Open SreenijaK opened 5 years ago
Are you using your own custom dataset? Looks like the image you are using has only one channel which means black & white. If that is true, please make it to have 3 channels.
Hi thank you . i changed the channels to one previously changing it back to 3 worked. I'm training with your data. Now i get the following error InvalidArgumentError: Input to reshape is a tensor with 1227400 values, but the requested shape requires a multiple of 10830 [[Node: Reshape = Reshape[T=DT_FLOAT, Tshape=DT_INT32, _device="/job:localhost/replica:0/task:0/device:CPU:0"](BiasAdd_22, Reshape/shape)]]
I am wondering where "reshape" is happening?
F:/bp/computer_vision/yolo_train/data/yolo_custom.cfg parsing F:/bp/computer_vision/yolo_train/data/annotations/ Parsing for ['ball'] [====================>]100% scene21261.xml Statistics: ball: 177 Dataset size: 191 Dataset of 191 instance(s) Training statistics: Learning rate : 1e-05 Batch size : 8 Epoch number : 100 Backup every : 2000 Traceback (most recent call last):
File "
File "C:\Users\hi\Anaconda3\lib\site-packages\spyder\utils\site\sitecustomize.py", line 705, in runfile execfile(filename, namespace)
File "C:\Users\hi\Anaconda3\lib\site-packages\spyder\utils\site\sitecustomize.py", line 102, in execfile exec(compile(f.read(), filename, 'exec'), namespace)
File "F:/bp/computer_vision/yolo_train/train.py", line 26, in
File "f:\bp\computer_vision\yolo_train\darkflow\darkflow\net\flow.py", line 56, in train fetched = self.sess.run(fetches, feed_dict)
File "C:\Users\hi\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 877, in run run_metadata_ptr)
File "C:\Users\hi\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1100, in _run feed_dict_tensor, options, run_metadata)
File "C:\Users\hi\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1272, in _do_run run_metadata)
File "C:\Users\hi\Anaconda3\lib\site-packages\tensorflow\python\client\session.py", line 1291, in _do_call raise type(e)(node_def, op, message)
InvalidArgumentError: Input to reshape is a tensor with 1227400 values, but the requested shape requires a multiple of 10830 [[Node: Reshape = Reshape[T=DT_FLOAT, Tshape=DT_INT32, _device="/job:localhost/replica:0/task:0/device:CPU:0"](BiasAdd_22, Reshape/shape)]]
Caused by op 'Reshape', defined at:
File "C:\Users\hi\Anaconda3\lib\site-packages\spyder\utils\ipython\start_kernel.py", line 269, in
InvalidArgumentError (see above for traceback): Input to reshape is a tensor with 1227400 values, but the requested shape requires a multiple of 10830 [[Node: Reshape = Reshape[T=DT_FLOAT, Tshape=DT_INT32, _device="/job:localhost/replica:0/task:0/device:CPU:0"](BiasAdd_22, Reshape/shape)]]
Looks like you have an invalid shape of image to convert to 10830 which is? I want to know, each values of H, W, B, and C. Also shape or net_out should be delivered as well.
where do i check for those values
why dont you write codes?
These are the values h-19, w-19, b-5, c-1 net_out_reshape is Tensor("Reshape:0", shape=(?, 19, 19, 5, 6), dtype=float32)
yeah net_out_reshape must be (?, 19, 19, 5, 6), but would it be convertible for net_out to net_out_reshape? In order to know, I need to know the shape of net_out as well.
when i run tfnet.train() i get the following error ValueError: Cannot feed value of shape (8, 608, 608, 3) for Tensor 'input:0', which has shape '(?, 608, 608, 1)'