Ops: Adds poetry template to replace setup.py and requirements.txt #18
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
I've also installed all the libraries requirements from requirements.txt and updated aif360 to a fork that contains a bug fix.
JoaoGranja
added a commit
that referenced
this pull request
Mar 13, 2023
* Adds poetry template to replace setup.py and requirements.txt (#18) Added all libraries requirements from requirements.txt. Updated aif360 package to a fork that contains a bug fix. * Add PyPI README. Update pyproject.toml with urls, classifiers, license, authors --------- Co-authored-by: Ahmed Elghareeb <ahmeds.elghareeb@gmail.com>
JoaoGranja
added a commit
that referenced
this pull request
Mar 17, 2023
* Add models_comparison.py file. This file has functions for comparing the performance of 2 models using mcnemar test or 5x2cv paired t test. * Add method to FAIR class which compare the performance and fairness metric of several bias mitigation methods. Apply bias mitigation before comparing the performance of 2 methods. * Add plot option to compare_mitigation_methods method * Add 'fairness metric score' to compute paired test 5x2cv. * Add discrimination threshold argument to FAIR class. Add draft threshold.py file to calculate discrimination threshold. * Update ml model before assessing fairness metrics in threshold.py. Enable matplotlib to show figures. * Scale fairness metrics value. * Added docstrings and did some refactoring code * Fix bug with FAIR class. Refactor stats.py code. * Refactor threshold.py code * Fix bug with FAIR unit tests. Add threshold to __init__.py. * Add discrimination_threshold function to threshold module. * Add FAIR unit tests for new methods "compare_mitigation_methods" and "model_mitigation" * Add stats unit tests. * Add threshold unit tests and check input arguments. * Add docstrings to new methods/functions. Remove metric_name property of FAIR class. * Improve docstrings to stats functions. * Issues fix * Bug fixed in FAIR bias mitigation method. * Add binary threshold score function to compute a score based on a discrimination threshold. * Add input metrics to choose which metrics to show on figure. * Add unit test to binary_threshold_score function. * Add some notebook examples with different uses cases. Add accuracy metric to the Discrimination Threshold class. * Refactor threshold.py code * Add docstrings to DiscriminationThreshold class and discrimination_threshold function. Refactor threshold.py code. * Fix some issues with unit tests. Copy internally the machine learning models. * Revert matplotlib backend after using function of dalex package. Update threshold-accuracy.ipynb and threshold-f1-score.ipynb notebooks * Increase default number of thresholds from 50 to 100. Update threshold-queue-rate.ipynb notebook. * Add utility cost option to binary_threshold_score function. Add threshold to paired t test. Add threshold-utility-cost.ipynb notebook. * Fix issue #16 - Add random_seed argument to FAIR class to set numpy random seed. * Fix issue #11 - Inform user of 1/metric calculation. * Add a check for testing data structure. Rename variables. * Add random seed to unit tests. * Update notebooks adding random seed to get reproducible results. * Fix issue #10 - The training and testing sets should use the same correlation remover object. I * Add option to compute the optimal threshold just evaluating the model (no training is computed). * Add the option to compute the optimal threshold when comparing the mitigation methods. * Add minimum and maximum bound to threshold range. * For paired t-test method, when a FAIR object and mitigation method are provided as input arguments, the second model to use will be computed, otherwise the input model_2 is used. * Update notebooks with new paired t-test and discrimination threshold function options * Update accuracy notebook * Change discrimination_threshold arguments, data and target variable, to X, y. Update discrimination_threshold calls and unit tests. * Update stats.py with the new discrimination_threshold arguments. * Change threshold-accuracy.ipynb notebook to use the new arguments of the discrimination_threshold function. * Change threshold-utility-cost.ipynb, threshold-queue-rate.ipynb and threshold-f1-score.ipynb notebooks to use the new arguments of the discrimination_threshold function. * Use poetry (#19) * Adds poetry template to replace setup.py and requirements.txt (#18) Added all libraries requirements from requirements.txt. Updated aif360 package to a fork that contains a bug fix. * Add PyPI README. Update pyproject.toml with urls, classifiers, license, authors --------- Co-authored-by: Ahmed Elghareeb <ahmeds.elghareeb@gmail.com> * Update ci.yml with poetry * Update poetry.lock * Update dependencies versions * Update ci.yaml to use poetry * Update ci.yaml to use poetry * Update ci.yaml to use poetry * Update ci.yaml to use poetry * Update ci.yaml to use poetry * Update ci.yaml to use poetry * Update ci.yaml to use poetry * Update ci.yaml to use poetry * Update pytest version requirement * Update poetry.lock * Update ci.yaml * Update ci.yaml * Update setup.py to use PyPI_README.md --------- Co-authored-by: Ahmed Elghareeb <ahmeds.elghareeb@gmail.com>
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.
I've also installed all the libraries requirements from requirements.txt
and updated aif360 to a fork that contains a bug fix.
This PR targets this issue #17
The file will need some more improvements to match the existing setup.py but the baseline requirements are there.