SunLC

有解决吗，我也遇到这个问题了

有解决这个问题吗？

num_classes = 24 num_training_classes = 24 train_dataset = dict( _delete_=True, type='MultiModalDataset', dataset=dict( type='YOLOv5CocoDataset', data_root='/huhong', ann_file='/huhong/lincheng/mmdetection-main/data/48K_24_train.json', data_prefix=dict(img='/defaultShare'), metainfo=data_meta_info, filter_cfg=dict(filter_empty_gt=False, min_size=32)), class_text_path='/huhong/lincheng/YOLO-UniOW-main/data/texts/aeb_texts.json', pipeline=train_pipeline) persistent_workers = False train_dataloader = dict(persistent_workers=persistent_workers, batch_size=train_batch_size_per_gpu, collate_fn=dict(type='yolow_collate'), dataset=train_dataset)...

好的，感谢作者，已经解决了！