[
https://issues.apache.org/jira/browse/SPARK-11086?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14955336#comment-14955336
]
Shivaram Venkataraman commented on SPARK-11086:
-----------------------------------------------
This is a great suggestion. Could you open a PR for this ?
> createDataFrame should dropFactor column-wise not cell-wise
> ------------------------------------------------------------
>
> Key: SPARK-11086
> URL: https://issues.apache.org/jira/browse/SPARK-11086
> Project: Spark
> Issue Type: Improvement
> Components: SparkR
> Reporter: Maciej Szymkiewicz
>
> At this moment SparkR {{createDataFrame}} [is using nested
> loop|https://github.com/apache/spark/blob/896edb51ab7a88bbb31259e565311a9be6f2ca6d/R/pkg/R/SQLContext.R#L99]
> to convert {{factors}} to {{character}} when called on a local
> {{data.frame}}.
> {code}
> data <- lapply(1:n, function(i) {
> lapply(1:m, function(j) { dropFactor(data[i,j]) })
> })
> {code}
> It works but is incredibly slow especially with {{data.table}} (~ 2 orders of
> magnitude compared to PySpark / Pandas version on a DateFrame of size 1M
> rows x 2 columns).
> A simple improvement is to apply {{dropFactor}} column-wise and then reshape
> output list:
> {code}
> args <- list(FUN=list, SIMPLIFY=FALSE, USE.NAMES=FALSE)
> data <- do.call(mapply, append(args, setNames(lapply(data, dropFactor),
> NULL)))
> {code}
> It should at least partially address
> [SPARK-8277|https://issues.apache.org/jira/browse/SPARK-8277].
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
