These propagate Batches sampled from a DataSet using a Sampler through a Module in order to evaluate a Loss, provide Feedback or train the model :
- Propagator : abstract class;
Abstract Class for propagating a sampling distribution (a Sampler) through a Module. A Propagator can be sub-classed to build task-tailored training or evaluation algorithms.
A Propagator constructor which takes key-value arguments:
lossis a Criterion which the Model output will need to evaluate or minimize.
callbackis a user-defined function(model, report) that does things like update the
modelparameters, gather statistics, decay learning rate, etc.
epoch_callbackis a user-defined function(model, report) that is called between epochs. Typically used for learning rate decay and such;
sampleris, you guessed it, a Sampler instance which iterates through a DataSet. Defaults to
observeris an Observer instance that is informed when an event occurs.
feedbackis a Feedback instance that takes Model input, output and targets as input to provide I/O feedback to the user or system.
progressis a boolean that, when true, displays the progress of examples seen in the epoch. Defaults to
statsis a boolean for displaying statistics. Defaults to
Evaluates a Model on a