Overview of Inference Engine Plugin Library¶
The plugin architecture of the Inference Engine allows to develop and plug independent inference solutions dedicated to different devices. Physically, a plugin is represented as a dynamic library exporting the single
CreatePluginEngine function that allows to create a new plugin instance.
Inference Engine Plugin Library¶
Inference Engine plugin dynamic library consists of several main components:
Provides information about devices of a specific type.
Can create an executable network instance which represents a Neural Network backend specific graph structure for a particular device in opposite to the InferenceEngine::ICNNNetwork interface which is backend-independent.
Can import an already compiled graph structure from an input stream to an executable network object.
Is an execution configuration compiled for a particular device and takes into account its capabilities.
Holds a reference to a particular device and a task executor for this device.
Can create several instances of Inference Request.
Can export an internal backend specific graph structure to an output stream.
Runs an inference pipeline serially.
Can extract performance counters for an inference pipeline execution profiling.
Wraps the Inference Request class and runs pipeline stages in parallel on several task executors based on a device-specific pipeline structure.
This documentation is written based on the
Template plugin, which demonstrates plugin development details. Find the complete code of the
Template, which is fully compilable and up-to-date, at
<dldt source dir>/docs/template_plugin.