How to save a data.tree in R?

Question

I have been looking over the manual of the data.tree package, however I was not able to find a way to save a data.tree structure once you created one. Is there a way to write a data.tree structure instead of transforming it to another structure(i.e. data.frame) and then saving it?

ok guys, to be fair because I appreciate both the answers I rolled a dice to pick the accepted answer and the winner is @Vandenman. — 20-roso, Aug 02 '16 at 13:36
No worries on that, but just FYI the R gurus (like Hadley Wickham) tell us it's computationally faster, more reliable and more efficient to save things as .RDS files rather than .RData. — Hack-R, Aug 02 '16 at 13:43
@Hack-R thank you for the piece of information, much appreciated!!! — 20-roso, Aug 02 '16 at 16:00
I tried to save a data.tree of 500 Mb in memory to a RData-file, but had to cancel it after it had surpassed 160 Gb. I haven't tried saving as a RDS-file though - do you know if this works? — Esben Eickhardt, Sep 05 '17 at 06:43

score 1 · Accepted Answer · answered Aug 02 '16 at 13:24

1

Your question is a bit vague as I have no clue whatsoever what type of object you are referring to. However, this seems to work for me.

data("acme")
a = acme
class(a)
[1] "Node" "R6" 

x = tempdir()
setwd(tempdir())
save(a, file = 'test.Rdata')
rm(a)
load('test.Rdata')
a

answered Aug 02 '16 at 13:24

Vandenman

3,046
20
33

Oh ok, thnx, thought it had some special way. – 20-roso Aug 02 '16 at 13:29
1

Saving like this works fine. Note, however, that saved objects tend to be very large, as each Node is an environment. See here for an explanation why: http://stackoverflow.com/questions/13912867/empty-r-environment-becomes-large-file-when-saved. If that bothers too much, you can convert it to a list (e.g. `as.list`) or data.frame before saving it, and then convert it back after loading it (using `as.Node`). – Christoph Glur Aug 03 '16 at 18:22
@ChristophGlur makes a great point! My data.tree object of 26000+ nodes could not be saved as a RData-file due to the massive size. It just seems very slow having to convert back and forth just to save an object! – Esben Eickhardt Sep 05 '17 at 06:45

score 1 · Answer 2 · answered Aug 02 '16 at 13:25

data(acme)
print(acme)
acme$fieldsAll
acme$count
acme$totalCount
acme$isRoot
acme$height
print(acme, "p", "cost")

outsource <- acme$IT$Outsource
class(outsource)
print(outsource)
outsource$fields
outsource$isLeaf
outsource$level
outsource$path
outsource$p
outsource$parent$name
outsource$root$name
outsource$expCost <- outsource$p * outsource$cost
print(acme, "expCost")

acme$Get("p")
acme$Do(function(x) x$expCost <- x$p * x$cost)
acme$Get("expCost", filterFun = isLeaf)

ToDataFrameTable(acme, "name", "p", "cost", "level", "pathString")
ToDataFrameTree(acme, "name", "p", "cost", "level")
ToDataFrameNetwork(acme, "p", "cost")

tree <- ToDataFrameTree(acme, "name", "p", "cost", "level")
saveRDS(tree, "tree.RDS")

it_works <- readRDS("tree.RDS")
it_works

                          levelName                     name    p    cost level
1  Acme Inc.                                       Acme Inc.   NA      NA     1
2   ¦--Accounting                                 Accounting   NA      NA     2
3   ¦   ¦--New Software                         New Software 0.50 1000000     3
4   ¦   °--New Accounting Standards New Accounting Standards 0.75  500000     3
5   ¦--Research                                     Research   NA      NA     2
6   ¦   ¦--New Product Line                 New Product Line 0.25 2000000     3
7   ¦   °--New Labs                                 New Labs 0.90  750000     3
8   °--IT                                                 IT   NA      NA     2
9       ¦--Outsource                               Outsource 0.20  400000     3
10      ¦--Go agile                                 Go agile 0.05  250000     3
11      °--Switch to R                           Switch to R 1.00   50000     3
>

A ok, I searched for "save" or "write" in the manual could not find anything, now it makes sense to me, thnx. — 20-roso, Aug 02 '16 at 13:39

score 1 · Answer 3 · answered Sep 05 '17 at 08:12

This is an answer to a related question: How does one save a data.tree into another object?

This is pretty important as several methods alter the original data.tree, a problem I have been running into.

# What one would expect
myDataTree_clone <- myDataTree

If you alter this "clone" you actually also alter the myDataTree object.

# What one should do instead
myDataTree_clone <- Clone(myDataTree)

Now you can work with the clone without altering the original myDataTree object.

How to save a data.tree in R?

3 Answers3