wt_preprocess: wt_preprocess

In MoritzFeigl/wateRtemp: Water temperature prediction modelling

Description

Preprocessing data for stream water temperature prediction. All results are stored automatically in the folder catchment in the current working directory.

Usage

wt_preprocess(
  data,
  nlags = 4,
  training_fraction = 0.8,
  year_range = c(min(data$year), max(data$year)),
  catchment = deparse(substitute(data))
)

Arguments

data	data frame containing the variables: year, month, day, wt and additional covariates.
nlags	Number of lags used for for all features except for time features.
training_fraction	Numeric of fraction of data used for training, e.g. 0.8. Testing data will consist of the (1-training_fraction) fraction of data.
year_range	Vector containing the first and last year to use. Only relevant in case not all available years should be used.
catchment	Character used to create the folder in which the resulting preprocessed training and test data sets get stored.

Examples

## Not run: 
data(test_catchment)
wt_preprocess(test_catchment)
train_data <- feather::read_feather("test_catchment/train_data.feather")
test_data <- feather::read_feather("test_catchment/test_data.feather")

## End(Not run)

MoritzFeigl/wateRtemp documentation built on Sept. 6, 2021, 6:58 a.m.