I’m new to yolox and am using the voc2007 dataset.

  File "/content/YOLOX/yolox/data/datasets/voc.py", line 264, in _write_voc_results_file
    if dets == []:
       └ array([[-4.2242885e+00,  3.8780212e+02,  3.7010245e+02,  4.9919244e+02,
                  3.5381533e-02],
                [-4.9626112e+00,  3.1...

ValueError: operands could not be broadcast together with shapes (3,5) (0,) 

Further, this bug only appears whenever I evaluate an epoch. so if I evaluate later it won’t appear.
I’ve updated the config and as far as im aware I’ve ensured the dataset paths are correct. what steps do you think i should take? I tried decreasing the confidence threshold to detect more objects, and it did change the shape of (0,5) in the error below as expected but it’s still being broadcasted with (0,) . I’ve tried to fix this for a long time and can’t think of other things to try.
yolox_base and yolox_voc_s matches in depth, width, and num_classes. The custom dataset is in the VocDevkit > VOC2007> ANnotations, imagesets, jpegimages. The self.data_dir is vocdevkit, and self.train_ann is ‘trainval’ and so on.
For the full bug, it says:

Writing aeroplane VOC results file
2024-06-11 14:43:11 | INFO     | yolox.core.trainer:195 - Training of experiment is done and the best AP is 0.00
2024-06-11 14:43:11 | ERROR    | yolox.core.launch:98 - An error has been caught in function 'launch', process 'MainProcess' (14805), thread 'MainThread' (136600321028736):
Traceback (most recent call last):

  File "/content/YOLOX/tools/train.py", line 138, in <module>
    launch(
    └ <function launch at 0x7c3bec9cab00>

> File "/content/YOLOX/yolox/core/launch.py", line 98, in launch
    main_func(*args)
    │          └ (╒═══════════════════╤═══════════════════════════════════════════════════════════════════════════════════════════════════════...
    └ <function main at 0x7c3bdb502a70>

  File "/content/YOLOX/tools/train.py", line 118, in main
    trainer.train()
    │       └ <function Trainer.train at 0x7c3b5729f370>
    └ <yolox.core.trainer.Trainer object at 0x7c3b572c3070>

  File "/content/YOLOX/yolox/core/trainer.py", line 76, in train
    self.train_in_epoch()
    │    └ <function Trainer.train_in_epoch at 0x7c3b5729f400>
    └ <yolox.core.trainer.Trainer object at 0x7c3b572c3070>

  File "/content/YOLOX/yolox/core/trainer.py", line 86, in train_in_epoch
    self.after_epoch()
    │    └ <function Trainer.after_epoch at 0x7c3b5729f760>
    └ <yolox.core.trainer.Trainer object at 0x7c3b572c3070>

  File "/content/YOLOX/yolox/core/trainer.py", line 222, in after_epoch
    self.evaluate_and_save_model()
    │    └ <function Trainer.evaluate_and_save_model at 0x7c3b5729fa30>
    └ <yolox.core.trainer.Trainer object at 0x7c3b572c3070>

  File "/content/YOLOX/yolox/core/trainer.py", line 336, in evaluate_and_save_model
    (ap50_95, ap50, summary), predictions = self.exp.eval(
                                            │    │   └ <function Exp.eval at 0x7c3b5729f2e0>
                                            │    └ ╒═══════════════════╤════════════════════════════════════════════════════════════════════════════════════════════════════════...
                                            └ <yolox.core.trainer.Trainer object at 0x7c3b572c3070>

  File "/content/YOLOX/yolox/exp/yolox_base.py", line 353, in eval
    return evaluator.evaluate(model, is_distributed, half, return_outputs=return_outputs)
           │         │        │      │               │                    └ True
           │         │        │      │               └ False
           │         │        │      └ False
           │         │        └ YOLOX(
           │         │            (backbone): YOLOPAFPN(
           │         │              (backbone): CSPDarknet(
           │         │                (stem): Focus(
           │         │                  (conv): BaseConv(
           │         │                    (conv): ...
           │         └ <function VOCEvaluator.evaluate at 0x7c3b5729d7e0>
           └ <yolox.evaluators.voc_evaluator.VOCEvaluator object at 0x7c3c09328850>

  File "/content/YOLOX/yolox/evaluators/voc_evaluator.py", line 114, in evaluate
    eval_results = self.evaluate_prediction(data_list, statistics)
                   │    │                   │          └ tensor([ 19.0988,   4.1687, 309.0000], device='cuda:0')
                   │    │                   └ {0: (tensor([[ 5.9809e+01,  2.5041e+01,  2.9600e+02,  4.4769e+02],
                   │    │                             [ 1.1087e+02,  3.4427e+01,  4.0809e+02,  4.7695e+0...
                   │    └ <function VOCEvaluator.evaluate_prediction at 0x7c3b5729d900>
                   └ <yolox.evaluators.voc_evaluator.VOCEvaluator object at 0x7c3c09328850>

  File "/content/YOLOX/yolox/evaluators/voc_evaluator.py", line 186, in evaluate_prediction
    mAP50, mAP70 = self.dataloader.dataset.evaluate_detections(all_boxes, tempdir)
                   │    │          │       │                   │          └ '/tmp/tmphog54lk2'
                   │    │          │       │                   └ [[array([[-4.2242885e+00,  3.8780212e+02,  3.7010245e+02,  4.9919244e+02,
                   │    │          │       │                              3.5381533e-02],
                   │    │          │       │                            [-4.9626112e+00,  3...
                   │    │          │       └ <function VOCDetection.evaluate_detections at 0x7c3b5729e200>
                   │    │          └ <yolox.data.datasets.voc.VOCDetection object at 0x7c3b502b6980>
                   │    └ <torch.utils.data.dataloader.DataLoader object at 0x7c3b502b7430>
                   └ <yolox.evaluators.voc_evaluator.VOCEvaluator object at 0x7c3c09328850>

  File "/content/YOLOX/yolox/data/datasets/voc.py", line 230, in evaluate_detections
    self._write_voc_results_file(all_boxes)
    │    │                       └ [[array([[-4.2242885e+00,  3.8780212e+02,  3.7010245e+02,  4.9919244e+02,
    │    │                                  3.5381533e-02],
    │    │                                [-4.9626112e+00,  3...
    │    └ <function VOCDetection._write_voc_results_file at 0x7c3b5729e320>
    └ <yolox.data.datasets.voc.VOCDetection object at 0x7c3b502b6980>

  File "/content/YOLOX/yolox/data/datasets/voc.py", line 264, in _write_voc_results_file
    if dets == []:
       └ array([[-4.2242885e+00,  3.8780212e+02,  3.7010245e+02,  4.9919244e+02,
                  3.5381533e-02],
                [-4.9626112e+00,  3.1...

ValueError: operands could not be broadcast together with shapes (3,5) (0,)