[change] Separate action outputs into OutputDistributions object #3514
Conversation
This file contains hidden or 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
added 30 commits
January 17, 2020 10:45
added 12 commits
February 20, 2020 15:51
andrewcoh
approved these changes
Feb 27, 2020
| pass | ||
|
|
||
| @abc.abstractproperty | ||
| def total_log_probs(self) -> tf.Tensor: |
Contributor
There was a problem hiding this comment.
Is this equivalent to reduce_sum(log_probs, axis=1) or something different?
Contributor
Author
There was a problem hiding this comment.
Pretty much - are we 100% sure that reduce_sum would work for all distribution types?
| ) | ||
| # Make entropy the right shape | ||
| self.entropy = tf.ones_like(tf.reshape(mu[:, 0], [-1])) * single_dim_entropy | ||
| self.entropy = distribution.entropy |
Contributor
There was a problem hiding this comment.
can we aggregate all occurrences of this line i.e. line 283
Contributor
Author
There was a problem hiding this comment.
👍 will do it in a follow-up PR. TODO: broader discussion about standardization of discrete and continuous (requires inference changes on C#)
added 6 commits
February 27, 2020 17:38
10 tasks
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
Proposed change(s)
Rather than have the output distribution built into the Policy, we separate these out into a separate class. Child classes of
OutputDistributioninclude both continuous and discrete versions. This opens the door to adding more types of output distributions in the future without major changes to Policy code.Types of change(s)
Checklist
Other comments