In WinVector/cdata: Fluid Data Transformations

Showing that cdata works the same with an rquery pipeline both in memory and on databases. Example from "Fluid data reshaping with cdata".

library("cdata")
library("rqdatatable")


# example from: http://winvector.github.io/FluidData/FluidDataReshapingWithCdata.html
d <- wrapr::build_frame(
  "val_loss", "val_acc", "loss" , "acc" , "epoch" |
  -0.377    , 0.8722   , -0.5067, 0.7852, 1       |
  -0.2997   , 0.8895   , -0.3002, 0.904 , 2       |
  -0.2964   , 0.8822   , -0.2166, 0.9303, 3       |
  -0.2779   , 0.8899   , -0.1739, 0.9428, 4       |
  -0.2843   , 0.8861   , -0.1411, 0.9545, 5       |
  -0.312    , 0.8817   , -0.1136, 0.9656, 6       )

controlTable = build_frame(
  "measure"                   , "training", "validation" |
  "minus binary cross entropy", "loss"    , "val_loss"   |
  "accuracy"                  , "acc"     , "val_acc"    )

d1 <- rowrecs_to_blocks(
  d,
  controlTable = controlTable,
  columnsToCopy = "epoch")

knitr::kable(d1)

d2 <- blocks_to_rowrecs(
  d1,
  controlTable = controlTable,
  keyColumns = "epoch")

knitr::kable(d2)

ops1 <- local_td(d) %.>%
  rowrecs_to_blocks(
    .,
    controlTable = controlTable,
    columnsToCopy = "epoch")

cat(format(ops1))

d %.>% 
  ops1 %.>%
  knitr::kable(.)

ops2 <- local_td(d1) %.>%
  blocks_to_rowrecs(
    .,
    controlTable = controlTable,
    keyColumns = "epoch")

cat(format(ops2))

d1 %.>% 
  ops2 %.>%
  knitr::kable(.)

db <- DBI::dbConnect(RSQLite::SQLite(), ":memory:")

rq_copy_to(db, "d", d)

db %.>% 
  ops1 %.>%
  knitr::kable(.)

rq_copy_to(db, "d1", d1)

db %.>% ops2 %.>%
  knitr::kable(.)

DBI::dbDisconnect(db)