loadModel | R Documentation |
Generates a new model of class loadModel (loadModel-class
).
loadModel(inner.fit.function, pred.format, data, metadata,
y.trans.function = NULL, retrans.function = exp, store = c("data",
"fitting.function"))
inner.fit.function |
function that accepts one argument, a training data
frame, and returns a [re]fitted statistical model, such as an |
pred.format |
character string that describes the output of retrans.function(predict(fit)) [NOT of predictSolute.loadModel(), which can do either!] |
data |
data.frame containing the initial training observations |
metadata |
An object of class |
y.trans.function |
function that accepts a vector of observed response values (e.g., concentrations or flux rates) and transforms them into the values on the left-hand side of the calibration formula. Because load models are frequently fit to log(y) ~ ..., the default is y.trans.function=log. |
retrans.function |
function that accepts a vector of predictions straight from the fitted model and retransforms them, if appropriate, to predictions in linear space and with units of concentration*flow. Because load models are frequently fit to log(y) ~ ..., the default is retrans.function=exp. After retrans.function has been applied to the predictions, the retransformed predictions will automatically undergo further units conversion from conc*flow to load.rate, according to the metadata elements conc.units, flow.units, and load.rate.units. |
store |
One or more character strings specifying which information to write within the model. Options are 'data': the original fitting data; 'fitting.function': a fitting function that can produce a new loadComp object from new data |
A fitted loadModel.
Other load.model.inits: loadComp
,
loadInterp
, loadLm
,
loadReg2
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.