 |
ImFusion C++ SDK 4.5.0
|
|
ImFusion C++ SDK 4.5.0
|
#include <TorchPlugin/include/ImFusion/Torch/TorchEngine.h>
Main wrapper around a Torch neural network performing a prediction on an image. More...
Main wrapper around a Torch neural network performing a prediction on an image.
Public Member Functions | |
| TorchEngine (const Properties &properties) | |
| Constructor from properties Note: the creation of the correct properties is typically taken care by the ModelConfiguration object, via the method 'getEngineProperties()'. | |
| ML::DataItem | predict (const ML::DataItem &input) override |
| Compute predictons of all classes for multiple images. | |
| std::vector< ML::ExecutionProvider > | availableProviders () const override |
| Returns which provider is supported by the engine and available on the host. | |
| std::optional< ML::ExecutionProvider > | provider () const override |
| Returns which provider is used by the engine to run the model. | |
| Public Member Functions inherited from ImFusion::ML::Engine | |
| Engine (const std::string &name) | |
| std::string | name () const |
| Returns the name of the operation as defined in the factories. | |
| virtual void | configure (const Properties *properties) override |
| Configure the Engine according to the properties. | |
| virtual void | configuration (Properties *properties) const override |
| Serialize the current object configuration into the given Properties object. | |
| virtual bool | isIdentical (const Engine *other) const |
| Check whether this engine instance is the same as other. | |
| virtual void | checkInputFields (const DataItem &input) |
| Check that inputs specified in the config are present in the input item. | |
| virtual void | checkOutputFields (const DataItem &input) |
| Check that outputs specified in the config are present in the predicted item. | |
| std::optional< ByteBuffer > | loadModelArtifact () const |
| Convenience function to load the model artifact specified in p_modelFile either from the model file or from a resource repository. | |
| Public Member Functions inherited from ImFusion::Configurable | |
| virtual void | configureDefaults () |
| Retrieve the properties of this object, replaces values with their defaults and sets it again. | |
| void | registerParameter (ParameterBase *param) |
| Register the given Parameter or SubProperty, so that it will be configured during configure()/configuration(). | |
| void | unregisterParameter (const ParameterBase *param) |
| Remove the given Parameter or SubProperty from the list of registered parameters. | |
| Configurable (const Configurable &rhs) | |
| Configurable (Configurable &&rhs) noexcept | |
| Configurable & | operator= (const Configurable &) |
| Configurable & | operator= (Configurable &&) noexcept |
Public Attributes | |
| Parameter< bool > | p_squeezeInput = {"EngineSqueezeInput", false, this} |
| Run squeeze on input tensor before running the model Squeezing will remove the batch dimension if tensor is single batch. | |
| Parameter< bool > | p_disableJITRecompile = {"DisableJITRecompile", false, this} |
| Disable JIT recompilation only for the next forward pass of the model. | |
| Parameter< std::string > | p_autocast = {"Autocast", "None", this} |
| Configure autocast precision mode for improved performance. | |
| Public Attributes inherited from ImFusion::ML::Engine | |
| Parameter< std::string > | p_modelFile = {"ModelFile", "", nullptr} |
| Parameters shared by all engines. | |
| Parameter< bool > | p_forceCPU = {"ForceCPU", false, nullptr} |
| Whether to force run on CPU even if GPU is present. | |
| Parameter< std::vector< std::string > > | p_inputFields = {"EngineInputFields", std::vector<std::string>{}, nullptr} |
| Input field names expected by the model run by the engine. | |
| Parameter< std::vector< std::string > > | p_outputFields = {"EngineOutputFields", std::vector<std::string>{}, nullptr} |
| Output field names returned by the model run by the engine. | |
| Parameter< std::vector< std::string > > | p_ignoreOutputFields = {"EngineOutputFieldsToIgnore", std::vector<std::string>{}, nullptr} |
| Output field names to be ignored by the engine. | |
| Parameter< std::string > | p_version = {"Version", "", nullptr} |
| Version of the engine used to train and save the model. | |
| Parameter< bool > | p_logInputToDataLogger = {"LogInputToDataLogger", false, this} |
When true, each engine input SharedImageSet is sent to the Data Logger. | |
| Public Attributes inherited from ImFusion::Configurable | |
| Signal | signalParametersChanged |
| Emitted whenever one of the registered Parameters' or SubPropertys' signalValueChanged signal was emitted. | |
Protected Member Functions | |
| bool | loadModel () |
| Load model architecture and weights from file, and return whether it was successful This method has to be called before any prediction. | |
| void | clearCache () |
| Clear the GPU cache allocated by torch. | |
| void | onModelFileChanged () override |
| Function that gets called when the model file of the engine changes. | |
| void | onForceCPUChanged () override |
| Function that gets called when the forceCPU flag of the engine changes. | |
| bool | useCuda () const |
| Returns whether the engine and model use CUDA. | |
| bool | useMPS () const |
| Returns whether the engine and model use MPS (Apple). | |
| Protected Member Functions inherited from ImFusion::ML::Engine | |
| virtual void | init (const Properties &properties) |
| Configures the engine from properties + connect signals. | |
| void | logInputToDataLogger (const DataItem &input) const |
Sends all SISBasedElement inputs to the Data Logger when p_logInputToDataLogger is enabled. | |
| virtual void | connectSignals () |
| Connect the signals like onModelFileChanged, onForceCPUChanged, ... (We do not do this automatically because children class need to call this after they register their parameters). | |
Protected Attributes | |
| std::unique_ptr< TorchModule > | m_torchModule |
| bool | m_hasCuda = false |
| bool | m_cudaDeviceAvailable = false |
| std::optional< int > | m_cudaDeviceIndex |
| bool | m_hasMPS = false |
| Protected Attributes inherited from ImFusion::ML::Engine | |
| EngineConfiguration | m_config |
| Configuration object. | |
| std::string | m_name = "" |
| Name of the engine interface (e.g. "torch"). | |
| Protected Attributes inherited from ImFusion::Configurable | |
| std::vector< Param > | m_params |
| List of all registered Parameter and SubProperty instances. | |
Additional Inherited Members | |
| Static Public Member Functions inherited from ImFusion::ML::Engine | |
| static std::vector< std::string > | availableEngines (EngineLanguage language=EngineLanguage::Any) |
| List all registered inference engines for a given implementation language. | |
| static bool | isRegistered (const std::string &engineName, bool quiet=true, EngineLanguage language=EngineLanguage::Any) |
| Check if an engine is registered for the selected implementation language. | |
| static std::unique_ptr< Engine > | create (const std::string &engineName, const Properties &properties, EngineLanguage language=EngineLanguage::Any) |
| Factory function to create an engine for the selected implementation language. | |
|
explicit |
Constructor from properties Note: the creation of the correct properties is typically taken care by the ModelConfiguration object, via the method 'getEngineProperties()'.
If the some required parameters is missing, it throws. In this case no extra parameter wrt ML::Engine is needed, so everything is set up in the base class
|
overridevirtual |
Compute predictons of all classes for multiple images.
Implements ImFusion::ML::Engine.
|
overridevirtual |
Returns which provider is supported by the engine and available on the host.
Implements ImFusion::ML::Engine.
|
overridevirtual |
Returns which provider is used by the engine to run the model.
Implements ImFusion::ML::Engine.
|
overrideprotectedvirtual |
Function that gets called when the model file of the engine changes.
Reimplemented from ImFusion::ML::Engine.
|
overrideprotectedvirtual |
Function that gets called when the forceCPU flag of the engine changes.
Reimplemented from ImFusion::ML::Engine.
| Parameter<std::string> ImFusion::TorchEngine::p_autocast = {"Autocast", "None", this} |
Configure autocast precision mode for improved performance.
Autocast automatically uses lower precision (FP16/BF16) for operations that can benefit from it while maintaining FP32 for operations that need higher precision.
Supported Values:
Example YAML Configuration: