Trainer (Deep Java Library 0.2.0 API specification)

All Superinterfaces:

java.lang.AutoCloseable
```
public interface Trainer
extends java.lang.AutoCloseable
```
An interface that represents a single training iteration.
Trainer provides an easy, and manageable interface for training. Trainer is not thread-safe.
See the tutorials on:

Method Summary

All Methods Instance Methods Abstract Methods Default Methods
Modifier and Type	Method and Description
`void`	`close()`
`NDList`	`forward(NDList input)` Applies the forward function of the model once on the given input `NDList`.
`Loss`	`getLoss()` Gets the training `Loss` function of the trainer.
`NDManager`	`getManager()` Gets the `NDManager` from the model.
`Model`	`getModel()` Returns the model used to create this trainer.
`<T extends TrainingMetric> T`	`getTrainingMetric(java.lang.Class<T> clazz)` Gets the training `TrainingMetric` that is an instance of the given `Class`.
`Loss`	`getValidationLoss()` Gets the validation `Loss` function of the trainer.
`<T extends TrainingMetric> T`	`getValidationMetric(java.lang.Class<T> clazz)` Gets the validation `TrainingMetric` that is an instance of the given `Class`.
`void`	`initialize(Shape... shapes)` Initializes the `Model` that the `Trainer` is going to train.
`default java.lang.Iterable<Batch>`	`iterateDataset(Dataset dataset)` Fetches an iterator that can iterate through the given `Dataset`.
`GradientCollector`	`newGradientCollector()` Returns a new instance of `GradientCollector`.
`void`	`resetTrainingMetrics()` Resets each of the training metrics and loss to its respective initial value.
`void`	`setMetrics(Metrics metrics)` Attaches a Metrics param to use for benchmark.
`void`	`setTrainingListener(TrainingListener listener)` Sets a `TrainingListener` to the `Trainer`.
`void`	`step()` Updates all of the parameters of the model once.
`void`	`trainBatch(Batch batch)` Trains the model with one iteration of the given `Batch` of data.
`void`	`validateBatch(Batch batch)` Validates the given batch of data.

- Method Detail
  - initialize
```
void initialize(Shape... shapes)
```
    Initializes the Model that the Trainer is going to train.
    
    Parameters:
    
    shapes - an array of Shape of the inputs
  - iterateDataset
```
default java.lang.Iterable<Batch> iterateDataset(Dataset dataset)
```
    Fetches an iterator that can iterate through the given Dataset.
    
    Parameters:
    
    dataset - the dataset to iterate through
    
    Returns:
    
    an Iterable of Batch that contains batches of data from the dataset
  - newGradientCollector
```
GradientCollector newGradientCollector()
```
    Returns a new instance of GradientCollector.
    
    Returns:
    
    a new instance of GradientCollector
  - trainBatch
```
void trainBatch(Batch batch)
```
    Trains the model with one iteration of the given Batch of data.
    
    Parameters:
    
    batch - a Batch that contains data, and its respective labels
  - forward
```
NDList forward(NDList input)
```
    Applies the forward function of the model once on the given input NDList.
    
    Parameters:
    
    input - the input NDList
    
    Returns:
    
    the output of the forward function
  - validateBatch
```
void validateBatch(Batch batch)
```
    Validates the given batch of data.
    During validation, the loss and training metrics are computed, but gradients aren't computed, and parameters aren't updated.
    
    Parameters:
    
    batch - a Batch of data
  - step
```
void step()
```
    Updates all of the parameters of the model once.
  - setMetrics
```
void setMetrics(Metrics metrics)
```
    Attaches a Metrics param to use for benchmark.
    
    Parameters:
    
    metrics - the Metrics class
  - setTrainingListener
```
void setTrainingListener(TrainingListener listener)
```
    Sets a TrainingListener to the Trainer.
    
    Parameters:
    
    listener - the TrainingListener to be set
  - resetTrainingMetrics
```
void resetTrainingMetrics()
```
    Resets each of the training metrics and loss to its respective initial value.
  - getLoss
```
Loss getLoss()
```
    Gets the training Loss function of the trainer.
    
    Returns:
    
    the Loss function
  - getValidationLoss
```
Loss getValidationLoss()
```
    Gets the validation Loss function of the trainer.
    
    Returns:
    
    the validation Loss function
  - getModel
```
Model getModel()
```
    Returns the model used to create this trainer.
    
    Returns:
    
    the model associated with this trainer
  - getTrainingMetric
```
<T extends TrainingMetric> T getTrainingMetric(java.lang.Class<T> clazz)
```
    Gets the training TrainingMetric that is an instance of the given Class.
    
    Type Parameters:
    
    T - the type of the training metric
    
    Parameters:
    
    clazz - the Class of the TrainingMetric sought
    
    Returns:
    
    the training metric requested
  - getValidationMetric
```
<T extends TrainingMetric> T getValidationMetric(java.lang.Class<T> clazz)
```
    Gets the validation TrainingMetric that is an instance of the given Class.
    
    Type Parameters:
    
    T - the type of the validation metric
    
    Parameters:
    
    clazz - the Class of the TrainingMetric sought
    
    Returns:
    
    the validation metric requested
  - getManager
```
NDManager getManager()
```
    Gets the NDManager from the model.
    
    Returns:
    
    the NDManager
  - close
```
void close()
```
    Specified by:
    
    close in interface java.lang.AutoCloseable

Interface Trainer

Method Summary

Method Detail

initialize

iterateDataset

newGradientCollector

trainBatch

forward

validateBatch

step

setMetrics

setTrainingListener

resetTrainingMetrics

getLoss

getValidationLoss

getModel

getTrainingMetric

getValidationMetric

getManager

close