Ensemble bayesian learning clean #3
Open
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
* add repeated kfold * working repeated k fold * working stacking evaluator without changing dataset and no final predict * replace datamanager * fix prediction with stack ensembles * adaptive repeats * working version of stacking with changing dataset preserving categorical info * working version of ensemble selection per layer, TODO: send predictions according to the weights associated with the model * finish previous todo: send predictions according to the weights associated with the model * working version of base repeat stacked ensembles, todo: check if other methods still work, add autogluon stacking * working all stacking versions * rename optimisation stacking ensemble * Add autogluon stacking (#1) * add working traditional models according autofluon * working pytorch embedding with skew and embed column splitting * work in progress: autogluon ensembling * working autogluon ensemble * important fix for more than 2 stacking layers * fix for running more than 2 stacking layers * working autogluon with default nn config from autogluon * working xgboost model * add configurationspace to traditional classification models * working autogluon stacking and stacking optimisation, todo: search for autogluon models and post hoc ensemble selection for ensemble optimisation * added post fit ensemble optimization, working per layer selection, repeat models, stacking optimisation * update config space for search, fix stratified resampling, fix printing model with weights for soe * fix running traditional pipeline for all the ensembles, fix get config from run history * fix cut off num run for all ensembles * __init__ file for column splittin * all requirements * add __init__.py for trad ml * pass smbo class to custom callback * early stop also ensemble opt * remove -1 from autogluon stacking * reduce number of models stored after stcking * fix issue wiuth null identifiers in selected ensemble identifiers * remove pointless line for debug * set multiprocessing context to forkserver for n workers 1 * fix error when all repeats do not finish * examples changed
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.
Types of changes
Note that a Pull Request should only contain one of refactoring, new features or documentation changes.
Please separate these changes and send us individual PRs for each.
For more information on how to create a good pull request, please refer to The anatomy of a perfect pull request.
Checklist:
Description
Motivation and Context
How has this been tested?