I love Tensorflow! But tf.contrib is so volatile, I write some code that works in the morning it becomes obsolete in the afternoon.
-
TF team has some serious standardization work to do.
@tensorflow
tf.slim.conv, tf.contrib.layers.conv, tf.contrib.keras.conv are meant to be the same thing, same implementation.
-
-
tf.contrib.layers.conv and tf.contrib.keras.conv will be literally the same object in the near future.
-
That's a step in the right direction. Although I think a single high level API consistent with core layers would be less confusing to users.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.