Description Usage Arguments Value Author(s) Examples
make a mlr model + its metadata for a specific task using a specific learner
1 | makeModel(task, learner)
|
task |
an object of class mlr::makeRegrTask() |
learner |
an object of class mlr::makeLearner() |
A 2 elements named list
snitch
: a boolean. Is TRUE
if function has provided the expected result. Is FALSE
is function throws an error
output
: a named list which elements are :
value
: a named list which elements are :
trained
: an element of class mlr::makeWrappedMode()
predictions
an element of class data.frame
containing the predictions made by the trained model at station locations. Colnames are truth
(= observations), response
(= predictions), se
(= standard error) and residuals
(= truth - response)
perfs
: a named list which elements are :
iters
: a dataframe containing the information relative to the leave-one-out cross validation of the model. Colnames are learner.id
, iter
, rmse
, mse
, mae
, timetrain
agg
: a dataframe containing the aggregated performances of the LOOCV. Same colnames as iters
condition
: a character specifying the condition encountered by the function : success, warning, or error.
message
: a character specifying the message relative to the condition.
Thomas Goossens
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 | ## Not run:
# load magrittr for pipe use : %>%
library(magrittr)
# create the dataset
myDataset = makeDataset(
dfrom = "2017-03-04T15:00:00Z",
dto = "2017-03-04T18:00:00Z",
sensor = "tsa")
# extract the list of hourly sets of records
myDataset = myDataset$output$value
# create the tasks
myTasks = purrr::map(myDataset, makeTask, target = "tsa")
# extract the tasks from the outputs
myTasks = myTasks %>% purrr::modify_depth(1, ~.$"output"$"value"$"task")
# keep the first task
myTask = myTasks[[1]]
# create the model
myModel = makeModel(
task = myTask,
learner = agrometeorLearners$mulLR_lonLatAlt_NA)
# extract the relevant information
myModel = myModel$output$value
# show the results of the LOOCV of this hour forthe choosen learner
head(myModel$predictions)
head(myModel$perfs$iters)
# show the model information
myModel$trained
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.