refactor PostProc tracing and debug message #3379
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
Summary:
# context
* There's quite some limitations on the postproc support for TorchRec's train pipeline
* add better warning message for debugging
## symptoms
* unable to run input_dist in "-1" batch with the `SparseDistTrainPipeline`, AKA, SDD (Sparse Data Dist) pipeline
* warning in log: `Module '{node.target}' will NOT be pipelined, due to input modifications`
## typical issues
* root cause: input KJT is modified or passed through some module/function potentially modifies the KJT
* pipeline_postproc is not enabled
* check the error message for `fx node {child_node.name, child_node.op, child_node.target} can't be handled correctly for postproc module`
* postproc module has trainable weights (sorry we don't support this)
* a postproc function modifies the input KJT
* two postproc modules have certain execution order
## workaround
* make the postproc function a nn.Module
* put order-dependent functions/modules under the same nn.Module to preserve the order.
Differential Revision: D82591429
meta-cla
bot
added
the
CLA Signed
|
@TroyGarden has exported this pull request. If you are a Meta employee, you can view the originating diff in D82591429. |
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Outdated suggestions cannot be applied.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.
Summary:
context
symptoms
SparseDistTrainPipeline, AKA, SDD (Sparse Data Dist) pipelineModule '{node.target}' will NOT be pipelined, due to input modificationstypical issues
fx node {child_node.name, child_node.op, child_node.target} can't be handled correctly for postproc moduleworkaround
Differential Revision: D82591429