[Fixbug] Fix for softmmax cpu causing issues #437
Closed
Conversation
This file contains 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
works on 8x8 at least but bad exp save for omp changes working and faster than pytorch works and is fast but exp is WIP remove useless files minor changes for rebase delete trash fix trash fix trash initial commit works on 8x8 at least but bad exp save for omp changes working and faster than pytorch works and is fast but exp is WIP remove useless files minor changes for rebase delete trash fix trash fix trash change imports fix for diff size, compiledmodule error fix
works on 8x8 at least but bad exp save for omp changes working and faster than pytorch works and is fast but exp is WIP remove useless files minor changes for rebase delete trash fix trash fix trash
works on 8x8 at least but bad exp save for omp changes working and faster than pytorch works and is fast but exp is WIP remove useless files minor changes for rebase delete trash fix trash fix trash
works on 8x8 at least but bad exp save for omp changes working and faster than pytorch works and is fast but exp is WIP remove useless files minor changes for rebase delete trash fix trash fix trash
yaoyaoding
changed the title
vadiklyutiy
pushed a commit
that referenced
this pull request
Dec 19, 2024
…rm` during operator fusion pass (#437) Closes #393 I spent some time looking into the issue without much progress, but I first found that the error message in the linked issue disappeared after commenting out either the `resolve_variant_pass()` or `fuse_operator_pass()` [here](https://github.com/CentML/hidet/blob/bfbb4db6d7792ed3de3be4e9702e597b8fbbe373/python/hidet/graph/transforms/__init__.py#L55-L61). Then, I found that simply adding the `EmbeddingBagOp` to the `NOT_FUSIBLE` set resolves the error. It is a workaround for now, but I am unaware of better solutions.
vadiklyutiy
pushed a commit
that referenced
this pull request
Dec 20, 2024
…rm` during operator fusion pass (#437) Closes #393 I spent some time looking into the issue without much progress, but I first found that the error message in the linked issue disappeared after commenting out either the `resolve_variant_pass()` or `fuse_operator_pass()` [here](https://github.com/CentML/hidet/blob/bfbb4db6d7792ed3de3be4e9702e597b8fbbe373/python/hidet/graph/transforms/__init__.py#L55-L61). Then, I found that simply adding the `EmbeddingBagOp` to the `NOT_FUSIBLE` set resolves the error. It is a workaround for now, but I am unaware of better solutions.
|
@yaoyaoding |
vadiklyutiy
pushed a commit
that referenced
this pull request
Dec 26, 2024
…rm` during operator fusion pass (#437) Closes #393 I spent some time looking into the issue without much progress, but I first found that the error message in the linked issue disappeared after commenting out either the `resolve_variant_pass()` or `fuse_operator_pass()` [here](https://github.com/CentML/hidet/blob/bfbb4db6d7792ed3de3be4e9702e597b8fbbe373/python/hidet/graph/transforms/__init__.py#L55-L61). Then, I found that simply adding the `EmbeddingBagOp` to the `NOT_FUSIBLE` set resolves the error. It is a workaround for now, but I am unaware of better solutions.
|
Seems there is some problem in the non-fp32 softmax but I don't remember the exact problem. But it's okay to close this and add a PR to fix the problem by fixing the operator template. |
|
I don't remember the issue exactly but I think it had something to do with the kernel working in isolation but not when it was included in a larger model graph because I put one of the functions in the wrong place. The change is simple so it can just be incorporated in another PR. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
moved implement_cpu to the cpu task