[Train V2] Adding Ray Train V2 Codebase, implementing the "Train + Tune API Revamp" REP #49376
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
Signed-off-by: Hongpeng Guo <[email protected]>
hongpeng-guo
requested review from
justinvyu,
matthewdeng,
raulchen and
woshiyyya
as code owners
Signed-off-by: Hongpeng Guo <[email protected]>
hongpeng-guo
force-pushed
the
hpguo/v2/train_v2_init
branch
from
3896f50 to
654a7ff
Compare
Signed-off-by: Hongpeng Guo <[email protected]>
|
Good to review. @matthewdeng @justinvyu |
Signed-off-by: Hongpeng Guo <[email protected]>
srinathk10
pushed a commit
that referenced
this pull request
Jan 3, 2025
…ne API Revamp" REP (#49376) Ray Tune and Ray Train have been tightly coupled since Ray 2.0, when Ray Tune became the common execution engine for both libraries. Ray Train execution invokes Tune’s execution logic under the hood, which leads to a complex, layered system. The original intention behind this was to increase the interoperability of the two libraries, but the dependency of Ray Train on Ray Tune has led to many usability and stability issues, and it has stalled feature development. ray-project/enhancements#57 proposed a much clearer design to improve the **Usability**, **Extensibility**, **Interoperability**, and **Testability**. This PR contains the implementation of the above REP for the revamped Ray Train. This implementation is contained in the `python/ray/train/v2` directory. These changes pave the way for improved feature development and enhanced user experience. Please refer to the REP for details on the design, as well as the remaining changes which will be added shortly in follow-up PRs. --------- Signed-off-by: Hongpeng Guo <[email protected]>
anyadontfly
pushed a commit
to anyadontfly/ray
that referenced
this pull request
Feb 13, 2025
…ne API Revamp" REP (ray-project#49376) Ray Tune and Ray Train have been tightly coupled since Ray 2.0, when Ray Tune became the common execution engine for both libraries. Ray Train execution invokes Tune’s execution logic under the hood, which leads to a complex, layered system. The original intention behind this was to increase the interoperability of the two libraries, but the dependency of Ray Train on Ray Tune has led to many usability and stability issues, and it has stalled feature development. ray-project/enhancements#57 proposed a much clearer design to improve the **Usability**, **Extensibility**, **Interoperability**, and **Testability**. This PR contains the implementation of the above REP for the revamped Ray Train. This implementation is contained in the `python/ray/train/v2` directory. These changes pave the way for improved feature development and enhanced user experience. Please refer to the REP for details on the design, as well as the remaining changes which will be added shortly in follow-up PRs. --------- Signed-off-by: Hongpeng Guo <[email protected]> Signed-off-by: Puyuan Yao <[email protected]>
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
Ray Tune and Ray Train have been tightly coupled since Ray 2.0, when Ray Tune became the common execution engine for both libraries.
Ray Train execution invokes Tune’s execution logic under the hood, which leads to a complex, layered system. The original intention behind this was to increase the interoperability of the two libraries, but the dependency of Ray Train on Ray Tune has led to many usability and stability issues, and it has stalled feature development.
ray-project/enhancements#57 proposed a much clearer design to improve the Usability, Extensibility, Interoperability, and Testability.
This PR contains the implementation of the above REP for the revamped Ray Train. This implementation is contained in the
python/ray/train/v2directory. These changes pave the way for improved feature development and enhanced user experience. Please refer to the REP for details on the design, as well as the remaining changes which will be added shortly in follow-up PRs.Ray Train V2 can be enabled by setting the
RAY_TRAIN_V2_ENABLED=1environment variable on the driver process.