org.deeplearning4j.ui.stats.api

Interface StatsReport

All Superinterfaces:

Persistable, java.io.Serializable

All Known Implementing Classes:

JavaStatsReport, SbeStatsReport

public interface StatsReport
extends Persistable

StatsReport: An interface for storing and serializing update information (such as scores, parameter histograms etc) for use in the StatsListener

Method Summary

Modifier and Type	Method and Description
java.util.List<java.io.Serializable>	getDataSetMetaData() Get the DataSet metadata, if any (null otherwise).
java.lang.String	getDataSetMetaDataClassName() Get the class
long[]	getDeviceCurrentBytes() Get device (GPU, etc) current bytes - may be null if no compute devices are present in the system
long[]	getDeviceMaxBytes() Get device (GPU, etc) maximum bytes - may be null if no compute devices are present in the system
double	getExamplesPerSecond() Get examples per second since the last report
java.util.List<Pair<java.lang.String,int[]>>	getGarbageCollectionStats() Get the garbage collection stats: Pair contains GC name and the delta count/time values
java.util.Map<java.lang.String,Histogram>	getHistograms(StatsType statsType) Get the histograms for all parameters, for a given StatsType (Parameters/Updates/Activations)
int	getIterationCount() Get the current iteration number
long	getJvmCurrentBytes() Get JVM memory - current bytes used
long	getJvmMaxBytes() Get JVM memory - max available bytes
java.util.Map<java.lang.String,java.lang.Double>	getLearningRates() Get the learning rates by parameter
java.util.Map<java.lang.String,java.lang.Double>	getMean(StatsType statsType) Get the mean values for each parameter for the given StatsType (Parameters/Updates/Activations)
java.util.Map<java.lang.String,java.lang.Double>	getMeanMagnitudes(StatsType statsType) Get the mean magnitude values for each parameter for the given StatsType (Parameters/Updates/Activations)
double	getMinibatchesPerSecond() Get the number of minibatches per second, since the last report
long	getOffHeapCurrentBytes() Get off-heap memory - current bytes used
long	getOffHeapMaxBytes() Get off-heap memory - max available bytes
double	getScore() Get the score at the current iteration
int	getStatsCollectionDurationMs() Get the number of millisecons required to calculate al of the stats.
java.util.Map<java.lang.String,java.lang.Double>	getStdev(StatsType statsType) Get the standard deviation values for each parameter for the given StatsType (Parameters/Updates/Activations)
long	getTotalExamples() Get total number of examples that have been processed since initialization
long	getTotalMinibatches() Get the total number of minibatches that have been processed since initialization
long	getTotalRuntimeMs() Get the total runtime since listener/model initialization
boolean	hasDataSetMetaData() Return whether any DataSet metadata is present or not
boolean	hasGarbageCollection() Return whether garbage collection information has been reported
boolean	hasHistograms(StatsType statsType) Return whether histograms have been reported, for the given stats type (Parameters, Updates, Activations)
boolean	hasLearningRates() Return whether the learning rates are present (have been reported)
boolean	hasMemoryUse() Return whether memory use has been reported
boolean	hasPerformance() Return whether performance stats (total time, total examples etc) have been reported
boolean	hasScore() Return whether the score is present (has been reported)
boolean	hasSummaryStats(StatsType statsType, SummaryType summaryType) Return whether the summary stats (mean, standard deviation, mean magnitudes) have been reported for the given stats type (Parameters, Updates, Activations)
void	reportDataSetMetaData(java.util.List<java.io.Serializable> dataSetMetaData, java.lang.Class<?> metaDataClass) Report any metadata for the DataSet
void	reportDataSetMetaData(java.util.List<java.io.Serializable> dataSetMetaData, java.lang.String metaDataClass) Report any metadata for the DataSet
void	reportGarbageCollection(java.lang.String gcName, int deltaGCCount, int deltaGCTime) Report Garbage collection stats
void	reportHistograms(StatsType statsType, java.util.Map<java.lang.String,Histogram> histogram) Report histograms for all parameters, for a given StatsType
void	reportIDs(java.lang.String sessionID, java.lang.String typeID, java.lang.String workerID, long timestamp)
void	reportIterationCount(int iterationCount) Report the current iteration number
void	reportLearningRates(java.util.Map<java.lang.String,java.lang.Double> learningRatesByParam) Report the learning rates by parameter
void	reportMean(StatsType statsType, java.util.Map<java.lang.String,java.lang.Double> mean) Report the mean values for each parameter, the given StatsType (Parameters/Updates/Activations)
void	reportMeanMagnitudes(StatsType statsType, java.util.Map<java.lang.String,java.lang.Double> meanMagnitudes) Report the mean magnitude values for each parameter for the given StatsType (Parameters/Updates/Activations)
void	reportMemoryUse(long jvmCurrentBytes, long jvmMaxBytes, long offHeapCurrentBytes, long offHeapMaxBytes, long[] deviceCurrentBytes, long[] deviceMaxBytes) Report the memory stats at this iteration
void	reportPerformance(long totalRuntimeMs, long totalExamples, long totalMinibatches, double examplesPerSecond, double minibatchesPerSecond) Report the performance stats (since the last report)
void	reportScore(double currentScore) Report model score at the current iteration
void	reportStatsCollectionDurationMS(int statsCollectionDurationMS) Report the number of milliseconds required to calculate all of the stats.
void	reportStdev(StatsType statsType, java.util.Map<java.lang.String,java.lang.Double> stdev) Report the standard deviation values for each parameter for the given StatsType (Parameters/Updates/Activations)

Methods inherited from interface org.deeplearning4j.api.storage.Persistable

decode, decode, decode, encode, encode, encode, encodingLengthBytes, getSessionID, getTimeStamp, getTypeID, getWorkerID

Method Detail

reportIDs

void reportIDs(java.lang.String sessionID,
               java.lang.String typeID,
               java.lang.String workerID,
               long timestamp)

reportIterationCount

void reportIterationCount(int iterationCount)

Report the current iteration number

getIterationCount

int getIterationCount()

Get the current iteration number

reportStatsCollectionDurationMS

void reportStatsCollectionDurationMS(int statsCollectionDurationMS)

Report the number of milliseconds required to calculate all of the stats. This is effectively the amount of listener overhead

getStatsCollectionDurationMs

int getStatsCollectionDurationMs()

Get the number of millisecons required to calculate al of the stats. This is effectively the amount of listener overhead.

reportScore

void reportScore(double currentScore)

Report model score at the current iteration

getScore

double getScore()

Get the score at the current iteration

reportLearningRates

void reportLearningRates(java.util.Map<java.lang.String,java.lang.Double> learningRatesByParam)

Report the learning rates by parameter

getLearningRates

java.util.Map<java.lang.String,java.lang.Double> getLearningRates()

Get the learning rates by parameter

reportMemoryUse

void reportMemoryUse(long jvmCurrentBytes,
                     long jvmMaxBytes,
                     long offHeapCurrentBytes,
                     long offHeapMaxBytes,
                     long[] deviceCurrentBytes,
                     long[] deviceMaxBytes)

Report the memory stats at this iteration

Parameters:

jvmCurrentBytes - Current bytes used by the JVM

jvmMaxBytes - Max bytes usable by the JVM (heap)

offHeapCurrentBytes - Current off-heap bytes used

offHeapMaxBytes - Maximum off-heap bytes

deviceCurrentBytes - Current bytes used by each device (GPU, etc). May be null if no devices are present

deviceMaxBytes - Maximum bytes for each device (GPU, etc). May be null if no devices are present

getJvmCurrentBytes

long getJvmCurrentBytes()

Get JVM memory - current bytes used

getJvmMaxBytes

long getJvmMaxBytes()

Get JVM memory - max available bytes

getOffHeapCurrentBytes

long getOffHeapCurrentBytes()

Get off-heap memory - current bytes used

getOffHeapMaxBytes

long getOffHeapMaxBytes()

Get off-heap memory - max available bytes

getDeviceCurrentBytes

long[] getDeviceCurrentBytes()

Get device (GPU, etc) current bytes - may be null if no compute devices are present in the system

getDeviceMaxBytes

long[] getDeviceMaxBytes()

Get device (GPU, etc) maximum bytes - may be null if no compute devices are present in the system

reportPerformance

void reportPerformance(long totalRuntimeMs,
                       long totalExamples,
                       long totalMinibatches,
                       double examplesPerSecond,
                       double minibatchesPerSecond)

Report the performance stats (since the last report)

Parameters:

totalRuntimeMs - Overall runtime since initialization

totalExamples - Total examples processed since initialization

totalMinibatches - Total number of minibatches (iterations) since initialization

examplesPerSecond - Examples per second since last report

minibatchesPerSecond - Minibatches per second since last report

getTotalRuntimeMs

long getTotalRuntimeMs()

Get the total runtime since listener/model initialization

getTotalExamples

long getTotalExamples()

Get total number of examples that have been processed since initialization

getTotalMinibatches

long getTotalMinibatches()

Get the total number of minibatches that have been processed since initialization

getExamplesPerSecond

double getExamplesPerSecond()

Get examples per second since the last report

getMinibatchesPerSecond

double getMinibatchesPerSecond()

Get the number of minibatches per second, since the last report

reportGarbageCollection

void reportGarbageCollection(java.lang.String gcName,
                             int deltaGCCount,
                             int deltaGCTime)

Report Garbage collection stats

Parameters:

gcName - Garbage collector name

deltaGCCount - Change in the total number of garbage collections, since last report

deltaGCTime - Change in the amount of time (milliseconds) for garbage collection, since last report

getGarbageCollectionStats

java.util.List<Pair<java.lang.String,int[]>> getGarbageCollectionStats()

Get the garbage collection stats: Pair contains GC name and the delta count/time values

reportHistograms

void reportHistograms(StatsType statsType,
                      java.util.Map<java.lang.String,Histogram> histogram)

Report histograms for all parameters, for a given StatsType

Parameters:

statsType - StatsType: Parameters, Updates, Activations

histogram - Histogram values for all parameters

getHistograms

java.util.Map<java.lang.String,Histogram> getHistograms(StatsType statsType)

Get the histograms for all parameters, for a given StatsType (Parameters/Updates/Activations)

Parameters:

statsType - Stats type (Params/updatse/activations) to get histograms for

Returns:

Histograms by parameter name, or null if not available

reportMean

void reportMean(StatsType statsType,
                java.util.Map<java.lang.String,java.lang.Double> mean)

Report the mean values for each parameter, the given StatsType (Parameters/Updates/Activations)

Parameters:

statsType - Stats type to report

mean - Map of mean values, by parameter

getMean

java.util.Map<java.lang.String,java.lang.Double> getMean(StatsType statsType)

Get the mean values for each parameter for the given StatsType (Parameters/Updates/Activations)

Parameters:

statsType - Stats type to get mean values for

Returns:

Map of mean values by parameter

reportStdev

void reportStdev(StatsType statsType,
                 java.util.Map<java.lang.String,java.lang.Double> stdev)

Report the standard deviation values for each parameter for the given StatsType (Parameters/Updates/Activations)

Parameters:

statsType - Stats type to report std. dev values for

stdev - Map of std dev values by parameter

getStdev

java.util.Map<java.lang.String,java.lang.Double> getStdev(StatsType statsType)

Get the standard deviation values for each parameter for the given StatsType (Parameters/Updates/Activations)

Parameters:

statsType - Stats type to get std dev values for

Returns:

Map of stdev values by parameter

reportMeanMagnitudes

void reportMeanMagnitudes(StatsType statsType,
                          java.util.Map<java.lang.String,java.lang.Double> meanMagnitudes)

Report the mean magnitude values for each parameter for the given StatsType (Parameters/Updates/Activations)

Parameters:

statsType - Stats type to report mean magnitude values for

meanMagnitudes - Map of mean magnitude values by parameter

reportDataSetMetaData

void reportDataSetMetaData(java.util.List<java.io.Serializable> dataSetMetaData,
                           java.lang.Class<?> metaDataClass)

Report any metadata for the DataSet

Parameters:

dataSetMetaData - MetaData for the DataSet

metaDataClass - Class of the metadata. Can be later retieved using getDataSetMetaDataClassName()

reportDataSetMetaData

void reportDataSetMetaData(java.util.List<java.io.Serializable> dataSetMetaData,
                           java.lang.String metaDataClass)

Report any metadata for the DataSet

Parameters:

dataSetMetaData - MetaData for the DataSet

metaDataClass - Class of the metadata. Can be later retieved using getDataSetMetaDataClassName()

getMeanMagnitudes

java.util.Map<java.lang.String,java.lang.Double> getMeanMagnitudes(StatsType statsType)

Get the mean magnitude values for each parameter for the given StatsType (Parameters/Updates/Activations)

Parameters:

statsType - Stats type to get mean magnitude values for

Returns:

Map of mean magnitude values by parameter

getDataSetMetaData

java.util.List<java.io.Serializable> getDataSetMetaData()

Get the DataSet metadata, if any (null otherwise). Note: due to serialization issues, this may in principle throw an unchecked exception related to class availability, serialization etc.

Returns:

List of DataSet metadata, if any.

getDataSetMetaDataClassName

java.lang.String getDataSetMetaDataClassName()

Get the class

Returns:

hasScore

boolean hasScore()

Return whether the score is present (has been reported)

hasLearningRates

boolean hasLearningRates()

Return whether the learning rates are present (have been reported)

hasMemoryUse

boolean hasMemoryUse()

Return whether memory use has been reported

hasPerformance

boolean hasPerformance()

Return whether performance stats (total time, total examples etc) have been reported

hasGarbageCollection

boolean hasGarbageCollection()

Return whether garbage collection information has been reported

hasHistograms

boolean hasHistograms(StatsType statsType)

Return whether histograms have been reported, for the given stats type (Parameters, Updates, Activations)

Parameters:

statsType - Stats type

hasSummaryStats

boolean hasSummaryStats(StatsType statsType,
                        SummaryType summaryType)

Return whether the summary stats (mean, standard deviation, mean magnitudes) have been reported for the given stats type (Parameters, Updates, Activations)

Parameters:

statsType - stats type (Parameters, Updates, Activations)

summaryType - Summary statistic type (mean, stdev, mean magnitude)

hasDataSetMetaData

boolean hasDataSetMetaData()

Return whether any DataSet metadata is present or not

Returns:

True if DataSet metadata is present