Upgrade Kubeflow Pipelines to version v2 sdk #1
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: Updating to a new version of pyre since the one we used has been removed from pip (too old). Recently it seems pyre configs have changed and we need a special argument to enable strict "no Any type" checks. Differential Revision: D75996666 Co-authored-by: Daniel Ohayon <[email protected]>
…a-pytorch#1075) (meta-pytorch#1075) Summary: Pull Request resolved: meta-pytorch#1075 Add support for cases like: ```lang=python # some_file.py # ==================== def my_component(...) -> specs.AppDef: ... # other_file.py # ==================== from some_file import my_component ``` where the component is invoked with `torchx run ... other_file.py:my_component` This was currently failing with a validation error because in the step where we inspect the AST of the component, we assume that the file where the component is being looked up is the same as the file where it is implemented. Reviewed By: kiukchung Differential Revision: D75496839 Co-authored-by: Daniel Ohayon <[email protected]>
Differential Revision: D72677312 Pull Request resolved: meta-pytorch#1069
…'s describe API Differential Revision: D76485112 Pull Request resolved: meta-pytorch#1080
Co-authored-by: Alexander Zhipa <[email protected]>
…x/runner:lib_core with no plugin bundling Differential Revision: D77400710 Pull Request resolved: meta-pytorch#1084
…x/runner:lib_core with no plugin bundling (meta-pytorch#1083) Summary: See details in [torchx-lite](https://docs.google.com/document/d/1Y_DzoRMAKhqbzE2LQDb7k3H5R-4PnkYdA9R4YLLFLk4/edit?tab=t.0#heading=h.unqedx38434n) doc. Creates a pure core library target for `torchx.runner` called `//torchx/runner:lib_core`. Similar to D77394064 which creates `//torchx/schedulers:lib_core` NOTE: I added a `autodeps_redirect` to `//torchx/runner:lib` from `:lib_core`. I've noticed that autodeps honors this in some cases but not in others (read IMPORTANT note below if you are finding that autodeps is removing `:lib` in favor of `:lib_core`) IMPORTANT: If you wind up here because autodeps is removing `//torchx/runner:lib` in favor of `//torchx/runner:lib_core`, you have two options (see below). OPTION 2 is preferred since you'll only pull in what you need (not the entire universe of plugins). OPTION 1. Add `# manual` tag to the existing `//torchx/runner:lib` ``` ... deps = [ "//torchx/runner:lib_core", # <-- autodeps will still add this (which is OK) "//torchx/runner:lib", # manual ... ] ... ``` OPTION 2: Accept the swap to `//torchx/runner:lib_core` and add any plugins that you need with `# manual` ``` ... deps = [ "//torchx/runner:lib_core", "//torchx/schedulers/fb:mast_scheduler", # manual (mast scheduler plugin) ... ] ... ``` Reviewed By: highker Differential Revision: D77400710
Differential Revision: D77621041 Pull Request resolved: meta-pytorch#1088
… importlib_metadata Differential Revision: D77619564 Pull Request resolved: meta-pytorch#1087
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.
Test plan: