python train.py raise MXNetError(py_str(_LIB.MXGetLastError())) mxnet.base.MXNetError: [10:30:40] src/io/iter_image_det_recordio.cc:326: ImageDetRecordIOParser: label_pad_width: 300 smaller than estimated width: 338

[QJXX]

Traceback (most recent call last):
File "train.py", line 156, in
voc07_metric=args.use_voc07_metric)
File "/home/zsl/qjx/mxnet-dssd/train/train_net.py", line 188, in train_net
label_pad_width=label_pad_width, path_imglist=train_list, **cfg.train)
File "/home/zsl/qjx/mxnet-dssd/dataset/iterator.py", line 77, in init
**kwargs)
File "/home/zsl/Programs/anaconda2/lib/python2.7/site-packages/mxnet/io.py", line 886, in creator
ctypes.byref(iter_handle)))
File "/home/zsl/Programs/anaconda2/lib/python2.7/site-packages/mxnet/base.py", line 129, in check_call
raise MXNetError(py_str(_LIB.MXGetLastError()))
mxnet.base.MXNetError: [10:30:40] src/io/iter_image_det_recordio.cc:326: ImageDetRecordIOParser: label_pad_width: 300 smaller than estimated width: 338