Hi, I do not how what is the "post_ops" in the following codes for? could you help explain for it?
batches = layers.interleave([x] + y, batch)skip_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)logits = [classifier(batches[0], training=True)]post_ops = [v for v in tf.get_collection(tf.GraphKeys.UPDATE_OPS) if v not in skip_ops]
In this code snippet, post_ops collects the operations to update batch norm estimates of mean and variance. Tensorflow automatically puts them inside the tf.GraphKeys.UPDATE_OPS collection.
Hi, I do not how what is the "post_ops" in the following codes for? could you help explain for it?
batches = layers.interleave([x] + y, batch)
skip_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
logits = [classifier(batches[0], training=True)]
post_ops = [v for v in tf.get_collection(tf.GraphKeys.UPDATE_OPS) if v not in skip_ops]