[Operator][Graph][Regression] CNN optimizations#356
Merged
hjjq merged 33 commits intohidet-org:mainfrom Sep 8, 2023
Merged
Conversation
added 30 commits
September 3, 2023 11:28
yaoyaoding
approved these changes
Sep 8, 2023
Member
yaoyaoding
left a comment
There was a problem hiding this comment.
Thanks @hjjq, the PR looks good to me. Good job!
yaoyaoding
pushed a commit
to yaoyaoding/hidet
that referenced
this pull request
Sep 25, 2023
This PR aims to improve the performance of CNNs. The main changes include: 1. Add a graph pass that converts and maintains image tensors to channel last format throughout the graph as long as possible. 2. Move the padding operation into the schedule of conv2d. Also get rid of register writeback to global memory and force shared memory writeback. 3. Add various torchvision models to regression.
vadiklyutiy
added a commit
that referenced
this pull request
Jul 22, 2024
When we used `__builtin_unreachable()` for hint the info about bounds lost after some code. There was an introduced workaround that added additional hints after loops. After switching to `__builtin_assume()` the issue disappeared. This PR removes the workaround. No performance changes. http://10.24.10.108:8868/Build_History 66fd65c after 3f955de before
vadiklyutiy
added a commit
that referenced
this pull request
Jul 23, 2024
When we used `__builtin_unreachable()` for hint the info about bounds lost after some code. There was an introduced workaround that added additional hints after loops. After switching to `__builtin_assume()` the issue disappeared. This PR removes the workaround. No performance changes. http://10.24.10.108:8868/Build_History 66fd65c after 3f955de before
vadiklyutiy
added a commit
that referenced
this pull request
Dec 26, 2024
When we used `__builtin_unreachable()` for hint the info about bounds lost after some code. There was an introduced workaround that added additional hints after loops. After switching to `__builtin_assume()` the issue disappeared. This PR removes the workaround. No performance changes. http://10.24.10.108:8868/Build_History 66fd65c after 3f955de before
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
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.
This PR aims to improve the performance of CNNs. The main changes include: