lasso回归分析时遇到警告信息

Question

rt <- read.csv("merged_matrix.csv",header = T,row.names = 1)
###构建模型
set.seed(10)   
x=as.matrix(rt[,c(2:ncol(rt))])
y=data.matrix(rt$fustat)
x <- scale(x)  # 标准化x
#y <- scale(y) 
fit=glmnet(x, y, family = "binomial", maxit = 3000)
plot(fit, xvar = "lambda", label = TRUE)

cvfit = cv.glmnet(x, y, family="binomial", maxit = 30000)


Warning messages:
1: In lognet(xd, is.sparse, ix, jx, y, weights, offset, alpha, nobs,  :
  one multinomial or binomial class has fewer than 8  observations; dangerous ground
2: In lognet(xd, is.sparse, ix, jx, y, weights, offset, alpha, nobs,  :
  one multinomial or binomial class has fewer than 8  observations; dangerous ground
3: In lognet(xd, is.sparse, ix, jx, y, weights, offset, alpha, nobs,  :
  one multinomial or binomial class has fewer than 8  observations; dangerous ground
4: In lognet(xd, is.sparse, ix, jx, y, weights, offset, alpha, nobs,  :
  one multinomial or binomial class has fewer than 8  observations; dangerous ground
5: In lognet(xd, is.sparse, ix, jx, y, weights, offset, alpha, nobs,  :
  one multinomial or binomial class has fewer than 8  observations; dangerous ground
6: In lognet(xd, is.sparse, ix, jx, y, weights, offset, alpha, nobs,  :
  one multinomial or binomial class has fewer than 8  observations; dangerous ground

进行losso分析时一直跳出以上警告信息，最终制作出的图中AUC为1，请求各位的帮助

Answer 1

试试下面这个：

# 1. 读取数据
rt <- read.csv("merged_matrix.csv", header = TRUE, row.names = 1)

# 2. 构建模型
set.seed(10)   
x <- as.matrix(rt[, c(2:ncol(rt))])
y <- as.numeric(rt$fustat)  # 将因变量转为数值型
x <- scale(x)  # 标准化x

# 3. 使用cv.glmnet进行LASSO交叉验证
cvfit <- cv.glmnet(x, y, family = "binomial", maxit = 30000)

# 4. 绘制交叉验证图
plot(cvfit, xvar = "lambda", label = TRUE)

# 5. 查看交叉验证的最佳lambda值
best_lambda <- cvfit$lambda.min
cat("Best lambda:", best_lambda, "\n")

# 6. 使用最佳lambda值构建最终模型
final_fit <- glmnet(x, y, family = "binomial", lambda = best_lambda, maxit = 3000)