Add global_counter to Trainer as optional argument #271
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| else: | ||
| self.global_step = tf.Variable(0., name='Global_Step', | ||
| trainable=False) | ||
| self.global_step = global_step |
There was a problem hiding this comment.
This line shouldn't be necessary as it would have already been set. This conflicts with my other comment but either way we can make this more simple in one way or the other.
There was a problem hiding this comment.
The last line is a stupid mistake - sorry!
| @@ -87,8 +87,11 @@ def __init__(self, train_ops, graph=None, clip_gradients=5.0, | |||
| self.validate_trainop_names() | |||
|
|
|||
| self.global_loss = None | |||
| self.global_step = tf.Variable(0., name='Global_Step', | |||
| trainable=False) | |||
| if global_step: | |||
There was a problem hiding this comment.
Here you need to set:
if global_step is None:
global_step = tf.Variable(0., name='Global_Step', trainable=False)
self.global_step = global_step
because 'if Tensor' raises an exception in TensorFlow.
|
That looks good @wielandbrendel! I just added a comment to use: as @braddengross suggested, because the other way will raise a TensorFlow error. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR allows to use tf.train.exponential_decay together with Trainer. It adds an optional argument to pass a global_counter to Trainer, and to hence initialise the global_counter before initializing Trainer. This is necessary if the parameter to be decayed is part of the loss function.