R语言recommendlab包的ALS implicit算法问题

COS论坛 | 统计之都 COS论坛 | 统计之都 统计学世界 数据挖掘和机器学习 R语言recommendlab包的ALS implicit算法问题

该主题包含 0 条回复,1个帖子,最后由  littleyang1 周, 4 天 之前 更新。

查看 1 个帖子(总计 1 个)
  • 作者
    帖子
  • 1 楼

    littleyang
    Participant

    我使用Movielens的rating数据使用ALS——implicit可以正常执行,使用MSWeb的binary数据时就会出错

    ALS_implicit run fold/sample [model time/prediction time]
    1 Used parameters:
    lambda = 0.1
    alpha = 0.5
    n_factors = 10
    n_iterations = 10
    min_item_nr = 1
    seed = 1234
    verbose = TRUE
    Timing stopped at: 0.02 0 0.02
    Error in matrix2[only_new_users, , drop = FALSE] :
    invalid or not-yet-implemented ‘Matrix’ subsetting
    Warning message:
    In .local(x, method, …) :
    Recommender ‘ALS_implicit’ has failed and has been removed from the results!

    换成别的二值数据也是一样的错误?求大神帮忙看一下是什么问题
    以下是程序:
    data(MSWeb)
    MSWeb10 <-MSWeb[which(rowSums(MSWeb)>10),]
    scheme <- evaluationScheme(MSWeb10, method=”cross”, k=4, given=10)
    algorithms <- list(
    “random items” = list(name=”RANDOM”, param=list(normalize = “Z-score”)),
    “item-based CF” = list(name=”IBCF”, param=list(k=50)),
    “ALS_implicit” = list(name=”ALS_implicit”,
    param = list(lambda=0.1, alpha = 0.5, n_factors=10,
    n_iterations=10, seed = 1234, verbose = TRUE))

    )
    results <- evaluate(scheme, algorithms, n=c(1, 3, 5, 10,15, 20))

查看 1 个帖子(总计 1 个)

您必须先登录才能回复该主题。