Bug confidence score prediction Yolov5

[CesareDavidePace]

Hi,
I noticed that if I set the confidence score to 0.1 on yolov5 when predicting with sliced, the results are different than predicting with 0.001 and then filtering out those with a confidence score above 0.1.

[fcakyon]

This is a normal behaviour. Filtering predictions before the postprocess and after the postprocess should yield different results.

[CesareDavidePace]

quindi se voglio prendere tutte le predizioni e poi filtrare come posso fare?
Devo per forza eseguire l'inferenza a diversi livelli di confidenza?

[fcakyon]

so if i want to take all the predictions and then filter how can i do?
Do I have to run the inference at different confidence levels?

Try using

• postprocess_type='NMS'
• postprocess_match_metric='IOU'
• postprocess_match_threshold= similar to your original models nms iou threshold

This way, you should be able to use a lower model threshold in sahi.