Closed liuyanzhi1214 closed 3 years ago
Which result? Could you give me an example?
xiaoyu_20140901 | |
---|---|
xiaoyu_20140901@163.com | 签名由网易邮箱大师定制 On 12/21/2020 23:01,liuyanzhi1214notifications@github.com wrote:
The result of voc_eval_visualize is different with the result of mmdetection,it is worse than mmdetection.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.
this is the result of voc_eval_visualize.py: python tools/voc_eval_visualize.py ./test_result/epoch10.pkl ./work_dirs/faster_rcnn_s50_fpn_syncbn-backbone+head_mstrain-range_1x_coco/faster_rcnn_s50_fpn_syncbn-backbone+head_mstrain-range_1x_coco.py 4805 4805 /media/e706/disk_1/liuyanzhi/mmdetection-master/mmdet/core/evaluation/mean_ap_visualize.py:390: RuntimeWarning: invalid value encountered in true_divide f_measure = np.mean(2*(top/down)) /media/e706/disk_1/liuyanzhi/mmdetection-master/mmdet/core/evaluation/mean_ap_visualize.py:403: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance. In a future version, a new instance will always be created and returned. Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance. plt.subplot(211) /media/e706/disk_1/liuyanzhi/mmdetection-master/mmdet/core/evaluation/mean_ap_visualize.py:412: MatplotlibDeprecationWarning: Adding an axes using the same arguments as a previous axes currently reuses the earlier instance. In a future version, a new instance will always be created and returned. Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance. plt.subplot(212) posx and posy should be finite values posx and posy should be finite values posx and posy should be finite values posx and posy should be finite values 2020-12-21 22:44:26,548 - print - INFO - +-----------+------+-------+--------+-------+ | class | gts | dets | recall | ap | +-----------+------+-------+--------+-------+ | nail_good | 7365 | 10176 | 0.803 | 0.674 | | nail_bad | 2349 | 5753 | 0.609 | 0.309 | +-----------+------+-------+--------+-------+ | mAP | | | | 0.492 | +-----------+------+-------+--------+-------+ mAP: 0.4919012188911438 this is the result of mmdetection: python tools/test.py work_dirs/faster_rcnn_s50_fpn_syncbn-backbone+head_mstrain-range_1x_coco/faster_rcnn_s50_fpn_syncbn-backbone+head_mstrain-range_1x_coco.py work_dirs/faster_rcnn_s50_fpn_syncbn-backbone+head_mstrain-range_1x_coco/epoch_10.pth --eval mAP [>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 4805/4805, 10.7 task/s, elapsed: 451s, ETA: 0s +-----------+------+------+--------+-------+ | class | gts | dets | recall | ap | +-----------+------+------+--------+-------+ | nail_good | 7365 | 9990 | 0.967 | 0.873 | | nail_bad | 2349 | 5180 | 0.932 | 0.839 | +-----------+------+------+--------+-------+ | mAP | | | | 0.856 | +-----------+------+------+--------+-------+ OrderedDict([('mAP', 0.8556843996047974)])
The result of voc_eval_visualize is different with the result of mmdetection,it is worse than mmdetection.