Atten4Vis / ConditionalDETR

This repository is an official implementation of the ICCV 2021 paper "Conditional DETR for Fast Training Convergence". (https://arxiv.org/abs/2108.06152)
Apache License 2.0
358 stars 48 forks source link

How extract precision, recall and f1-score metrics #13

Closed jackonealll closed 2 years ago

jackonealll commented 2 years ago

❓ How to do something using DETR

Describe what you want to do, including:

  1. what inputs you will provide, if any: I would like to know how to extract precision, recall and f1-score metrics.
  2. what outputs you are expecting: I only have AP and AR metrics provided by coco evaluater.

I am trying to use the following code but it gives me a numpy matrix:

precision = coco_eval.eval['precision']
recall = coco_eval.eval['recall']

Can you help me?

DeppMeng commented 2 years ago

Hi, we use standard COCOAPI for evaluation. If you want to extract certain metrics, you may need to refer to the COCOAPI documents. It is not directly related to our method.