Class representing the settings of a Deep Learning benchmark. More...


Public Member Functions | |
| EDeepLearningBenchmarkSettings () | |
| Constructor. | |
| EDeepLearningBenchmarkSettings (eVision.EasyDeepLearning.EDeepLearningBenchmarkSettings other) | |
| Constructor. | |
| virtual void | Load (string filePath) |
| Loads a EDeepLearningBenchmarkSettings object. The given ESerializer must have been created for reading. | |
| virtual void | Load (eVision.ESerializer serializer) |
| Loads a EDeepLearningBenchmarkSettings object. The given ESerializer must have been created for reading. | |
| virtual void | Save (string filePath) |
| Saves the EDeepLearningBenchmarkSettings object. The given ESerializer must have been created for writing. | |
| virtual void | Save (eVision.ESerializer serializer) |
| Saves the EDeepLearningBenchmarkSettings object. The given ESerializer must have been created for writing. | |
Properties | |
| string | Name [get, set] |
| Name of the benchmark. | |
| string | Engine [get, set] |
| Deep Learning engine to use for the benchmark. | |
| eVision.EasyDeepLearning.EDeepLearningDevice | Device [get, set] |
| Device to use for the benchmark. | |
| eVision.EasyDeepLearning.EDeepLearningInferencePrecision | InferencePrecision [get, set] |
| Inference precision to use for the benchmark. | |
| int | NumExternalThreads [get, set] |
| Number of external threads. | |
| int | NumInternalThreads [get, set] |
| Number of internal threads. | |
| float | CameraSpeed [get, set] |
| Camera speed to simulate. If equal to 0, the benchmark will produce images as fast as possible. | |
| int | BatchSize [get, set] |
| Batch size. | |
| int | NumInferences [get, set] |
| Number of iterations to use during the benchmark. | |
| bool | InternalResizeDisabled [get, set] |
| Whether to disable the internal resize operation performed by the Deep Learning tools. When disabled, the benchmark will generate images at the optimal size for the tool. | |
| bool | OptimizedBatchSize [get, set] |
| Whether to enable the auto setup of the batch size. | |
Class representing the settings of a Deep Learning benchmark.
License(s): EasyClassify or EasySegment or EasyLocate
| eVision.EasyDeepLearning.EDeepLearningBenchmarkSettings.EDeepLearningBenchmarkSettings | ( | ) |
Constructor.
| eVision.EasyDeepLearning.EDeepLearningBenchmarkSettings.EDeepLearningBenchmarkSettings | ( | eVision.EasyDeepLearning.EDeepLearningBenchmarkSettings | other | ) |
Constructor.
| other | The other EDeepLearningBenchmarkSettings that we will copy into this one. |
|
virtual |
Loads a EDeepLearningBenchmarkSettings object. The given ESerializer must have been created for reading.
| serializer | Pointer to the ESerializer created for reading. |
|
virtual |
Loads a EDeepLearningBenchmarkSettings object. The given ESerializer must have been created for reading.
| filePath | File path. |
|
virtual |
Saves the EDeepLearningBenchmarkSettings object. The given ESerializer must have been created for writing.
| serializer | Pointer to the ESerializer created for writing. |
|
virtual |
Saves the EDeepLearningBenchmarkSettings object. The given ESerializer must have been created for writing.
| filePath | File path. |
|
getset |
Batch size.
|
getset |
Camera speed to simulate. If equal to 0, the benchmark will produce images as fast as possible.
Camera speed higher than 500 frame per seconds will lead to unreliable results.
|
getset |
Device to use for the benchmark.
|
getset |
Deep Learning engine to use for the benchmark.
|
getset |
Inference precision to use for the benchmark.
|
getset |
Whether to disable the internal resize operation performed by the Deep Learning tools. When disabled, the benchmark will generate images at the optimal size for the tool.
|
getset |
Name of the benchmark.
|
getset |
Number of external threads.
|
getset |
Number of iterations to use during the benchmark.
|
getset |
Number of internal threads.
|
getset |
Whether to enable the auto setup of the batch size.