BenchmarkResult

AI-generated Key Takeaways

• BenchmarkResult is a data class storing performance and accuracy data from TFLite model benchmarking.

• It includes metrics like inference time, initialization time, memory usage, and accuracy validation results.

• The class provides access to model output and a list of accuracy metrics through its methods.

• BenchmarkResult contains two nested classes, BenchmarkMetric for accuracy and InferenceOutput for model output.

• Developers can use BenchmarkResult to evaluate the performance and accuracy of different TFLite model configurations.

public abstract class BenchmarkResult extends Object

Data class that contains BenchmarkResult.

Nested Class Summary

class	BenchmarkResult.BenchmarkMetric		Accuracy metrics.
class	BenchmarkResult.InferenceOutput		Model output of one tensor.

Public Method Summary

abstract List<BenchmarkResult.InferenceOutput>	actualOutput() Returns model output of running with this config.
abstract boolean	hasPassedAccuracyCheck() Returns whether accuracy validation check has passed.
abstract List<Long>	inferenceTimeMicros() Returns inference time of each sample input, in microseconds.
abstract List<Long>	initializationTimeMicros() Returns initialization time of each sample input, in microseconds.
abstract int	maxMemoryKb() Returns max memory used of all sample inputs, in kilobytes.
abstract List<BenchmarkResult.BenchmarkMetric>	metrics() Returns the list of accuracy metrics.

Inherited Method Summary

From class java.lang.Object

Object	clone()
boolean	equals(Object arg0)
void	finalize()
final Class<?>	getClass()
int	hashCode()
final void	notify()
final void	notifyAll()
String	toString()
final void	wait(long arg0, int arg1)
final void	wait(long arg0)
final void	wait()

Public Methods

public abstract List<BenchmarkResult.InferenceOutput> actualOutput ()

Returns model output of running with this config. Each BenchmarkResult.InferenceOutput maps to one output tensor.

public abstract boolean hasPassedAccuracyCheck ()

Returns whether accuracy validation check has passed.

public abstract List<Long> inferenceTimeMicros ()

Returns inference time of each sample input, in microseconds.

public abstract List<Long> initializationTimeMicros ()

Returns initialization time of each sample input, in microseconds.

public abstract int maxMemoryKb ()

Returns max memory used of all sample inputs, in kilobytes.

public abstract List<BenchmarkResult.BenchmarkMetric> metrics ()

Returns the list of accuracy metrics.