xgboost错误：setinfo.xgb.DMatrix（dmat，names（p），p [[1]]）错误：标签的长度必须等于输入数据中的行数

Question

sparse_matrix <- sparse.model.matrix(state ~ . -Month -LanID, data = smoted_data)
smoted_data_mat_label<-as.matrix(as.numeric(smoted_data$state))
smoted_data_mat_label = smoted_data[,"state"] == 1

    xgb.fit <- xgboost(
    data = sparse_matrix,
    label = smoted_data_mat_label,
    eta = 1.8,
    max_depth = 7,
    min_child_weight = 11, # could be anything all are showing same results
    nrounds = 9, # could be anything all are showing same results
    nfold = 10,
    objective = "binary:logistic",  # for regression models
    verbose = 1,               # silent,
    early_stopping_rounds = 10 # stop if no improvement for 10 consecutive trees
  )

Error in setinfo.xgb.DMatrix(dmat, names(p), p[[1]]) : 
  The length of labels must equal to the number of rows in the input data

length(smoted_data_mat_label)
[1] 22721
>   length(sparse_matrix)
[1] 41739681
  dim(smoted_data)
[1] 22721    31
   length(smoted_data$LanID)
[1] 22721

以前的代码运行良好，现在导致下面的错误，不确定为什么？我检查了sparse_matrix的类，它是矩阵，标号也是一个矩阵。

Answer 1

之所以发生此问题，是因为在新数据中引入了一些NA。删除了NA，算法运行良好。