mdeff/cnn_graph

ValueError: feeding

magnumical opened this issue · 2 comments

Hi!
I encountered this issue while I was trying to feed my data.

My input shape:

  • X_train(967680,64)
  • train_labels (967680,1)
  • X_test(1075020,64)
  • test_labels(1075020,1)

Paraneters
params['batch_size'] = 1024


Traceback (most recent call last):
  File "onEEGcode.py", line 98, in <module>
    accuracy, loss, t_step = model.fit(X_train, train_labels, X_test, test_labels)
  File "D:\GRAPHneuralNET\cnn_graph-master\lib\models.py", line 109, in fit
    learning_rate, loss_average = sess.run([self.op_train, self.op_loss_average], feed_dict)
  File "C:\Users\REZA\AppData\Roaming\Python\Python36\site-packages\tensorflow\python\client\session.py", line 929, in run
    run_metadata_ptr)
  File "C:\Users\REZA\AppData\Roaming\Python\Python36\site-packages\tensorflow\python\client\session.py", line 1128, in _run
    str(subfeed_t.get_shape())))
ValueError: Cannot feed value of shape (1024, 1) for Tensor 'inputs/labels:0', which has shape '(1024,)'

python 3.6
tensorflow 1.13.1

mdeff commented

Maybe make train_labels and test_labels as (967680,) instead of (967680,1)?

It worked by reshape() but still can be solved by using squeeze()

THANKS