|
/Projection/DOTA_python/Faster_RCNN_for_DOTA-master/faster_rcnn/../lib/utils/symbol.py:38: UserWarning: Cannot decide shape for the following arguments (0s in shape means unknown dimensions). Consider providing them as input:
gt_boxes: ()
arg_shape, out_shape, aux_shape = self.sym.infer_shape(**data_shape_dict)
Traceback (most recent call last):
File "/Projection/DOTA_python/Faster_RCNN_for_DOTA-master/experiments/faster_rcnn/rcnn_dota_quadrangle_e2e.py", line 25, in
train_quadrangle_end2end.main()
File "/Projection/DOTA_python/Faster_RCNN_for_DOTA-master/faster_rcnn/train_quadrangle_end2end.py", line 166, in main
config.TRAIN.begin_epoch, config.TRAIN.end_epoch, config.TRAIN.lr, config.TRAIN.lr_step)
File "/Projection/DOTA_python/Faster_RCNN_for_DOTA-master/faster_rcnn/train_quadrangle_end2end.py", line 93, in train_net
sym_instance.infer_shape(data_shape_dict)
File "/Projection/DOTA_python/Faster_RCNN_for_DOTA-master/faster_rcnn/../lib/utils/symbol.py", line 41, in infer_shape
self.arg_shape_dict = dict(zip(self.sym.list_arguments(), arg_shape))
TypeError: zip argument #2 must support iteration
|