R 中 xgboost 中的 xgb.create.features 返回异常答案

【问题标题】：xgb.create.features in xgboost in R returns exceptionR 中 xgboost 中的 xgb.create.features 返回异常
【发布时间】：2020-03-14 19:34:17
【问题描述】：

在尝试应用包含在 R 中 xgboost 包中并在用户手册中介绍的 xgb.create.features(model, data, ...) 函数时，我遇到了一个我不明白的异常。这是一个可重现的示例：

library(xgboost)
data(mtcars)

X = as.matrix(mtcars[, -9])
Y = mtcars$am

dtrain = xgb.DMatrix(data = X, label = Y)

model = xgb.train(data = dtrain, 

                                     feval = performance_fun_F,

                                     verbose =0,  maximize = TRUE, 

                                     params = list(objective = "binary:logistic",

                                                   eta = 0.1,

                                                   max_depth = 6,

                                                   subsample = 0.8,

                                                   lambda = 0.1 ), 

                                    nrounds = 10)

dtrain1 = xgb.create.features(model, dtrain)

Error: not-yet-implemented method for cbind2(<xgb.DMatrix>, <dgCMatrix>).
 ->>  Ask the package authors to implement the missing feature.
Traceback:

1. xgb.create.features(model, dtrain)
2. cBind(data, sparse.model.matrix(~. - 1, cols))
3. base::cbind(..., deparse.level = deparse.level)
4. cbind(deparse.level, ...)
5. cbind2(..1, r)
6. cbind2(..1, r)
7. .bail.out.2(.Generic, class(x), class(y))
8. stop(gettextf("not-yet-implemented method for %s(<%s>, <%s>).\n ->>  Ask the package authors to implement the missing feature.", 
 .     fun, cl1[1L], cl2[1L]), call. = FALSE, domain = NA)

【问题讨论】：

什么是performance_fun_F？

标签： r xgboost

【解决方案1】：

您应该将矩阵X 提供给xgb.create.features 而不是dtrain：

例子：

library(xgboost)

data(mtcars)
X = as.matrix(mtcars[, -9])
dtrain = xgb.DMatrix(data = X, label = Y)

model = xgb.train(data = dtrain, 
                  eval = "auc",
                  verbose =0,  maximize = TRUE, 
                  params = list(objective = "binary:logistic",
                                eta = 0.1,
                                max_depth = 6,
                                subsample = 0.8,
                                lambda = 0.1 ), 
                  nrounds = 10)

dtrain1 = xgb.create.features(model, X)
dtrain1 = xgb.DMatrix(data = dtrain1, label = Y)

现在：

bst <- xgb.train(data = dtrain1, 
                 eval = "auc",
                 verbose =0,  maximize = TRUE, 
                 params = list(objective = "binary:logistic",
                               eta = 0.1,
                               max_depth = 6,
                               subsample = 0.8,
                               lambda = 0.1 ), 
                 nrounds = 10)

【讨论】：