train_eval_funs

Description

Basic methods for training and evaluation of models with callback calls.

Functions

device_of(model)	Return the device of the given pytorch model.
evaluate(model, kpi_fns, val_loader[, ...])	Evaluate the model wrt loss and metric_fns on the test data.
loader([data, batch_size, shuffle, device, ...])	Prepare and return a torch data loader with device-dependent multi-processing settings.
predict_laplace(model, data[, device, var0])	Performs prediction with probit approximation of the Bayesian posterior.
second_stage_train(model, nll_fn, ...)	Evaluate the model wrt loss and metric_fns on the test data.
train_one_epoch(model, loss_fn, metric_fns, ...)	Train for one epoch, evaluate, and return history and test results.