MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T > Class Template Reference

The TrainerRNNSimple implements a very simple RNN trainer inspired by adepierre's GitHub site referenced. More...

Inheritance diagram for MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >:

Public Member Functions
	TrainerRNNSimple (MyCaffeControl< T > mycaffe, PropertySet properties, CryptoRandom random, IxTrainerCallback icallback, BucketCollection rgVocabulary)
	The constructor. More...
void	Dispose ()
	Releases all resources used. More...
bool	Initialize ()
	Initialize the trainer. More...
bool	Shutdown (int nWait)
	Shutdown the trainer. More...
float[]	Run (int nN, PropertySet runProp)
	Run a single cycle on the environment after the delay. More...
byte[]	Run (int nN, PropertySet runProp, out string type)
	Run a single cycle on the environment after the delay. More...
bool	Test (int nN, ITERATOR_TYPE type)
	Run the test cycle - currently this is not implemented. More...
bool	Train (int nN, ITERATOR_TYPE type, TRAIN_STEP step)
	Train the network using a modified PG training algorithm optimized for GPU use. More...

Detailed Description

The TrainerRNNSimple implements a very simple RNN trainer inspired by adepierre's GitHub site referenced.

See also

1. The Unreasonable Effectiveness of Recurrent Neural Networks, by Andrej Karpathy, 2015, Github.io

2. karpathy/char-rnn sample.lua

3. GitHub: adepierre/caffe-char-rnn, by adepierre, 2017, Github

4. MyCaffe: A Complete C# Re-Write of Caffe with Reinforcement Learning by D. Brown, 2018, arXiv

Definition at line 28 of file TrainerRNNSimple.cs.

Constructor & Destructor Documentation

TrainerRNNSimple()

MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >.TrainerRNNSimple	(	MyCaffeControl< T >	mycaffe,
		PropertySet	properties,
		CryptoRandom	random,
		IxTrainerCallback	icallback,
		BucketCollection	rgVocabulary
	)

The constructor.

Parameters

mycaffe	Specifies the MyCaffeControl to use for learning and prediction.
properties	Specifies the property set containing the key/value pairs of property settings.
random	Specifies the random number generator to use.
icallback	Specifies the callback for parent notifications and queries.
rgVocabulary	Specifies the vocabulary to use.

The 'bUsePreloadData' parameter has been replaced with a property 'UsePreLoadData'. When this property does not exist, the 'UsePreLoadData' defaults to 'true'.

Definition at line 51 of file TrainerRNNSimple.cs.

Member Function Documentation

Dispose()

void MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >.Dispose

(

)

Releases all resources used.

Definition at line 64 of file TrainerRNNSimple.cs.

Initialize()

bool MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >.Initialize

(

)

Initialize the trainer.

Returns

Returns true.

Implements MyCaffe.trainers.IxTrainer.

Definition at line 72 of file TrainerRNNSimple.cs.

Run() [1/2]

float[] MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >.Run	(	int	nN,
		PropertySet	runProp
	)

Run a single cycle on the environment after the delay.

Parameters

nN	specifies the number of samples to run.
runProp	Optionally specifies properties to use when running.

Returns

The results of the run containing the action are returned.

Implements MyCaffe.trainers.IxTrainerRNN.

Definition at line 136 of file TrainerRNNSimple.cs.

Run() [2/2]

byte[] MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >.Run	(	int	nN,
		PropertySet	runProp,
		out string	type
	)

Run a single cycle on the environment after the delay.

Parameters

nN	Specifies the number of samples to run.
runProp	Optionally specifies properties to use when running.
type	Returns the data type contained in the byte stream.

Returns

The results of the run containing the action are returned as a byte stream.

Implements MyCaffe.trainers.IxTrainerRNN.

Definition at line 149 of file TrainerRNNSimple.cs.

Shutdown()

bool MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >.Shutdown

(

int

nWait

)

Shutdown the trainer.

Parameters

nWait

Specifies a wait in ms. for the shutdown to complete.

Returns

Returns true.

Implements MyCaffe.trainers.IxTrainer.

Definition at line 96 of file TrainerRNNSimple.cs.

Test()

bool MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >.Test	(	int	nN,
		ITERATOR_TYPE	type
	)

Run the test cycle - currently this is not implemented.

Parameters

nN	Specifies the number of iterations (based on the ITERATION_TYPE) to run, or -1 to ignore.
type	Specifies the iteration type (default = ITERATION).

Returns

A value of true is returned when handled, false otherwise.

Implements MyCaffe.trainers.IxTrainer.

Definition at line 162 of file TrainerRNNSimple.cs.

Train()

bool MyCaffe.trainers.rnn.simple.TrainerRNNSimple< T >.Train	(	int	nN,
		ITERATOR_TYPE	type,
		TRAIN_STEP	step
	)

Train the network using a modified PG training algorithm optimized for GPU use.

Parameters

nN	Specifies the number of iterations (based on the ITERATION_TYPE) to run, or -1 to ignore.
type	Specifies the iteration type (default = ITERATION).
step	Specifies the stepping mode to use (when debugging).

Returns

A value of true is returned when handled, false otherwise.

Implements MyCaffe.trainers.IxTrainer.

Definition at line 174 of file TrainerRNNSimple.cs.

---

The documentation for this class was generated from the following file:

• C:/Data/Data/SS_Projects/Intelligence/GitHub/MyCaffe/MyCaffe.trainers/rnn.simple/TrainerRNNSimple.cs