experiencor / keras-yolo3

Training and Detecting Objects with YOLO3
MIT License
1.6k stars 861 forks source link

How do you determine recall and precision? #318

Closed ghost closed 3 years ago

ghost commented 3 years ago

In utils.utils, in the evaluate() fuction, there are variables recall and precision. However, printing them returns a long np array Precision [1. 1. 1. ... 0.05946994 0.05943152 0.05939316] Recall [0.00892857 0.01785714 0.02678571 ... 0.82142857 0.82142857 0.82142857] How do you extract a singe value for each? I tried averaging them, but the precision value I get seems too low: Precision 0.19783977406044592 Recall 0.7449852439361802 mAP: 0.5763