public interface Trainer
extends java.lang.Runnable
ParallelWrapper
for handling training in multi core situations.Modifier and Type | Method and Description |
---|---|
void |
feedDataSet(org.nd4j.linalg.dataset.api.DataSet dataSet)
Train on a
DataSet |
void |
feedMultiDataSet(org.nd4j.linalg.dataset.api.MultiDataSet dataSet)
Train on a
MultiDataSet |
Model |
getModel()
THe current model for the trainer
|
boolean |
isRunning() |
void |
setUncaughtExceptionHandler(java.lang.Thread.UncaughtExceptionHandler handler)
Set the
Thread.UncaughtExceptionHandler
for this Trainer |
void |
shutdown()
Shutdown this worker
|
void |
start()
Start this trainer
|
void |
updateModel(Model model)
Update the current
Model
for the worker |
void |
waitTillRunning()
Block the main thread
till the trainer is up and running.
|
void feedMultiDataSet(@NonNull org.nd4j.linalg.dataset.api.MultiDataSet dataSet)
MultiDataSet
dataSet
- the data set to train onvoid feedDataSet(@NonNull org.nd4j.linalg.dataset.api.DataSet dataSet)
DataSet
dataSet
- the data set to train onModel getModel()
Model
for the workervoid updateModel(@NonNull Model model)
Model
for the workermodel
- the new model for this workerboolean isRunning()
void shutdown()
void waitTillRunning()
void setUncaughtExceptionHandler(java.lang.Thread.UncaughtExceptionHandler handler)
Thread.UncaughtExceptionHandler
for this Trainer
handler
- the handler for uncaught errorsvoid start()