Open xiaowanzizz opened 3 years ago
@xiaowanzizz 你检查一下EvalReader中设置的anno_path的json文件是否正确,报错信息是json文件解析错误。
@xiaowanzizz 你检查一下EvalReader中设置的anno_path的json文件是否正确,报错信息是json文件解析错误。
没错呀,是dataset/coco/annotations/instances_test2017.json
我上传一下我的json 文件,您可以检查一下。 jsonfile.zip
File "tools/eval.py", line 193, in
main()
File "tools/eval.py", line 175, in main
save_only=save_only)
File "/home/zzf/Desktop/PaddleDetection/ppdet/utils/eval_utils.py", line 241, in eval_results
save_only=save_only)
File "/home/zzf/Desktop/PaddleDetection/ppdet/utils/coco_eval.py", line 102, in bbox_eval
map_stats = cocoapi_eval(outfile, 'bbox', coco_gt=coco_gt)
File "/home/zzf/Desktop/PaddleDetection/ppdet/utils/coco_eval.py", line 238, in cocoapi_eval
coco_dt = coco_gt.loadRes(jsonfile)
File "/home/zzf/miniconda3/envs/torch1.7.1/lib/python3.7/site-packages/pycocotools/coco.py", line 388, in loadRes
'Results do not correspond to current coco set'
AssertionError: Results do not correspond to current coco set
infer.py可以得到正确的图片和标注结果 但是eval.py对自己训练的model 不能得出结果。