-
Notifications
You must be signed in to change notification settings - Fork 204
Add pruning checkpoints for the compress algorithm #607
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
Merged
Merged
Changes from all commits
Commits
Show all changes
52 commits
Select commit
Hold shift + click to select a range
694c317
Add decilm modelling code
danielkorzekwa 991659f
Add decilm modelling code.
danielkorzekwa 8489cee
Add transformers codebase
danielkorzekwa f0afefe
Add transformers code
danielkorzekwa b3ed5bc
Add decilm modelling code
danielkorzekwa a700da5
Add decilm modelling code
danielkorzekwa b59b679
Correct licence headers
danielkorzekwa 1abdf3e
Correct licence headers
danielkorzekwa 66609b1
Add decilm code
danielkorzekwa 7da0a8a
Add decilm code
danielkorzekwa 6e09a81
Add decilm code
danielkorzekwa 2e3f5da
Add decilm code
danielkorzekwa 418890e
Add decilm code
danielkorzekwa 01f4fc1
Make llama3 converter self-contained (no deps on internal Nvidia code)
danielkorzekwa c57eed4
Add common module
danielkorzekwa 3dc37b3
module refactoring
danielkorzekwa 10ffdfe
refactoring
danielkorzekwa 27a4456
add shared_checkpointing_utils
danielkorzekwa b0e22b7
Add json tools
danielkorzekwa 52e7827
add logger
danielkorzekwa f5c1c87
import refactoring
danielkorzekwa 0aa6320
add post_init_sparse module
danielkorzekwa 35d0dbc
Add post_init_sparse
danielkorzekwa e39a1ad
merginy hydra.py and hydra_utils.py
danielkorzekwa 1bd0c67
Add integrationt test for attention pruning
danielkorzekwa 0ecd52b
add score_pruning_activations
danielkorzekwa 278c6b7
import refactoring
danielkorzekwa 7a0af16
add dist_utils
danielkorzekwa 0f0cbbd
Add validate_model
danielkorzekwa cb5cf25
Add activation scoring hooks for pruning
danielkorzekwa 6f82a67
make validate_model self-contained
danielkorzekwa a87fb79
updage validatete_pipeline to use DeciLMForCausalLM from modelopt
danielkorzekwa b227521
fix imports
danielkorzekwa ca7ab3f
add sewing_kit
danielkorzekwa a7a4adc
add sewing_kit
danielkorzekwa ad84c26
fix imports
danielkorzekwa 3d7e8a2
fix imports
danielkorzekwa 3d755b2
add pruning_ckpts
danielkorzekwa 845d453
add pruning_ckpts
danielkorzekwa 4fd921b
import refactoring
danielkorzekwa 3641847
refactor imports
danielkorzekwa 8d6333b
import refactoring
danielkorzekwa b6b7ca9
Merge branch 'feature/compress' into dkorzekwa/pruning_ckpts_1
danielkorzekwa 7ab69e6
Delete not needed mistral tokenizer
danielkorzekwa 2217a2a
Improve doc strings
danielkorzekwa a281ff7
Delete empty module
danielkorzekwa c1fb32c
Add doc string
danielkorzekwa 5203169
Add doc string + remove references to 'lustre'
danielkorzekwa e2eee60
Add typeguard to compress dependencies in setup.py
danielkorzekwa 6e26074
Improve dpcs
danielkorzekwa 42da180
fix imports
danielkorzekwa 6200962
fix import ordering
danielkorzekwa File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
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
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
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
Large diffs are not rendered by default.
Oops, something went wrong.
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
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
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
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
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
Oops, something went wrong.
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.
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.
Fyi, All this is natively supported in DynamicModule if we implement the DeciLM model as a DynamicModule. That way
child_init.pycould also be greatly simplifiedWe just need to set hyperparameter active value to the pruned value and assign a ranking then exporting it will export sorted + pruned module. If ranking order is not assigned, then it will just truncate.
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.
will create an issue (once gitlab is up again:)