Calculating gradients for partial graph #7269
Merged
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
JiayiFeng
reviewed
Jan 8, 2018
| return op_path | ||
|
|
||
|
|
||
| def calc_gradient(targets, inputs, target_gradients=None, no_grad_set=None): |
There was a problem hiding this comment.
I have a few questions about the calc_gradient:
- It seems that this API appends partial backward to calculate gradients of
targetstoinputs. Can this API be used by users directly? And when will users use it? - Here are some special handlings(like
_rename_grad_) to make sure this API can be invoked for multiple times. However, in each invoking theprogramitself will be changed(some backward ops are appended). So during the second invoke, the backward part generated in the first time is still in the program and is regarded as forwarding part. Is this reasonable?
There was a problem hiding this comment.
- Yes. Some algorithm have its own way of calculating gradients. One example is Learning to learn by gradient descent by gradient descent.
- Yes. This happens in cases where gradients from different parts needs to be combined in some way to calculate the final result.
added 3 commits
January 9, 2018 18:21
Added backward.calc_gradient to backpropagate gradient from given targets to inputs.
emailweixu
force-pushed
the
calc_gradient
branch
from
5cdbefe to
6e5eae1
Compare
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Outdated suggestions cannot be applied.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.
Added backward.calc_gradient to backpropagate gradient from given targets to inputs.