Error at inference

thanksf or opensourcing the work , when i try to run the borderdet r50 model i am getting the following error

command : pods_test --num-gpus 1 MODEL.WEIGHTS weights/BorderDet_R_50_FPN_1x.pth OUTPUT_DIR /BorderDet/op/
by going into the folder "BorderDet/playground/detection/coco/borderdet/borderdet.res50.fpn.coco.800size.1x"
Error :
09/09 11:50:28 c2.utils.env.env]: Using a generated random seed 28549847
[09/09 11:50:30 c2.checkpoint.checkpoint]: Loading checkpoint from /BorderDet/weights/BorderDet_R_50_FPN_1x.pth
WARNING [09/09 11:50:30 c2.checkpoint.checkpoint]: 'backbone.top_block.p6.weight' has shape (256, 2048, 3, 3) in the checkpoint but (256, 256, 3, 3) in the model! Skipped.
[09/09 11:50:30 c2.checkpoint.checkpoint]: Some model parameters are not in the checkpoint:
backbone.top_block.p6.weight
[09/09 11:50:30 c2.data.build]: TransformGens used: [ResizeShortestEdge(short_edge_length=(800, 800), max_size=1333, sample_style='choice')] in testing
[09/09 11:50:31 c2.data.datasets.coco]: Loaded 5000 images in COCO format from

Start by rerunning the shown pods_test command from playground/detection/coco/borderdet/borderdet.res50.fpn.coco.800size.1x and capture the complete output after the COCO dataset-loading line. Inspect the checkpoint and model-shape mismatch for backbone.top_block.p6.weight; done means the reported inference error is reproduced and its cause or a successful inference result is documented.