[Clustering] Small tidy up - removed unusued variables #557
Conversation
…trainable_weights has all we need. Change-Id: I271319fd9ff09f71df3178bf2459bde55e09b72c
github-actions
bot
added
the
technique:clustering
akarmi
added
the
ready to pull
|
|
||
| @property | ||
| def trainable_weights(self): | ||
| return self.layer.trainable_weights + self._trainable_weights |
There was a problem hiding this comment.
Maybe I've misunderstood but self._trainable_weights and self._non_trainable_weights should be updated automatically during calls to self.add_weight().
By removing "self._trainable_weights" here, someone calling layer.trainable_weights will no longer see all the weights that your wrapper is adding, which breaks the interface.
You can try creating a unit test that checks what layer.trainable_weights outputs and see if it makes sense and look at the base tf.keras.layers.Layer implementation to see what it does.
|
Marking as not ready to pull given my above comment. Please do add back if it changes. |
alanchiao
removed
the
ready to pull
|
This was accidentally merged - please followup on the above. |
|
Hi @alanchiao I will investigate. During debugging I have not seen them updating, but I will re-test this. |
|
Hi @alanchiao and @wwwind , |
5 participants
Small tidy up: removed variables that are not set and not used - some leftovers.