[WIP] [New Model] PrepLGBM #245
Open
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
Innixma
force-pushed
the
prep_lgb
branch
3 times, most recently
from
827655b to
73ce839
Compare
atschalz
force-pushed
the
prep_lgb
branch
2 times, most recently
from
6b62286 to
ea510d6
Compare
Innixma
force-pushed
the
prep_lgb
branch
2 times, most recently
from
83e94fe to
4e26a4b
Compare
Innixma
force-pushed
the
prep_lgb
branch
2 times, most recently
from
bf832b6 to
c0fdcd0
Compare
Innixma
force-pushed
the
prep_lgb
branch
2 times, most recently
from
a5256e4 to
e203149
Compare
Remove debugging code update pytabkit version tmp commit Update update CVSplitter import switch to AG SquashingScaler Update prep_mixin to be faster and fix crashes Update prep_mixin to work with lists instead of tuples Fix yaml serialization fix pyproject.toml Add passthrough_types Update passthrough logic Use `remove_unused_features="false_recursive"` and `post_drop_duplicates=True` Update memory estimation to use n_numeric, and n_categorical estimates from preprocessors. Minor improvements to linear residual Use BulkFeatureGenerator and parallel stage preprocessing, >2x inference speedup. Minor fix Reduce ModelAgnosticPrepMixin._estimate_memory_usage runtime by 4x cleanup Add missing __init__.py stop saving training job logs to s3 add `memory_usage_estimate` tracking cleanup torch memory and gc minor update Update RealTabPFN-2.5 to new prep logic refactor prep logic streamline prep logic update Small bug fix Extend preprocessing logic to TabM, TabPFN2.5, CatBoost and XGBoost; Move preprocessing models to a separate directory minor updates Add skrub to dependencies bug fix in search space Adjust search space Fix bug in memory estimation for OOF-TE Add memory estimation, switch to using AG preprocessors Add functions for linear residual boosting Undo example changes Remove unnecessary old code Add explorative search space Make preprocessors and residuals use the correct random seeds Adjust to new tabarena structure Initialize with AG LightGBM Update PrepGBM: Add residual logic and preprocessors, switch to use LGBModel as superclass; Adapt search space Add new model and search space
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
Issue #, if available:
Description of changes:
Adds a new model that extends the current LightGBM implemented in AutoGluon by adding several preprocessing techniques.
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.