reexports: Default compile function for ProcessTransformer model

In processpredictR: Process Prediction

Default compile function for ProcessTransformer model

Description

These objects are imported from other packages. Follow the links below to see their documentation.

keras

compile, evaluate, fit

stats

predict

Arguments

optimizer	Default optimizer for ppred_model
loss	Default loss for ppred_model
metrics	Default metrics for ppred_model
train_data	A training dataset
batch_size	A batch size
num_epochs	A number of epochs
verbose	A verbose
callbacks	list: A list of callbacks. keras default is NULL, but can be adjusted (ex. keras::callback_csv_logger(filename = paste("log_", object$task)), #or NULL keras::callback_tensorboard())
shuffle	logical (default TRUE): If TRUE shuffles the data
validation_split	A ratio to split on
object	ppred_model (default NULL): ProcessTransformer model of class ppred_model.
test_data	ppred_examples_df (default NULL): preprocessed test data.
append	logical (default FALSE): if TRUE, returns a passed data.frame with predicted values.
...	Additional arguments

See Also

See keras::fit() for documentation of parameters

processpredictR documentation built on Jan. 18, 2023, 1:09 a.m.