RuntimeError: Expected 3D (unbatched) or 4D (batched) input to conv2d, but got input of size: [1, 6, 3, 800, 800]

[talk2rajasimman]

Hello,
I am trying to use SAHI fine tuning for custom dataset, The annotation are already in the coco format, So directly I generated sliced dataset SLICE_SIZE_LIST = [480, 640] with a OVERLAP_RATIO_LIST = [0, 0.25], It's fine upto here,

While in the training I am using visdrone_fcos/fcos_crop_480_960_cls_60.py config file.
!python mmdet_tools/train.py mmdet_configs/visdrone_fcos/fcos_crop_480_960_cls_60.py

In the fcos_crop_480_960_cls_60.py I have chnaged the DATA_ROOT, NUM_CLASSES, CLASSES, (train, val, test)_dataset, path's.

While calling train.py I am getting the
Expected 3D (unbatched) or 4D (batched) input to conv2d, but got input of size: [1, 6, 3, 800, 800],

Can anyone please help me on this,