yunjey / davian-tensorflow Goto Github PK

Hello,

When I get to the part of creating the model:
model = Convnet ()
I get the following error:

`ValueError Traceback (most recent call last)
in ()
----> 1 model = Convnet()

in init(self, learning_rate)
22 correct_pred = tf.equal(pred, self.targets)
23 self.accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32))
---> 24 self.loss = tf.reduce_mean(tf.nn.sparse_softmax_cross_entropy_with_logits(out, self.targets))
25 self.train_op = tf.train.AdamOptimizer(self.learning_rate).minimize(self.loss)

~\Anaconda3\envs\tensorflow-gpu\lib\site-packages\tensorflow\python\ops\nn_ops.py in sparse_softmax_cross_entropy_with_logits(_sentinel, labels, logits, name)
2010 """
2011 _ensure_xent_args("sparse_softmax_cross_entropy_with_logits", _sentinel,
-> 2012 labels, logits)
2013
2014 # TODO(pcmurray) Raise an error when the label is not an index in

~\Anaconda3\envs\tensorflow-gpu\lib\site-packages\tensorflow\python\ops\nn_ops.py in _ensure_xent_args(name, sentinel, labels, logits)
1773 if sentinel is not None:
1774 raise ValueError("Only call %s with "
-> 1775 "named arguments (labels=..., logits=..., ...)" % name)
1776 if labels is None or logits is None:
1777 raise ValueError("Both labels and logits must be provided.")

ValueError: Only call sparse_softmax_cross_entropy_with_logits with named arguments (labels=..., logits=..., ...)`