TENSORFLOW. Les étiquettes et les logits ne correspondent pas, mais je suis sûr qu'ils ont la même forme

Question

Traceback (most recent call last): 
    File "CNN2.py", line 272, in <module> 
    _, l, predictions = session.run([optimizer, cost, y_pred], feed_dict=feed_dict) 
    File "/Users/Aldo/tensorflow/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 778, in run 
    run_metadata_ptr) 
    File "/Users/Aldo/tensorflow/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 961, in _run 
    % (np_val.shape, subfeed_t.name, str(subfeed_t.get_shape()))) 
ValueError: Cannot feed value of shape (128,) for Tensor 'y_true:0', which has shape '(128, 1)' 

Answer 1

Vous pouvez simplement remodeler la valeur des étiquettes pour (128, 1)

labels = np.reshape(labels, (128, 1)) 
feed_dict = {..., y_true:labels}