004,

lambda_l1 = .002,

lambda_l2 = .008,

drop_rate = seq(0, 1, .1)

)

perf_drop_rate_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

max_bin = grid_search[i, 'max_bin'],

min_data_in_bin = grid_search[i, 'min_data_in_bin'],

feature_fraction = grid_search[i, 'feature_fraction'],

min_sum_hessian = grid_search[i, 'min_sum_hessian'],

lambda_l1 = grid_search[i, 'lambda_l1'],

lambda_l2 = grid_search[i, 'lambda_l2'],

drop_rate = grid_search[i, 'drop_rate']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_drop_rate_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_drop_rate_1

ggplot(data = grid_search, aes(x = drop_rate, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在drop_rate=0.3时达到最大,在[.3, .4]之间较好；在[0, 1]变化不大

12.调试max_drop参数

grid_search <- expand.grid(

learning_rate = .125,

num_leaves = 600,

max_bin = 30,

min_data_in_bin = 64,

feature_fraction = .64,

min_sum_hessian = .004,

lambda_l1 = .002,

lambda_l2 = .008,

drop_rate = .3,

max_drop = seq(1, 10, 2)

)

perf_max_drop_1 <- numeric(length = nrow(grid_search))

for(i in 1:nrow(grid_search)){

lgb_weight <- (lgb_tr$TARGET * 2 + 1) / sum(lgb_tr$TARGET * 2 + 1)

lgb_train <- lgb.Dataset(

data = data.matrix(lgb_tr[, 1:137]),

label = lgb_tr$TARGET,

free_raw_data = FALSE,

weight = lgb_weight

)

# 参数

params <- list(

objective = 'binary',

metric = 'auc',

learning_rate = grid_search[i, 'learning_rate'],

num_leaves = grid_search[i, 'num_leaves'],

max_bin = grid_search[i, 'max_bin'],

min_data_in_bin = grid_search[i, 'min_data_in_bin'],

feature_fraction = grid_search[i, 'feature_fraction'],

min_sum_hessian = grid_search[i, 'min_sum_hessian'],

lambda_l1 = grid_search[i, 'lambda_l1'],

lambda_l2 = grid_search[i, 'lambda_l2'],

drop_rate = grid_search[i, 'drop_rate'],

max_drop = grid_search[i, 'max_drop']

)

# 交叉验证

lgb_tr_mod <- lgb.cv(

params,

data = lgb_train,

nrounds = 300,

stratified = TRUE,

nfold = 10,

num_threads = 2,

early_stopping_rounds = 10

)

perf_max_drop_1[i] <- unlist(lgb_tr_mod$record_evals$valid$auc$eval)[length(unlist(lgb_tr_mod$record_evals$valid$auc$eval))]}

grid_search$perf <- perf_max_drop_1

ggplot(data = grid_search, aes(x = max_drop, y = perf)) +

geom_point() +

geom_smooth()

结论：从此图可知auc值在max_drop=5时达到最大,在[1, 10]区间变化较小

==========未完待续==========

　往期精彩内容整理合集　

2017年R语言发展报告（国内）

R语言中文社区历史文章整理（作者篇）

R语言中文社区历史文章整理（类型篇）

公众号后台回复关键字即可学习

回复 R                  R语言快速入门及数据挖掘 
回复 Kaggle案例  Kaggle十大案例精讲（连载中）
回复 文本挖掘      手把手教你做文本挖掘
回复 可视化          R语言可视化在商务场景中的应用 
回复 大数据         大数据系列免费视频教程 
回复 量化投资      张丹教你如何用R语言量化投资 
回复 用户画像      京东大数据，揭秘用户画像
回复 数据挖掘     常用数据挖掘算法原理解释与应用
回复 机器学习     人工智能系列之机器学习与实践
回复 爬虫            R语言爬虫实战案例分享

lightgbm algorithm case of kaggle（上）相关推荐

  1. Kaggle上的犬种识别（ImageNet Dogs）

    Kaggle上的犬种识别(ImageNet Dogs) Dog Breed Identification (ImageNet Dogs) on Kaggle 在本节中,将解决在Kaggle竞赛中的犬种 ...

  2. 如何下载kaggle上的数据集

    如何下载kaggle上的数据集 1 直接下载 2 通过kaggle提供的API下载数据集 2.1 Cifar10 数据集地址 2.2 kaggle如何通过kaggle提供的API下载数据集 3 错误 ...

  3. Kaggle上主要有两大方法：梯度提升机和深度学习

    Python深度学习: 在2016年和2017年,Kaggle上主要有两大方法:梯度提升机和深度学习.具体而言,梯度提升机用于处理结构化数据的问题,而深度学习则用于图像分类等感知问题.使用前一种方法的 ...

  4. 在Kaggle上赢得大数据竞赛的技巧和窍门

    在Kaggle上赢得大数据竞赛的技巧和窍门 解决方案 平台 数据 应用 方法 阅读1906  原文:The tips and tricks I used to succeed on Kaggle  作 ...

  5. kaggle上面的score以及ROC_AUC数值咋回事

    TP:真的传染病患者被检测为传染病患者 FP:假的传染病患者被检测为传染病患者 假如现在10000人 其中9900正常,100人禽流感. 如果你的分类器全部设定为正常,那么精度就是9900,但是这个我 ...

  6. c罗python可视化分析_关于C罗“冲刺速度”这种隐私问题，我们在Kaggle上找到了数据集，然后。。。...

    原标题:关于C罗"冲刺速度"这种隐私问题,我们在Kaggle上找到了数据集,然后... 今日份知识你摄入了么? 你知道 FIFA 么?在线踢足球那种,贼老多人玩儿这个游戏,本仙女作 ...

  7. Kaggle上的狗品种识别（ImageNet Dogs）

    Kaggle上的狗品种识别(ImageNet Dogs) 在本节中,我们将解决Kaggle竞赛中的犬种识别挑战,比赛的网址是https://www.kaggle.com/c/dog-breed-ide ...

  8. Kaggle上传数据时出现错误，且不是网络问题

    Kaggle上传数据时出现错误,且不是网络问题 Kaggle上传数据processing 数据时出现错误 Kaggle上传数据processing 数据时出现错误 Kaggle上传数据时,能够上传数据 ...

  9. (python3)kaggle上matplotlib显示不了中文(title是框,坐标也是框)完全解决.md

    文章目录 0.问题的出现 1.查看支持的字体 2.解决title是框 2.1(方法一): 2.2(方法二:) 3.解决title和坐标轴是框框框的问题 3.1把它的代码放到kaggle上跑一下发现 3 ...

  10. 计算机视觉相关比赛,竞赛 | Kaggle上有哪些有趣又多金的计算机视觉类比赛？

    原标题:竞赛 | Kaggle上有哪些有趣又多金的计算机视觉类比赛? 本文阅读时间约5分钟 如今计算机视觉已经成为人工智能领域的热门研究对象,很多小伙伴都想通过一些有效的途径来学习和运用所学知识,有什 ...

最新文章

  1. python太阳花绘图_python中turtle如何画太阳花？
  2. Spring Boot 以 jar 包方式运行在后台
  3. 热点账户问题和常用解决方案【上】
  4. PHP语法（3.）环境变量
  5. [zz]用U盘装win7/XP系统的操作[图解]
  6. 数据结构——线性表：顺序表、单向链表、循环链表、双向链表
  7. python爬取大众点评_浅谈python爬取58同城，大众点评这两个网站
  8. 程序猿果真有前端后端client吗
  9. Git 少用 Pull 多用 Fetch 和 Merge
  10. 程序设计与算法----递归之爬楼梯问题
  11. 清华大学信息 计算机科学与技术,清华大学信息科学与技术国家实验室
  12. 零点起飞学php下载,零点起飞学PHP(附光盘)/零点起飞学编程
  13. 队列，计算机中的排队模型
  14. 【知易行难】RS485组网连接示意图
  15. 中小型园区网典型配置
  16. 手把手教你写电商爬虫-第四课 淘宝网商品爬虫自动JS渲染 1
  17. Android5.0系统下代码格式化sd卡方法
  18. 常见编码格式（中文编码）
  19. 利用腾讯会议开展远程面试的解决方案
  20. android img 制作工具,Android 镜像文件制作 【ramdisk.img system.img userdata.img】

热门文章

  1. 阿里为什么推荐使用LongAdder？而不是AtomicLong？
  2. IDEA 这个小技巧太实用了。。
  3. 我的天！！线上千万级大表排序，如何优化？
  4. 来感受一下别人的密码
  5. AWS 挂了 11 个小时：因多处光缆被挖断
  6. javascript-mqtt
  7. [js]promise学习2
  8. 强化学习读书笔记 - 03 - 有限马尔科夫决策过程
  9. PHP学习笔记之PDO
  10. group by having where order by