https://github.com/devbruce/calc4ap
Easy AP Calculator for Object Detection
https://github.com/devbruce/calc4ap
ap coco evaluation map metric object-detection pascalvoc
Last synced: about 1 month ago
JSON representation
Easy AP Calculator for Object Detection
- Host: GitHub
- URL: https://github.com/devbruce/calc4ap
- Owner: devbruce
- License: mit
- Created: 2021-06-03T02:30:49.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2021-06-20T06:50:10.000Z (over 4 years ago)
- Last Synced: 2025-03-06T05:37:48.978Z (11 months ago)
- Topics: ap, coco, evaluation, map, metric, object-detection, pascalvoc
- Language: Python
- Homepage:
- Size: 9.77 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# calc4ap
> Pascal VOC ()
## Install via PyPI
```bash
$ pip install calc4ap
```
## How to Use
```python
from calc4ap.voc import CalcVOCmAP
# (some codes ...)
voc_ap = CalcVOCmAP(labels=labels, preds=preds, iou_thr=0.5, conf_thr=0.0)
ap_summary = voc_ap.get_summary()
car_AP = ap_summary['car']
mAP = ap_summary['mAP']
```
### `CalcVOCmAP` Args
- `preds`: (list): list of `[left, top, right, bottom, confidence, class_name, image_id]`
- `labels`: (list): list of `[left, top, right, bottom, class_name, image_id]`
- `iou_thr`: (float): IoU Threshold (Default: 0.5)
- `conf_thr`: (float): Confidence Threshold (Default: 0.0)