Tensor flow compile error
prabirsinha opened this issue · comments
Hi ,
I am having the following error in running the CIFAR10 dataset training . Can you give me some where to fix this issue.
"
File "C:\Users\Prabir Sinha\AppData\Local\Programs\Python\Python35\lib\site-packages\tensorflow\python\ops\array_ops.py", line 1001, in concat
).assert_is_compatible_with(tensor_shape.scalar())
File "C:\Users\Prabir Sinha\AppData\Local\Programs\Python\Python35\lib\site-packages\tensorflow\python\framework\tensor_shape.py", line 756, in assert_is_compatible_with
raise ValueError("Shapes %s and %s are incompatible" % (self, other))
ValueError: Shapes (2, 1) and () are incompatible "
Thanks
Prabir
I think this is the first time I've heard about this error. Have you modified the code or perhaps not downloaded everything? Or are you not running the Notebook from the beginning?
You also don't write how this error is connected with the code for the CIFAR10 dataset. These two errors are inside TensorFlow. Please try and debug / make print-statements to find out where in the CIFAR10 code the problem is.
You copied the contents of the Notebook into a .py file? Did the Notebook run OK? Perhaps you made a mistake when copying the code?
The two error messages are inside TensorFlow, not in my code.
"Shapes (2, 1) and () are incompatible" tells you that two tensors have the wrong shapes. The error appears to be a call to concat
but I did a search and I don't seem to use concat
anywhere in my code.
Please try and debug this yourself, there is really nothing I can do to help. When you find the solution then please write it here in case others have the same problem in the future.
I had the same issue running the 06_CIFAR-10 tutorial, and I fix it updating my tensorflow version to 1.3.0 and now it works
python -m pip uninstall tensorflow
pip install --upgrade https://storage.googleapis.com/tensorflow/windows/cpu/tensorflow-1.3.0-cp35-cp35m-win_amd64.whl
Also, I had another issue with prettytensor '0.7.4'
TypeError: zeros_initializer() missing 1 required positional argument: 'shape'
and I fix it changing :
tf.zeros_initializer() to tf.constant_initializer(0.0)
In the files: pretty_tensor_image_methods.py, pretty_tensor_loss_methods.py, pretty_tensor_methods.py and recurrent_networks.py located in the folder C:\Python35\Lib\site-packages\prettytensor
Hope it works for you