vietanhdev / anylabeling

Effortless AI-assisted data labeling with AI support from YOLO, Segment Anything, MobileSAM!!
https://anylabeling.nrl.ai
GNU General Public License v3.0
2k stars 218 forks source link

config.yaml enhancement for custom models #172

Open mahesh11T opened 4 months ago

mahesh11T commented 4 months ago

YOLOv8:

type: yolov8
name: yolov8m-r20230415
display_name: YOLOv8m Ultralytics
model_path: yolov8m.onnx
confidence_threshold: 0.45
input_height: 640
input_width: 640
nms_threshold: 0.45
score_threshold: 0.5
classes:
- person
- bicycle

proposed

if we want to use any other onnx model apart from yolo, we should be able to give path to code where model can be loaded and also inference code for bbox.

custom_onnx_model

load_model_code: load_model.py
inference_model_code: inference_model.py
type: custom_model
name: yolov8m-r20230415
display_name: my_custom_model
model_path: custom.onnx
confidence_threshold: 0.45
input_height: 640
input_width: 640
nms_threshold: 0.45
score_threshold: 0.5
classes:
- abc
- 124