Error in transfer learning part

[Monster-Gaming-Studios]

---------------------------------------------------------------------------
InvalidArgumentError                      Traceback (most recent call last)
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py in _create_c_op(graph, node_def, inputs, control_inputs)
   1658   try:
-> 1659     c_op = c_api.TF_FinishOperation(op_desc)
   1660   except errors.InvalidArgumentError as e:

InvalidArgumentError: Dimension 0 in both shapes must be equal, but are 3 and 328. Shapes are [3,3,512,324] and [328,512,3,3]. for 'Assign_49' (op: 'Assign') with input shapes: [3,3,512,324], [328,512,3,3].

During handling of the above exception, another exception occurred:

ValueError                                Traceback (most recent call last)
11 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/ops.py in _create_c_op(graph, node_def, inputs, control_inputs)
   1660   except errors.InvalidArgumentError as e:
   1661     # Convert to ValueError for backwards compatibility.
-> 1662     raise ValueError(str(e))
   1663 
   1664   return c_op

ValueError: Dimension 0 in both shapes must be equal, but are 3 and 328. Shapes are [3,3,512,324] and [328,512,3,3]. for 'Assign_49' (op: 'Assign') with input shapes: [3,3,512,324], [328,512,3,3].

I am getting the above error when loading weights in the section 5.2 Build the model for weight_sampling_tutorial.ipynb

I am using tensorflow 1.13.2 and keras 2.1.4

Please help me @pierluigiferrari .

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[Monster-Gaming-Studios]

@pierluigiferrari

Somebody please help me.

[Sri-user]

@Monster-Gaming-Studios

I ran this today. I came across the same problem. U have to downgrade tensor flow and keras

Install tensorflow 1.4 and keras 2.0.4

After creating model and saving it, when u try to load it, u will face another problem, closure must be none or tuple error

Remove the line clear session in the load_model block. This part alone i did it in google colab

You can do all the above steps in google colab itself.

But you have to do this two steps %tensorflow_version 1.x !pip install q keras==2.1.5

[Monster-Gaming-Studios]

@Sri-user Right now I am out of station, but I will try once I get home. But thanks for your help.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.