Error when finetuning self-defined dataset: Results do not correspond to current coco set

[fireoil]

Evaluating speed.

Evaluating mAP by pycocotools.
Saving ./runs/train/exp/predictions.json...
loading annotations into memory...
Done (t=0.00s)
creating index...
index created!
Loading and preparing results...
ERROR in evaluate and save model.
ERROR in training loop or eval/save model.

Training completed in 0.002 hours.
Traceback (most recent call last):
File "tools/train.py", line 87, in
main(args)
File "tools/train.py", line 77, in main
trainer.train()
File "/zhangxin/training/YOLOv6/yolov6/core/engine.py", line 62, in train
self.train_in_loop()
File "/zhangxin/training/YOLOv6/yolov6/core/engine.py", line 81, in train_in_loop
self.eval_and_save()
File "/zhangxin/training/YOLOv6/yolov6/core/engine.py", line 107, in eval_and_save
self.eval_model()
File "/zhangxin/training/YOLOv6/yolov6/core/engine.py", line 132, in eval_model
task='train')
File "/root/anaconda3/lib/python3.7/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "/zhangxin/training/YOLOv6/tools/eval.py", line 76, in run
eval_result = val.eval_model(pred_result, model, dataloader, task)
File "/zhangxin/training/YOLOv6/yolov6/core/evaler.py", line 129, in eval_model
pred = anno.loadRes(pred_json)
File "/root/anaconda3/lib/python3.7/site-packages/pycocotools/coco.py", line 328, in loadRes
'Results do not correspond to current coco set'
AssertionError: Results do not correspond to current coco set

[mtjhl]

Hi， did you set is_coco to False here

[jcrdeep]

In my case, it by giving the default file name for the JSON file, i.e: instances_val2017.json and also by giving is_coco: False

data/coco.yaml

val: coco3/images/val2017
anno_path: coco3/annotations/instances_val2017.json
is_coco: False

[mtjhl]

@fireoil Does your problem solved? If you still have same proble, please tell us.