[nnx] fix transforms guide #4421
Merged
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
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
superbobry
reviewed
Dec 6, 2024
| @@ -345,7 +345,7 @@ To solve this issue pass all Module as arguments to the functions being transfor | |||
|
|
|||
| ### Consistent aliasing | |||
|
|
|||
| The main issue with allowing for reference semantics in transforms is that references can be shared across inputs and outputs. This can be problematic if it is not taken care of because it would lead to ill-defined or inconsistent behavior. In the example below you have a single `Weights` `nnx.Module` - `m` ` whose reference appears in multiple places in `arg1` and `arg2`. The problem here is that you also specify that you want to vectorize `arg1` in axis `0` and `arg2` in axis `1`. This would be fine in JAX because of referential transparency of pytrees. But this would be problematic in Flax NNX because you are trying to vectorize `m` in two different ways. Flax NNX will enforce consistency by raising an error. | |||
| The main issue with allowing for reference semantics in transforms is that references can be shared across inputs and outputs. This can be problematic if it is not taken care of because it would lead to ill-defined or inconsistent behavior. In the example below you have a single `Weights` Module `m` whose reference appears in multiple places in `arg1` and `arg2`. The problem here is that you also specify that you want to vectorize `arg1` in axis `0` and `arg2` in axis `1`. This would be fine in JAX because of referential transparency of pytrees. But this would be problematic in Flax NNX because you are trying to vectorize `m` in two different ways. Flax NNX will enforce consistency by raising an error. | |||
There was a problem hiding this comment.
Should this be "module" instead of "Module"?
There was a problem hiding this comment.
we use Module in other places so its fine
superbobry
approved these changes
Dec 6, 2024
cgarciae
force-pushed
the
nnx-fix-transforms-guide
branch
from
9a255ab to
10f3460
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.
What does this PR do?