Remove lowered execution #21674
Closed
Remove lowered execution #21674
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: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object).
pytorchbot
added
oncall: jit
This was referenced Jun 12, 2019
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
pytorchbot
added
the
module: internals
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
suo
approved these changes
Jun 13, 2019
| if(const auto& sub = module.find_module(field)) { | ||
| gatherParamsFirstClass(*sub, u.user->output(), params); | ||
| } else if(script::Slot* slot = module.find_parameter(field)) { | ||
| if (const auto &sub = module.find_module(field)) { |
Member
There was a problem hiding this comment.
seems like some weird autoformatting going on
| } | ||
| const std::string& name() const { | ||
| return name_; | ||
| const std::string& field_name() const { |
Member
There was a problem hiding this comment.
field_name is a little confusing—unqualified_name?
Contributor
Author
There was a problem hiding this comment.
There are now two names, the name of the Class and the name of the module as a member of a parent module. We still rely on this "field_name" in a few places and it is never qualified. There is a comment that explains this. My hope is that the "Module because a view patch" simply removes it.
| // Create a type representing a Module, | ||
| // These do not have methods, and are not globally registered | ||
| static ClassTypePtr createModuleType(std::shared_ptr<CompilationUnit> module); | ||
| std::shared_ptr<CompilationUnit> cu, bool is_module = false); |
Member
There was a problem hiding this comment.
this is_module flag is a suspicious to me, since it implies that we are branching somewhere on whether something is a module (and so modules are not quite first class actually)
Member
There was a problem hiding this comment.
Ah, seems like it is used for ONNX…
Contributor
Author
There was a problem hiding this comment.
Discussed in person: ModuleType will be introduced as subtype of ClassType in a future comment. This means that the ClassType has additional module-like behavior (e.g. it has a list of "parameter" objects)
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
pytorchbot
added
the
module: custom-operators
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
Remove lowered execution Summary: First class modules can now run on our entire test suite, and benchmarks indicate there is no obvious performance regressions. This PR removes lowered execution and cleans up dead code without changing the Module API. A followup PR further simplifies the Module API now that methods are simply a pair of (function, method_object). gh-metadata: pytorch pytorch 21674 gh/zdevito/63/head
zdevito
added a commit
to zdevito/ATen
that referenced
this pull request
Jun 16, 2019
Summary: Pull Request resolved: pytorch/pytorch#21674 ghimport-source-id: b8e27f0ce9b8b362daf73556ee67457fb5355062 Reviewed By: eellison Differential Revision: D15777726 Pulled By: zdevito fbshipit-source-id: 718ac676c9a1bcf99b856862fd29631d825645da
Contributor
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.
Stack from ghstack:
Summary: First class modules can now run on our entire test suite,
and benchmarks indicate there is no obvious performance regressions.
This PR removes lowered execution and cleans up dead code without
changing the Module API. A followup PR further simplifies the Module API
now that methods are simply a pair of (function, method_object).
Differential Revision: D15777726