fix node2vec neg affinity calculation
This commit is contained in:
parent
b48b92b5e7
commit
d928dc72f6
@ -477,7 +477,7 @@ class Node2VecModel(GeneralizedModel):
|
||||
|
||||
def _loss(self):
|
||||
aff = tf.reduce_sum(tf.multiply(self.outputs1, self.outputs2), 1) + self.outputs2_bias
|
||||
neg_aff = tf.matmul(self.outputs2, tf.transpose(self.neg_outputs)) + self.neg_outputs_bias
|
||||
neg_aff = tf.matmul(self.outputs1, tf.transpose(self.neg_outputs)) + self.neg_outputs_bias
|
||||
true_xent = tf.nn.sigmoid_cross_entropy_with_logits(
|
||||
labels=tf.ones_like(aff), logits=aff)
|
||||
negative_xent = tf.nn.sigmoid_cross_entropy_with_logits(
|
||||
|
Loading…
Reference in New Issue
Block a user