Enable open registration of VariableType objects #8540
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
We have 2 use cases where we want to experiment with new base ATen tensor types: * BatchTensor for matchbox * Tensors that live on accelerators It is possible to subclass TensorImpl to implement these but VariableType does not work with them because it cannot find the equivalent variable type in the registry. This commit changes the way we implement type -> variable(type) lookup so that torch::register_variable_type_for can be called on any at::Type. Lookups are still done using arrays so there should be no perf impact from the change.
zdevito
requested review from
apaszke,
colesbury,
ezyang,
gchanan and
soumith
as code owners
ezyang
reviewed
Jun 15, 2018
| return "VariableType"; | ||
| } | ||
|
|
||
| // Pre-condition: backend/scalar_type is a valid type in the type_registry |
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
ezyang
approved these changes
Jun 15, 2018
Contributor
ezyang
left a comment
ezyang
left a comment
There was a problem hiding this comment.
There might be some additional legwork to do when binding to Python in PyTypeObject, but the rest looks good.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
We have 2 use cases where we want to experiment with new base ATen
tensor types:
It is possible to subclass TensorImpl to implement these but VariableType
does not work with them because it cannot find the equivalent variable type
in the registry.
This commit changes the way we implement type -> variable(type) lookup so that
torch::register_variable_type_for can be called on any at::Type.
Lookups are still done using arrays so there should be no perf impact from the change.