Add example of reading learning rate from optimizer state #363
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Outdated suggestions cannot be applied.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.
Closes #312.
I extended the Optax 101 notebook to also include an example of extracting the learning rate, based on #206.
This example uses the
inject_hyperparams
wrapper.Some additional changes include:
step()
function, so it can reused later onI have also tried to do it by passing in the
count
frominner_state
to thescheduler
. I found this method quite hacky due to having to index intoinner_state[0]
, so I have not included the following way of doing it:Please let me know what you think. Happy to make any changes!