Hello, If we got a data frame like below, how to use "data" to predict type, if "data" is another data frame (called nested data frame): by_subject #> # A tibble: 1000 x 3 #> subject type data #> <fct> <fct> <list> #> 1 subject1 aa <tibble [100 ? 10]> #> 2 subject2 bb <tibble [100 ? 10]> #> 3 subject3 cc <tibble [100 ? 10]> #> # ? with 997 more rows by_subject$data[[1]] #> # A tibble: 100 x 10 #> parts weight length height #> <int> <dbl> <int> <dbl> #> 1 1 28.8 100 170 #> 2 2 30.3 105 169 #> 2 3 10.5 109 189 #> # ? with 97 more rows Sincerely yours,
Your question is too vague to answer. Many R functions have "predict" methods that can be used to make predictions using a new object (e.g. a data frame) from a fitted object fit on another data frame/object. See e.g. ?predict.lm for an exemplar. But that's the closest I can come to guessing what you want. I think you need to spend time with a tutorial or two on whatever functions/methods you are using for "machine learning." You should not expect us to do such homework for you. Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Sun, Oct 13, 2019 at 10:04 AM vod vos via R-help <r-help at r-project.org> wrote:> Hello, > > If we got a data frame like below, how to use "data" to predict type, > if "data" is another data frame (called nested data frame): > > > by_subject > #> # A tibble: 1000 x 3 > #> subject type data > #> <fct> <fct> <list> > #> 1 subject1 aa <tibble [100 ? 10]> > #> 2 subject2 bb <tibble [100 ? 10]> > #> 3 subject3 cc <tibble [100 ? 10]> > #> # ? with 997 more rows > > > by_subject$data[[1]] > #> # A tibble: 100 x 10 > #> parts weight length height > #> <int> <dbl> <int> <dbl> > #> 1 1 28.8 100 170 > #> 2 2 30.3 105 169 > #> 2 3 10.5 109 189 > #> # ? with 97 more rows > > Sincerely yours, > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >[[alternative HTML version deleted]]
Jeff Newmiller
2019-Oct-13 21:08 UTC
[R] Can nested data frame be used in machine learning?
predict doesn't know about nesting. You can either unnest or use lapply or purrr:map to create new nested data frames with predicted results, but it is up to you to decide where you want to put the results. On October 13, 2019 10:03:36 AM PDT, vod vos via R-help <r-help at r-project.org> wrote:>Hello, > >If we got a data frame like below, how to use "data" to predict type, > if "data" is another data frame (called nested data frame): > > >by_subject >#> # A tibble: 1000 x 3 >#> subject type data >#> <fct> <fct> <list> >#> 1 subject1 aa <tibble [100 ? 10]> >#> 2 subject2 bb <tibble [100 ? 10]> >#> 3 subject3 cc <tibble [100 ? 10]> >#> # ? with 997 more rows > > >by_subject$data[[1]] >#> # A tibble: 100 x 10 >#> parts weight length height >#> <int> <dbl> <int> <dbl> >#> 1 1 28.8 100 170 >#> 2 2 30.3 105 169 >#> 2 3 10.5 109 189 >#> # ? with 97 more rows > >Sincerely yours, > >______________________________________________ >R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see >https://stat.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide >http://www.R-project.org/posting-guide.html >and provide commented, minimal, self-contained, reproducible code.-- Sent from my phone. Please excuse my brevity.