search

hizhangp / yolo_tensorflow

Tensorflow implementation of YOLO, including training and test phase.
MIT License
795 stars 442 forks source link

InvalidArgumentError input and filter must have the same depth: 14 vs 1024 #57

Open raghavgurbaxani opened 6 years ago

raghavgurbaxani commented 6 years ago

Hi,

I encounter the following error. Would appreciate your help.

InvalidArgumentError (see above for traceback): input and filter must have the same depth: 14 vs 1024 [[Node: yolo/conv_28/BiasAdd = _MklConv2DWithBias[T=DT_FLOAT, _kernel="MklOp", data_format="NHWC", padding="VALID", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/device:CPU:0"](yolo/pad_27, yolo/conv_28/weights/read, yolo/conv_28/biases/read, DMT/_159, yolo/conv_28/weights/read:1, DMT/_160)]]

Caused by op 'yolo/conv_28/BiasAdd', defined at: File "train.py", line 165, in main() File "train.py", line 152, in main yolo = YOLONet() File "/mnt/c/scratch/training/tra658/project/yolo_tensorflow/yolo/yolo_net.py", line 41, in init is_training=is_training) File "/mnt/c/scratch/training/tra658/project/yolo_tensorflow/yolo/yolo_net.py", line 98, in build_network net, 1024, 3, 2, padding='VALID', scope='conv_28')

ganyi1996ppo commented 5 years ago

i meat this problem too,is anyone knows how to deal that,or what exactly happened,any help would be appreciated