Implement tensor.size(int) for BatchedTensor #40028
Closed
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 this call native::size directly. Some alternatives I considered were: - Call VariableType::size directly. That seems isomorphic to what we're doing now. - when creating a BatchedTensor from a regular tensor, put all of the keys on that tensor into the BatchedTensor's dispatch key set and use the dispatcher fallthrough mechanism. That seems weird because BatchedTensor is a tensor wrapper and also error prone because if BatchedTensor gets the VariableType key, there's a chance that if something goes wrong, an autogradmeta gets created on it... Test Plan: - `./build/bin/vmap_test` [ghstack-poisoned]
We have this call native::size directly. Some alternatives I considered were: - Call VariableType::size directly. That seems isomorphic to what we're doing now. - when creating a BatchedTensor from a regular tensor, put all of the keys on that tensor into the BatchedTensor's dispatch key set and use the dispatcher fallthrough mechanism. That seems weird because BatchedTensor is a tensor wrapper and also error prone because if BatchedTensor gets the VariableType key, there's a chance that if something goes wrong, an autogradmeta gets created on it... Test Plan: - `./build/bin/vmap_test` [ghstack-poisoned]
💊 CI failures summary and remediationsAs of commit 69ec2ca (more details on the Dr. CI page): ✅ None of the CI failures appear to be your fault 💚
❄️ 3 failures tentatively classified as flakybut reruns have not yet been triggered to confirm:
|
We have this call native::size directly. Some alternatives I considered were: - Call VariableType::size directly. That seems isomorphic to what we're doing now. - when creating a BatchedTensor from a regular tensor, put all of the keys on that tensor into the BatchedTensor's dispatch key set and use the dispatcher fallthrough mechanism. That seems weird because BatchedTensor is a tensor wrapper and also error prone because if BatchedTensor gets the VariableType key, there's a chance that if something goes wrong, an autogradmeta gets created on it... Test Plan: - `./build/bin/vmap_test` [ghstack-poisoned]
We have this call native::size directly. Some alternatives I considered were: - Call VariableType::size directly. That seems isomorphic to what we're doing now. - when creating a BatchedTensor from a regular tensor, put all of the keys on that tensor into the BatchedTensor's dispatch key set and use the dispatcher fallthrough mechanism. That seems weird because BatchedTensor is a tensor wrapper and also error prone because if BatchedTensor gets the VariableType key, there's a chance that if something goes wrong, an autogradmeta gets created on it... Test Plan: - `./build/bin/vmap_test` [ghstack-poisoned]
ezyang
approved these changes
Jun 16, 2020
We have this call native::size directly. Some alternatives I considered were: - Call VariableType::size directly. That seems isomorphic to what we're doing now. - when creating a BatchedTensor from a regular tensor, put all of the keys on that tensor into the BatchedTensor's dispatch key set and use the dispatcher fallthrough mechanism. That seems weird because BatchedTensor is a tensor wrapper and also error prone because if BatchedTensor gets the VariableType key, there's a chance that if something goes wrong, an autogradmeta gets created on it... Test Plan: - `./build/bin/vmap_test` [ghstack-poisoned]
xwang233
pushed a commit
to xwang233/pytorch
that referenced
this pull request
Jun 20, 2020
Summary: Pull Request resolved: pytorch#40028 We have this call native::size directly. Some alternatives I considered were: - Call VariableType::size directly. That seems isomorphic to what we're doing now. - when creating a BatchedTensor from a regular tensor, put all of the keys on that tensor into the BatchedTensor's dispatch key set and use the dispatcher fallthrough mechanism. That seems weird because BatchedTensor is a tensor wrapper and also error prone because if BatchedTensor gets the VariableType key, there's a chance that if something goes wrong, an autogradmeta gets created on it... Test Plan: - `./build/bin/vmap_test` Differential Revision: D22070655 Pulled By: zou3519 fbshipit-source-id: 18530579ad41f3c4f96589da41eb24a46caf7af9
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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.
Stack from ghstack:
vector<int64_t>#40042 Change VmapTransforms to use SmallVector instead ofvector<int64_t>We have this call native::size directly. Some alternatives I considered
were:
doing now.
keys on that tensor into the BatchedTensor's dispatch key set and use
the dispatcher fallthrough mechanism. That seems weird because
BatchedTensor is a tensor wrapper and also error prone because if
BatchedTensor gets the VariableType key, there's a chance that if
something goes wrong, an autogradmeta gets created on it...
Test Plan:
./build/bin/vmap_testDifferential Revision: D22070655