lib.sedna.service.multi_edge_inference

Subpackages

• lib.sedna.service.multi_edge_inference.interface
  • lib.sedna.service.multi_edge_inference.interface.detection_endpoint
  • lib.sedna.service.multi_edge_inference.interface.fe_endpoint
  • lib.sedna.service.multi_edge_inference.interface.reid_endpoint
• lib.sedna.service.multi_edge_inference.server
  • lib.sedna.service.multi_edge_inference.server.detection
  • lib.sedna.service.multi_edge_inference.server.feature_extraction
  • lib.sedna.service.multi_edge_inference.server.reid

Package Contents

Classes

Detection	Endpoint to trigger the Object Tracking component
FE	Endpoint to trigger the Feature Extraction
ReID_Endpoint	Endpoint to trigger the ReID
DetectionServer	REST api server for object detection component
FEServer	rest api server for feature extraction
ReIDServer	REST api server for reid

class lib.sedna.service.multi_edge_inference.Detection(service_name, version='', ip='127.0.0.1', port='8080', protocol='http')

Endpoint to trigger the Object Tracking component

check_server_status()

transmit(data, **kwargs)

Transfer enriched tracking object to video analytics job

update_service(data, **kwargs)

class lib.sedna.service.multi_edge_inference.FE(service_name, version='', ip='127.0.0.1', port='8080', protocol='http')

Endpoint to trigger the Feature Extraction

check_server_status()

transmit(data, **kwargs)

Transfer feature vector to FE worker

get_target_features(data, **kwargs)

Send target images to FE service and receive back the ReID features

update_service(data, **kwargs)

class lib.sedna.service.multi_edge_inference.ReID_Endpoint(service_name, version='', ip='127.0.0.1', port='8080', protocol='http')

Endpoint to trigger the ReID

check_server_status()

transmit(data: sedna.core.multi_edge_inference.data_classes.DetTrackResult, **kwargs)

Transfer feature vector to ReID worker

class lib.sedna.service.multi_edge_inference.DetectionServer(model, service_name, ip: str = '127.0.0.1', port: int = 8080, max_buffer_size: int = 1004857600, workers: int = 1)

Bases: sedna.service.server.base.BaseServer

REST api server for object detection component

start()

status(request: fastapi.Request)

async video_analytics(request: fastapi.Request)

async update_service(request: fastapi.Request)

class lib.sedna.service.multi_edge_inference.FEServer(model, service_name, ip: str = '127.0.0.1', port: int = 8080, max_buffer_size: int = 1004857600, workers: int = 1)

Bases: sedna.service.server.base.BaseServer

rest api server for feature extraction

start()

status(request: fastapi.Request)

async feature_extraction(request: fastapi.Request)

async get_target_features(request: fastapi.Request)

async update_service(request: fastapi.Request)

class lib.sedna.service.multi_edge_inference.ReIDServer(model, service_name, ip: str = '127.0.0.1', port: int = 8080, max_buffer_size: int = 104857600, workers: int = 1)

Bases: sedna.service.server.base.BaseServer

REST api server for reid

start()

status(request: fastapi.Request)

async reid(request: fastapi.Request)