NJT binary pointwise broadcasting support via jagged <-> padded dense conversion #133021
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
… conversion [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/133021
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 6a17e36 with merge base fd182b9 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
jbschlosser
requested review from
YuqingJ,
davidberard98 and
soulitzer
and removed request for
soulitzer
52 tasks
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. [ghstack-poisoned]
jbschlosser
added
module: nestedtensor
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
cpuhrsch
approved these changes
Sep 19, 2024
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
…added dense conversion" Related: #132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. This PR also adds (hacky) support for bool inputs to the jagged <-> padded dense conversions. The underlying CUDA kernels do not support integer / bool inputs; so the following workaround is employed: `convert input -> half, run conversion kernel, convert output -> bool`. Note that this bool support is needed specifically for the backward formula of `fmax`, and likely others. cc cpuhrsch bhosmer drisspg soulitzer davidberard98 YuqingJ [ghstack-poisoned]
|
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
BoyuanFeng
pushed a commit
to BoyuanFeng/pytorch
that referenced
this pull request
Sep 25, 2024
… conversion (pytorch#133021) Related: pytorch#132695 This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes: * `(B, j0, D) + (1, 1, 1)` * `(B, j0, D) + (B, 1, 1)` * `(B, j0, D) + (B, 1, D)` * etc. This PR also adds (hacky) support for bool inputs to the jagged <-> padded dense conversions. The underlying CUDA kernels do not support integer / bool inputs; so the following workaround is employed: `convert input -> half, run conversion kernel, convert output -> bool`. Note that this bool support is needed specifically for the backward formula of `fmax`, and likely others. Pull Request resolved: pytorch#133021 Approved by: https://github.com/cpuhrsch
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 (oldest at bottom):
Related: #132695
This PR uses padded dense <-> jagged conversions to handle binary pointwise broadcasting of (NT, T) and (T, NT). This includes:
(B, j0, D) + (1, 1, 1)(B, j0, D) + (B, 1, 1)(B, j0, D) + (B, 1, D)This PR also adds (hacky) support for bool inputs to the jagged <-> padded dense conversions. The underlying CUDA kernels do not support integer / bool inputs; so the following workaround is employed:
convert input -> half, run conversion kernel, convert output -> bool. Note that this bool support is needed specifically for the backward formula offmax, and likely others.cc @cpuhrsch @bhosmer @drisspg @soulitzer @davidberard98 @YuqingJ