THIS RELEASE SHOULD NOT BE USED.
The model file here is NOT COMPATIBLE with the current version of edgetpu_tfl.py and it runs slower than the current version of the model from release v1.0
yolov9-s-relu6-10epoch-17class_320_int8_edgetpu.zip
Then unzip the file to get the TFLite file that Frigate can work with.
Model weights file for YOLO v9 "s" with resolution 320x320 compiled to run on Google Coral using the modified edgetpu detector plug-in code
17 classes of objects can be detected (a subset of COCO's 80 classes)
Model performance after 10 epochs of fine-tuning.
epoch, train/box_loss, train/cls_loss, train/dfl_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/cls_loss, val/dfl_loss, x/lr0, x/lr1, x/lr2
9, 1.47, 1.3719, 1.649, 0.75295, 0.65368, 0.72426, 0.54616, 0, 0, 0, 0.00208, 0.00208, 0.00208
Additional epochs might improve the performance, and lead to future updated releases here.