Reorg for converters elu and selu (FX Converter Refactor [7/N]) <Target: converter_reorg_proto> #1903
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
There was a problem hiding this comment.
There are some changes that do not conform to Python style guidelines:
--- py/torch_tensorrt/fx/converters/aten_ops_converters.py 2023-05-10 05:51:44.730358 +0000
+++ py/torch_tensorrt/fx/converters/aten_ops_converters.py 2023-05-10 05:52:04.953773 +0000
@@ -177,11 +177,11 @@
args: Tuple[Argument, ...],
kwargs: Dict[str, Argument],
name: str,
) -> Union[TRTTensor, Sequence[TRTTensor]]:
- if (len(args) > 2) :
+ if len(args) > 2:
return activation.selu(
network,
target,
SourceIR.ATEN,
name,
--- py/torch_tensorrt/fx/converters/impl/activation.py 2023-05-10 05:51:44.730358 +0000
+++ py/torch_tensorrt/fx/converters/impl/activation.py 2023-05-10 05:52:05.422983 +0000
@@ -136,6 +136,6 @@
source_ir,
name,
operation_type,
input_val,
dyn_range_fn=elu_dyn_range_fn,
- )
\ No newline at end of file
+ )
--- py/torch_tensorrt/fx/converters/nn_ops_converters.py 2023-05-10 05:51:44.730358 +0000
+++ py/torch_tensorrt/fx/converters/nn_ops_converters.py 2023-05-10 05:52:05.556664 +0000
@@ -49,7 +49,7 @@
network=network,
target="torch.nn.functional.selu",
source_ir=SourceIR.NN,
name=layer_name,
input_val=kwargs["input"],
- alpha = kwargs["alpha"]
+ alpha=kwargs["alpha"],
)
--- py/torch_tensorrt/fx/converters/acc_ops_converters.py 2023-05-10 05:51:44.730358 +0000
+++ py/torch_tensorrt/fx/converters/acc_ops_converters.py 2023-05-10 05:52:07.665077 +0000
@@ -1043,11 +1043,11 @@
target: Target,
args: Tuple[Argument, ...],
kwargs: Dict[str, Argument],
name: str,
) -> Union[TRTTensor, Sequence[TRTTensor]]:
-
+
return activation.elu(
network,
target,
SourceIR.ACC,
name,
@@ -1070,11 +1070,10 @@
target,
SourceIR.ACC,
name,
kwargs["input"],
)
-
@tensorrt_converter(acc_ops.softsign)
def acc_ops_softsign(
network: TRTNetwork,
--- py/torch_tensorrt/fx/test/converters/aten_op/test_elu_aten.py 2023-05-10 05:51:44.734358 +0000
+++ py/torch_tensorrt/fx/test/converters/aten_op/test_elu_aten.py 2023-05-10 05:52:08.075263 +0000
@@ -46,6 +46,6 @@
TestModule(), input_specs, expected_ops={torch.ops.aten.elu.default}
)
if __name__ == "__main__":
- run_tests()
\ No newline at end of file
+ run_tests()
--- py/torch_tensorrt/fx/test/converters/aten_op/test_selu_aten.py 2023-05-10 05:51:44.734358 +0000
+++ py/torch_tensorrt/fx/test/converters/aten_op/test_selu_aten.py 2023-05-10 05:52:08.233990 +0000
@@ -46,6 +46,6 @@
TestModule(), input_specs, expected_ops={torch.ops.aten.elu.default}
)
if __name__ == "__main__":
- run_tests()
\ No newline at end of file
+ run_tests()1696cd2 to
12f545c
Compare
f87d034 to
84ff793
Compare
wushirong
approved these changes
May 25, 2023
84ff793 to
67cac89
Compare
wushirong
reviewed
May 30, 2023
Comment on lines
+182
to
+189
| if len(args) > 2: | ||
| return activation.selu( | ||
| network, | ||
| target, | ||
| SourceIR.ATEN, | ||
| name, | ||
| args[0], | ||
| ) |
There was a problem hiding this comment.
what is the purpose of this if logic?
There was a problem hiding this comment.
The aten trace for both selu and elu comes as torch.ops.aten.elu.default, but with different args. elu has args[0] and args[1] for alpha, while selu has only one arg args[0]. Just a way to differentiate between the two.
Adding selu converter Python linting correction
67cac89 to
67ade93
Compare
wushirong
approved these changes
Jun 10, 2023
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.
No description provided.