update tracin influence API #1072
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
|
This pull request was exported from Phabricator. Differential Revision: D41324297 |
99warriors
pushed a commit
to 99warriors/captum
that referenced
this pull request
Nov 21, 2022
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Reviewed By: cyrjano Differential Revision: D41324297 fbshipit-source-id: 7ac12211941172d86a0aadbb6b5dd41ae9e1b52b
99warriors
force-pushed
the
export-D41324297
branch
from
9b14f36
to
49f2399
Compare
|
This pull request was exported from Phabricator. Differential Revision: D41324297 |
99warriors
pushed a commit
to 99warriors/captum
that referenced
this pull request
Dec 3, 2022
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Reviewed By: cyrjano Differential Revision: D41324297 fbshipit-source-id: c5834f74e301b4ccbbc2cc0b9f331455ff04a4b2
99warriors
force-pushed
the
export-D41324297
branch
from
49f2399
to
3b1aa94
Compare
|
This pull request was exported from Phabricator. Differential Revision: D41324297 |
99warriors
pushed a commit
to 99warriors/captum
that referenced
this pull request
Dec 8, 2022
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Reviewed By: cyrjano Differential Revision: D41324297 fbshipit-source-id: 9fe108de1a6789d461c19d71b724cd18bbcffbd9
99warriors
force-pushed
the
export-D41324297
branch
from
3b1aa94
to
0d9ebb4
Compare
|
This pull request was exported from Phabricator. Differential Revision: D41324297 |
99warriors
added a commit
to 99warriors/captum
that referenced
this pull request
Dec 8, 2022
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Differential Revision: https://internalfb.com/D41324297 fbshipit-source-id: 78a982fbc07a5555c9eae1ef0b4177088cd217fd
99warriors
pushed a commit
to 99warriors/captum
that referenced
this pull request
Dec 8, 2022
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Reviewed By: cyrjano Differential Revision: D41324297 fbshipit-source-id: 7fb77eb9014adf846ed973a4f8edf82c44127595
99warriors
force-pushed
the
export-D41324297
branch
from
0d9ebb4
to
b452a17
Compare
|
This pull request was exported from Phabricator. Differential Revision: D41324297 |
99warriors
pushed a commit
to 99warriors/captum
that referenced
this pull request
Dec 9, 2022
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Reviewed By: cyrjano Differential Revision: D41324297 fbshipit-source-id: 9204b4c5b75f7bff1093ddf562a1cba4dfb83284
99warriors
force-pushed
the
export-D41324297
branch
from
b452a17
to
facd51c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D41324297 |
99warriors
added a commit
to 99warriors/captum
that referenced
this pull request
Dec 9, 2022
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Differential Revision: https://internalfb.com/D41324297 fbshipit-source-id: d962b940452685e7e488986f11c769633b3d3e2d
|
This pull request was exported from Phabricator. Differential Revision: D41324297 |
99warriors
pushed a commit
to 99warriors/captum
that referenced
this pull request
Dec 19, 2022
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Reviewed By: cyrjano Differential Revision: D41324297 fbshipit-source-id: f0098b83a486b49059c02f359f093ed3b791688c
99warriors
force-pushed
the
export-D41324297
branch
from
facd51c
to
24cd933
Compare
Summary: Pull Request resolved: pytorch#1072 This diff changes the API for implementations of `TracInCPBase` as discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of the `influence` method are changed from `inputs: Tuple, targets: Optional[Tensor]` to `inputs: Union[Tuple[Any], DataLoader]`, which is either a single batch, or a dataloader yielding batches. In both cases, `model(*batch)` is assumed to produce the predictions for a batch, and `batch[-1]` is assumed to be the labels for a batch. This is the same format assumed of the batches yielded by `train_dataloader`. We make this change for 2 reasons - it unifies the assumptions made of the test data and the assumptions made of the training data - for some implementations, we want to allow the test data to be represented by a dataloader. with the old API, there was no clean way to allow both a single as well as a dataloader to be passed in, since a batch required 2 arguments, but a dataloader only requires 1. For now, all implementations only allow `inputs` to be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e. `TracInCP`) to accept a dataloader as `inputs`. Other changes: - changes to make documentation. for example, documentation in `TracInCPBase.influence` now refers to the "test dataset" instead of test batch. - the `unpack_inputs` argument is no longer needed for the `influence` methods, and is removed - the usage of `influence` in all the tests is changed to match new API. - signature of helper methods `_influence_batch_tracincp` and `_influence_batch_tracincp_fast` are changed to match new representation of batches. Reviewed By: cyrjano Differential Revision: D41324297 fbshipit-source-id: 827350795bf2e5c6c1fab2e5ef8f2db1473dfe3d
99warriors
force-pushed
the
export-D41324297
branch
from
24cd933
to
3fde650
Compare
|
This pull request was exported from Phabricator. Differential Revision: D41324297 |
|
This pull request has been merged in fe13596. |
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:
This diff changes the API for implementations of
TracInCPBaseas discussed in https://fb.quip.com/JbpnAiWluZmI. In particular, the arguments representing test data of theinfluencemethod are changed frominputs: Tuple, targets: Optional[Tensor]toinputs: Union[Tuple[Any], DataLoader], which is either a single batch, or a dataloader yielding batches. In both cases,model(*batch)is assumed to produce the predictions for a batch, andbatch[-1]is assumed to be the labels for a batch. This is the same format assumed of the batches yielded bytrain_dataloader.We make this change for 2 reasons
For now, all implementations only allow
inputsto be a tuple (and not a dataloader). This is okay due to inheritance rules. Later on, we will allow some implementations (i.e.TracInCP) to accept a dataloader asinputs.Other changes:
TracInCPBase.influencenow refers to the "test dataset" instead of test batch.unpack_inputsargument is no longer needed for theinfluencemethods, and is removedinfluencein all the tests is changed to match new API._influence_batch_tracincpand_influence_batch_tracincp_fastare changed to match new representation of batches.Reviewed By: cyrjano
Differential Revision: D41324297