AttributeError: 'ConfigDict' object has no attribute 'pipeline'

[wcyy0123]

I have set

data = dict(
    train=dict(
        pipeline=train_pipeline,
        ann_file=data_root + 'train/txt_label/',
        img_prefix=data_root + 'train/trainimg/'),
    val=dict(
        ann_file=data_root + 'val/txt_label/',
        img_prefix=data_root + 'val/valimg/'),
    test=dict(
        ann_file=data_root + 'test/images/',
        img_prefix=data_root + 'test/images/'))

but the problem occurs:

  File "huge_image_demo.py", line 223, in <module>
    main(args)
  File "huge_image_demo.py", line 151, in main
    result = inference_detector_by_patches(model, img_path, args.patch_sizes,
  File "D:\software\program\Anaconda3\envs\open-mmlab1\lib\site-packages\mmrotate\apis\inference.py", line 42, in inference_detector_by_patches
    cfg.data.test.pipeline[0].type = 'LoadPatchFromImage'
  File "D:\software\program\Anaconda3\envs\open-mmlab1\lib\site-packages\mmcv\utils\config.py", line 50, in __getattr__
    raise ex
AttributeError: 'ConfigDict' object has no attribute 'pipeline'

[jbwang1997]

There is no pipeline in val and test. Please set pipelines in them.