register custom_op for fpEBC #2067
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
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
force-pushed
the
export-D57829276
branch
from
1e4a525 to
350b0e7
Compare
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 4, 2024
Summary: # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 4, 2024
Summary: # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
TroyGarden
force-pushed
the
export-D57829276
branch
from
350b0e7 to
e6ff80c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 5, 2024
Summary: # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
TroyGarden
force-pushed
the
export-D57829276
branch
from
e6ff80c to
4e41b4d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
force-pushed
the
export-D57829276
branch
from
4e41b4d to
048be62
Compare
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 6, 2024
Summary: # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 7, 2024
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
TroyGarden
force-pushed
the
export-D57829276
branch
from
048be62 to
1e6e30d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
force-pushed
the
export-D57829276
branch
from
1e6e30d to
3d0e187
Compare
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 17, 2024
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 18, 2024
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
TroyGarden
force-pushed
the
export-D57829276
branch
from
3d0e187 to
7eeb139
Compare
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 22, 2024
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
TroyGarden
force-pushed
the
export-D57829276
branch
from
7eeb139 to
d1203f4
Compare
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
force-pushed
the
export-D57829276
branch
from
d1203f4 to
04702f1
Compare
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 25, 2024
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
TroyGarden
pushed a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 25, 2024
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
added a commit
to TroyGarden/torchrec
that referenced
this pull request
Jun 25, 2024
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
TroyGarden
force-pushed
the
export-D57829276
branch
from
04702f1 to
903fb9c
Compare
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Reviewed By: PaulZhang12 Differential Revision: D57829276
|
This pull request was exported from Phabricator. Differential Revision: D57829276 |
TroyGarden
force-pushed
the
export-D57829276
branch
from
903fb9c to
10c247f
Compare
TroyGarden
pushed a commit
to TroyGarden/torchrec
that referenced
this pull request
Jul 1, 2024
Summary: Pull Request resolved: pytorch#2067 # context * convert `FeatureProcessedEmbeddingBagCollection` to custom op in IR export * add serialization and deserialization function for FPEBC * add an API for the `FeatureProcessorInterface` to export necessary paramters for create an instance * use this API (`get_init_kwargs`) in the serialize and deserialize functions to flatten and unflatten the feature processor # details 1. Added `FPEBCMetadata` schema for FP_EBC, use a `fp_json` string to store the necessary paramters 2. Added `FPEBCJsonSerializer`, converted the init_kwargs to json string and store in the `fp_json` field in the metadata 3. Added a fqn check for `serialized_fqns`, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module) 4. Added an API called `get_init_kwargs` for `FeatureProcessorsCollection` and `FeatureProcessor`, and use a `FeatureProcessorNameMap` to map the classname to the feature processor class 5. Added `_non_strict_exporting_forward` function for FPEBC so that in non_strict IR export it goes to the custom_op logic Differential Revision: D57829276
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
FeatureProcessedEmbeddingBagCollectionto custom op in IR exportFeatureProcessorInterfaceto export necessary paramters for create an instanceget_init_kwargs) in the serialize and deserialize functions to flatten and unflatten the feature processordetails
FPEBCMetadataschema for FP_EBC, use afp_jsonstring to store the necessary paramtersFPEBCJsonSerializer, converted the init_kwargs to json string and store in thefp_jsonfield in the metadataserialized_fqns, so that when a higher-level module is serialized, the lower-level module can be skipped (it's already included in the higher-level module)get_init_kwargsforFeatureProcessorsCollectionandFeatureProcessor, and use aFeatureProcessorNameMapto map the classname to the feature processor class_non_strict_exporting_forwardfunction for FPEBC so that in non_strict IR export it goes to the custom_op logicDifferential Revision: D57829276