unify the definition of EmbeddingTableType
#2391
Open
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
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D62596655 |
|
This pull request was exported from Phabricator. Differential Revision: D62596655 |
beekbin
pushed a commit
to beekbin/torchrec
that referenced
this pull request
Sep 13, 2024
Summary: Pull Request resolved: pytorch#2391 # Problem `EmbeddingTableType` is introduced recently for ZCH. There are two same definitions of `EmbeddingTableType` * one is defined in [torchrec/fb/distributed/sparsenn_configs.py](https://www.internalfb.com/code/fbsource/[3a83814d95f27eec3bf425d468a5234d6ca8a381]/fbcode/torchrec/fb/distributed/sparsenn_configs.py?lines=110) * The other is defined in [minimal_viable_ai/core/model_family_api/configs.py](https://www.internalfb.com/code/fbsource/[bbe727e13ae251889c5df57b58f2cbc05675662e]/fbcode/minimal_viable_ai/core/model_family_api/configs.py?lines=76) **Problem is that** The fields of `EmbeddingTable` defined in `minimal_viable_ai/core/model_family_api/configs.py` will be assigned to the `EmbeddingTable` defined in `torchrec/fb/distributed/sparsenn_configs.py`. The assignment happens at [minimal_viable_ai/sandbox/umia_hstu/umia_configs_utils.py ](https://fburl.com/code/5gb24p9j) for model publish. So if the EmbeddingTable are different, the assignment will fail. # solution from this diff * Option1: keep one of them, and the other import it. This would cause dependency loop; * Option2(this diff): move the definition of `EmbeddingTableType` to [torchrec/modules/embedding_configs.py](https://fburl.com/code/tvob11sz) Differential Revision: D62596655
beekbin
force-pushed
the
export-D62596655
branch
from
1523cbc
to
06cac81
Compare
Summary: Pull Request resolved: pytorch#2391 # Problem `EmbeddingTableType` is introduced recently for ZCH. There are two same definitions of `EmbeddingTableType` * one is defined in [torchrec/fb/distributed/sparsenn_configs.py](https://www.internalfb.com/code/fbsource/[3a83814d95f27eec3bf425d468a5234d6ca8a381]/fbcode/torchrec/fb/distributed/sparsenn_configs.py?lines=110) * The other is defined in [minimal_viable_ai/core/model_family_api/configs.py](https://www.internalfb.com/code/fbsource/[bbe727e13ae251889c5df57b58f2cbc05675662e]/fbcode/minimal_viable_ai/core/model_family_api/configs.py?lines=76) **Problem is that** The fields of `EmbeddingTable` defined in `minimal_viable_ai/core/model_family_api/configs.py` will be assigned to the `EmbeddingTable` defined in `torchrec/fb/distributed/sparsenn_configs.py`. The assignment happens at [minimal_viable_ai/sandbox/umia_hstu/umia_configs_utils.py ](https://fburl.com/code/5gb24p9j) for model publish. So if the EmbeddingTable are different, the assignment will fail. # solution from this diff * Option1: keep one of them, and the other import it. This would cause dependency loop; * Option2(this diff): move the definition of `EmbeddingTableType` to [torchrec/modules/embedding_configs.py](https://fburl.com/code/tvob11sz) Differential Revision: D62596655
|
This pull request was exported from Phabricator. Differential Revision: D62596655 |
beekbin
force-pushed
the
export-D62596655
branch
from
06cac81
to
19d3dc1
Compare
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:
Problem
EmbeddingTableTypeis introduced recently for ZCH.There are two same definitions of
EmbeddingTableTypeProblem is that
The fields of
EmbeddingTabledefined inminimal_viable_ai/core/model_family_api/configs.pywill be assigned to theEmbeddingTabledefined intorchrec/fb/distributed/sparsenn_configs.py.The assignment happens at minimal_viable_ai/sandbox/umia_hstu/umia_configs_utils.py for model publish.
So if the EmbeddingTable are different, the assignment will fail.
solution from this diff
EmbeddingTableTypeto torchrec/modules/embedding_configs.pyDifferential Revision: D62596655