| HailoInference_async_multimodel | index /home/bedrock/Desktop/project/AlertWatch/Production/detector/inference/HailoInference_async_multimodel.py |
Hailo Inference Module
This module provides functionality to perform asynchronous inference using the Hailo platform.
It supports running multiple models concurrently, configuring network groups, and handling inference pipelines.
Classes:
- HailoInferenceAsyncMultiModel: Manages multiple inference models on a Hailo device.
Functions:
- initialize_inference_models: Initializes face detection and face landmarks models for inference.
Dependencies:
- hailo_platform: For Hailo device and model configuration.
- numpy: For numerical operations and input data manipulation.
- subprocess: For interacting with external commands (e.g., `hailortcli`).
Usage:
Use the `initialize_inference_models` function to set up models for face detection and landmarks inference.
Author:
Lior Jigalo
License:
MIT
| Modules | ||||||
| ||||||
| Classes | ||||||||||
|
| ||||||||||
| Functions | ||
| ||