-
Notifications
You must be signed in to change notification settings - Fork 65
chore(release): merge set of changes for v2.4.0 #439
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Signed-off-by: Dushyant Behl <[email protected]>
fix: broken README.md link
* Allow hf dataset id to be loaded by training_data_path Signed-off-by: Dushyant Behl <[email protected]> * update README Signed-off-by: Dushyant Behl <[email protected]> * minor changes Signed-off-by: Abhishek <[email protected]> --------- Signed-off-by: Dushyant Behl <[email protected]> Signed-off-by: Abhishek <[email protected]> Co-authored-by: Abhishek <[email protected]>
* feat: accelerated MoE dataclass and init Signed-off-by: Will Johnson <[email protected]> * fix: author's note Signed-off-by: Will Johnson <[email protected]> * feat: accelerated moe in acceleration framework Signed-off-by: Will Johnson <[email protected]> * feat: accelerated moe to sft_trainer Signed-off-by: Will Johnson <[email protected]> * feat: fmt, testing Signed-off-by: Will Johnson <[email protected]> * fix: rename accelerated moe to fast moe Signed-off-by: Will Johnson <[email protected]> * test: add testing for scatter moe on accel framework Signed-off-by: Will Johnson <[email protected]> * fix: model, dtype, assertions Signed-off-by: Will Johnson <[email protected]> * fix: post init check removed from FastMoe, experimental set to True Signed-off-by: Will Johnson <[email protected]> * fix: if non-iterable nested dataclass, still initialize Signed-off-by: Will Johnson <[email protected]> * test: add failing test for wrong ep_degree Signed-off-by: Will Johnson <[email protected]> * fix: actually expect failure Signed-off-by: Will Johnson <[email protected]> * test: make sure fast moe doesn't work with non-moe model Signed-off-by: Will Johnson <[email protected]> * fix: regex of new test Signed-off-by: Will Johnson <[email protected]> * comment: explain iterable unpacking Signed-off-by: Will Johnson <[email protected]> * docs: fast MOE in README Signed-off-by: Will Johnson <[email protected]> * docs: Add note for post-processing Signed-off-by: Will Johnson <[email protected]> * fix: Dockerfile Signed-off-by: Will Johnson <[email protected]> * test: fix params Signed-off-by: Will Johnson <[email protected]> * fix: file path Signed-off-by: Will Johnson <[email protected]> * fix: expand on docs, remove from Dockerfile, move iterable data to else statement Signed-off-by: Will Johnson <[email protected]> * lint Signed-off-by: Will Johnson <[email protected]> * fix: spelling Signed-off-by: Will Johnson <[email protected]> --------- Signed-off-by: Will Johnson <[email protected]>
…ate (#430) * fix: allow for padding free + pretraining Signed-off-by: Harikrishnan Balagopal <[email protected]> * add data collator for padding free plugin scenario to be used for extended pretraining Signed-off-by: Dushyant Behl <[email protected]> * fix: update value error Signed-off-by: Mehant Kammakomati <[email protected]> * fix: delete images only when exists Signed-off-by: Mehant Kammakomati <[email protected]> --------- Signed-off-by: Harikrishnan Balagopal <[email protected]> Signed-off-by: Dushyant Behl <[email protected]> Signed-off-by: Mehant Kammakomati <[email protected]> Co-authored-by: Harikrishnan Balagopal <[email protected]> Co-authored-by: Mehant Kammakomati <[email protected]>
#434) * fix: function name from requires_agumentation to requires_augmentation Signed-off-by: Will Johnson <[email protected]> * fix: file path Signed-off-by: Will Johnson <[email protected]> * fmt Signed-off-by: Will Johnson <[email protected]> --------- Signed-off-by: Will Johnson <[email protected]>
Signed-off-by: Abhishek <[email protected]>
Signed-off-by: Will Johnson <[email protected]>
|
Thanks for making a pull request! 😃 |
|
This looks good, but could we wait 1 day to merge? Waiting to test something for PR #422 and see if I can get it merged soon |
Abhishek-TAMU
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The current commits are in sync with the main branch. Looks good to me.
release 2.4.0!