diff --git a/README.md b/README.md index 2b769b5a..962a4c1c 100644 --- a/README.md +++ b/README.md @@ -8,6 +8,8 @@ [![PyPI - License](https://img.shields.io/github/license/holistic-ai/holisticai)](https://img.shields.io/github/license/holistic-ai/holisticai) [![PyPI - Downloads](https://img.shields.io/pypi/dm/holisticai)](https://img.shields.io/pypi/dm/holisticai) [![Slack](https://img.shields.io/badge/Slack-Join-blue)](https://join.slack.com/t/holisticaicommunity/shared_invite/zt-2jamouyrn-BrMfeoBZIHT8HbLzB3P9QQ) +[![jax_badge][jax_badge_link]](https://github.com/google/jax) + --- @@ -107,3 +109,6 @@ brew install cmake ### Contributing We welcome contributions from the community To learn more about contributing to Holistic AI, please refer to our [Contributing Guide](CONTRIBUTING.md). + + +[jax_badge_link]: https://img.shields.io/badge/JAX-Accelerated-9cf.svg?style=flat-square&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAC0AAAAaCAYAAAAjZdWPAAAIx0lEQVR42rWWBVQbWxOAkefur%2B7u3les7u7F3ZIQ3N2tbng8aXFC0uAuKf2hmlJ3AapIgobMv7t0w%2Ba50JzzJdlhlvNldubeq%2FY%2BXrTS1z%2B6sttrKfQOOY4ns13ecFImb47pVvIkukNe4y3Junr1kSZ%2Bb3Na248tx7rKiHlPo6Ryse%2F11NKQuk%2FV3tfL52yHtXm8TGYS1wk4J093wrPQPngRJH9HH1x2fAjMhcIeIaXKQCmd2Gn7IqSvG83BueT0CMkTyESUqm3vRRggTdOBIb1HFDaNl8Gdg91AFGkO7QXe8gJInpoDjEXC9gbhtWH3rjZ%2F9yK6t42Y9zyiC1iLhZA8JQe4eqKXklrJF0MqfPv2bc2wzPZjpnEyMEVlEZCKQzYCJhE8QEtIL1RaXEVFEGmEaTn96VuLDzWflLFbgvqUec3BPVBmeBnNwUiakq1I31UcPaTSR8%2B1LnditsscaB2A48K6D9SoZDD2O6bELvA0JGhl4zIYZzcWtD%2BMfdvdHNsDOHciXwBPN18lj7sy79qQCTNK3nxBZXakqbZFO2jHskA7zBs%2BJhmDmr0RhoadIZjYxKIVHpCZngPMZUKoQKrfEoz1PfZZdKAe2CvP4XnYE8k2LLMdMumwrLaNlomyVqK0UdwN%2BD7AAz73dYBpPg6gPiCN8TXFHCI2s7AWYesJgTabD%2FS5uXDTuwVaAvvghncTdk1DYGkL0daAs%2BsLiutLrn0%2BRMNXpunC7mgkCpshfbw4OhrUvMkYo%2F0c4XtHS1waY4mlG6To8oG1TKjs78xV5fAkSgqcZSL0GoszfxEAW0fUludRNWlIhGsljzVjctr8rJOkCpskKaDYIlgkVoCmF0kp%2FbW%2FU%2F%2B8QNdXPztbAc4kFxIEmNGwKuI9y5gnBMH%2BakiZxlfGaLP48kyj4qPFkeIPh0Q6lt861zZF%2BgBpDcAxT3gEOjGxMDLQRSn9XaDzPWdOstkEN7uez6jmgLOYilR7NkFwLh%2B4G0SQMnMwRp8jaCrwEs8eEmFW2VsNd07HQdP4TgWxNTYcFcKHPhRYFOWLfJJBE5FefTQsWiKRaOw6FBr6ob1RP3EoqdbHsWFDwAYvaVI28DaK8AHs51tU%2BA3Z8CUXvZ1jnSR7SRS2SnwKw4O8B1rCjwrjgt1gSrjXnWhBxjD0Hidm4vfj3e3riUP5PcUCYlZxsYFDK41XnLlUANwVeeILFde%2BGKLhk3zgyZNeQjcSHPMEKSyPPQKfIcKfIqCf8yN95MGZZ1bj98WJ%2BOorQzxsPqcYdX9orw8420jBQNfJVVmTOStEUqFz5dq%2F2tHUY3LbjMh0qYxCwCGxRep8%2FK4ZnldzuUkjJLPDhkzrUFBoHYBjk3odtNMYoJVGx9BG2JTNVehksmRaGUwMbYQITk3Xw9gOxbNoGaA8RWjwuQdsXdGvpdty7Su2%2Fqn0qbzWsXYp0nqVpet0O6zzugva1MZHUdwHk9G8aH7raHua9AIxzzjxDaw4w4cpvEQlM84kwdI0hkpsPpcOtUeaVM8hQT2Qtb4ckUbaYw4fXzGAqSVEd8CGpqamj%2F9Q2pPX7miW0NlHlDE81AxLSI2wyK6xf6vfrcgEwb0PAtPaHM1%2BNXzGXAlMRcUIrMpiE6%2Bxv0cyxSrC6FmjzvkWJE3OxpY%2BzmpsANFBxK6RuIJvXe7bUHNd4zfCwvPPh9unSO%2BbIL2JY53QDqvdbsEi2%2BuwEEHPsfFRdOqjHcjTaCLmWdBewtKzHEwKZynSGgtTaSqx7dwMeBLRhR1LETDhu76vgTFfMLi8zc8F7hoRPpAYjAWCp0Jy5dzfSEfltGU6M9oVCIATnPoGKImDUJNfK0JS37QTc9yY7eDKzIX5wR4wN8RTya4jETAvZDCmFeEPwhNXoOlQt5JnRzqhxLZBpY%2BT5mZD3M4MfLnDW6U%2Fy6jkaDXtysDm8vjxY%2FXYnLebkelXaQtSSge2IhBj9kjMLF41duDUNRiDLHEzfaigsoxRzWG6B0kZ2%2BoRA3dD2lRa44ZrM%2FBW5ANziVApGLaKCYucXOCEdhoew5Y%2Btu65VwJqxUC1j4lav6UwpIJfnRswQUIMawPSr2LGp6WwLDYJ2TwoMNbf6Tdni%2FEuNvAdEvuUZAwFERLVXg7pg9xt1djZgqV7DmuHFGQI9Sje2A9dR%2FFDd0osztIRYnln1hdW1dff%2B1gtNLN1u0ViZy9BBlu%2BzBNUK%2BrIaP9Nla2TG%2BETHwq2kXzmS4XxXmSVan9KMYUprrbgFJqCndyIw9fgdh8dMvzIiW0sngbxoGlniN6LffruTEIGE9khBw5T2FDmWlTYqrnEPa7aF%2FYYcPYiUE48Ul5jhP82tj%2FiESyJilCeLdQRpod6No3xJNNHeZBpOBsiAzm5rg2dBZYSyH9Hob0EOFqqh3vWOuHbFR5eXcORp4OzwTUA4rUzVfJ4q%2FIa1GzCrzjOMxQr5uqLAWUOwgaHOphrgF0r2epYh%2FytdjBmUAurfM6CxruT3Ee%2BDv2%2FHAwK4RUIPskqK%2Fw4%2FR1F1bWfHjbNiXcYl6RwGJcMOMdXZaEVxCutSN1SGLMx3JfzCdlU8THZFFC%2BJJuB2964wSGdmq3I2FEcpWYVfHm4jmXd%2BRn7agFn9oFaWGYhBmJs5v5a0LZUjc3Sr4Ep%2FmFYlX8OdLlFYidM%2B731v7Ly4lfu85l3SSMTAcd5Bg2Sl%2FIHBm3RuacVx%2BrHpFcWjxztavOcOBcTnUhwekkGlsfWEt2%2FkHflB7WqKomGvs9F62l7a%2BRKQQQtRBD9VIlZiLEfRBRfQEmDb32cFQcSjznUP3um%2FkcbV%2BjmNEvqhOQuonjoQh7QF%2BbK811rduN5G6ICLD%2BnmPbi0ur2hrDLKhQYiwRdQrvKjcp%2F%2BL%2BnTz%2Fa4FgvmakvluPMMxbL15Dq5MTYAhOxXM%2FmvEpsoWmtfP9RxnkAIAr%2F5pVxqPxH93msKodRSXIct2l0OU0%2FL4eY506L%2B3GyJ6UMEZfjjCDbysNcWWmFweJP0Jz%2FA0g2gk80pGkYAAAAAElFTkSuQmCC diff --git a/docs/source/bmbench/index.rst b/docs/source/bmbench/index.rst new file mode 100644 index 00000000..274eb0c7 --- /dev/null +++ b/docs/source/bmbench/index.rst @@ -0,0 +1,6 @@ + +BMBENCH +========= + +The bias mitigation benchmark (BMBENCH) is a framework to simple evaluate the performance of bias mitigation algorithms. The benchmark is based on four empirical machine learning tasks: binary classification, regression, multi-class classification, and clustering. The benchmark is designed to be simple to use and easy to extend at the top of `holisticai` library. + diff --git a/docs/source/getting_started/datasets.csv b/docs/source/getting_started/datasets.csv index 10093ef3..67171561 100644 --- a/docs/source/getting_started/datasets.csv +++ b/docs/source/getting_started/datasets.csv @@ -1,9 +1,9 @@ adult,Adult,Binary Classification, race / sex clinical_records,Heart Clinical Records,Binary Classification, sex -law_school,Law School,Binary Classification, race / gender +law_school,Law School,Regression, race / gender lastfm,Last FM,Recommender, -student,Student,Binary Classification, sex / address -student_multiclass,Student (Multiclass),Multiclassification, sex / address +student,Student, Regression, sex / address +student_multiclass,Student (Multiclass), Multiclassification, sex / address us_crime,US Crime,Regression, race us_crime_multiclass,US Crime (Multiclass),Regression, race german_credit, German Credit, Binary Classification, sex diff --git a/docs/source/index.rst b/docs/source/index.rst index 9a958b4b..8ae688bc 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -65,6 +65,15 @@ Welcome to the Holistic AI library! This is an open-source tool to assess and im Learn how to contribute to the library + .. grid-item-card:: Bias Mitigation Benchmark (BMBENCH) + :img-top: _static/contributor.svg + :class-card: intro-card + :link: bmbench/index.html + :link-alt: bmbench + :shadow: md + + Bias mitigation benchmark for AI systems + .. admonition:: Learn more about the Holistic AI company :class: dropdown admonition-youtube diff --git a/src/holisticai/benchmark/BMBENCH.md b/src/holisticai/benchmark/BMBENCH.md new file mode 100644 index 00000000..aaf62c0e --- /dev/null +++ b/src/holisticai/benchmark/BMBENCH.md @@ -0,0 +1,43 @@ +# Empirical Benchmarking of Algorithmic Fairness in Machine Learning Models + +## Motivation + +The development and assessment of bias mitigation methods require rigorous benchmarks. This paper introduces BMBench, a comprehensive benchmarking framework to evaluate bias mitigation strategies across multitask machine learning predictions (binary classification, multiclass classification, regression, and clustering). Our benchmark leverages state-of-the-art and proposed datasets to improve fairness research, offering a broad spectrum of fairness metrics for a robust evaluation of bias mitigation methods. We provide an open-source repository to allow researchers to test and refine their bias mitigation approaches easily, promoting advancements in creating fair machine learning models. + +## Get started + +### Installation and usage + +```bash +pip install holisticai +``` + +```python +from holisticai.benchmark import BiasMitigationBenchmark + +stage = # preprocessing, inprocessing, postprocessing +task = # binary_classification, multiclass, regression, clustering + +benchmark = BiasMitigationBenchmark(task, stage) + +my_mitigator = MyMitigator() + +my_results = benchmark.run(custom_mitigator=my_mitigator) +benchmark.submit() +``` + +There are many other functionalities available in the `BiasMitigationBenchmark` class. Please refer to the [API documentation](https://holisticai.github.io/holistic-ai/). + + +## How to cite + +If you use BMBench in your research, please cite the following paper: + +```bibtex +@article{dacosta2025bmbench, + title={BMBENCH: Empirical Benchmarking of Algorithmic Fairness in Machine Learning Models}, + author={da Costa, Kleyton and Munoz, Cristian and Fernandez, Franklin and Modenesi, Bernardo and Kazim, Emre and Koshiyama, Adriano}, + url={}, + year={2025} +} +``` diff --git a/src/holisticai/benchmark/__init__.py b/src/holisticai/benchmark/__init__.py new file mode 100644 index 00000000..16dea9be --- /dev/null +++ b/src/holisticai/benchmark/__init__.py @@ -0,0 +1,5 @@ +from holisticai.benchmark._bias import BiasMitigationBenchmark + +__all__ = [ + "BiasMitigationBenchmark", +] diff --git a/src/holisticai/benchmark/_bias.py b/src/holisticai/benchmark/_bias.py new file mode 100644 index 00000000..8dcb84a3 --- /dev/null +++ b/src/holisticai/benchmark/_bias.py @@ -0,0 +1,481 @@ +from __future__ import annotations + +import logging +import math +import time +from functools import partial +from typing import Any + +import matplotlib.pyplot as plt +import numpy as np +import pandas as pd +import seaborn as sns +from sklearn.preprocessing import StandardScaler + +from holisticai.datasets._load_dataset import _load_dataset_benchmark +from holisticai.pipeline import Pipeline +from holisticai.utils._plotting import get_colors +from holisticai.utils.benchmark_config import DATASETS, METRICS, MITIGATORS, MODELS + +logging.basicConfig(level=logging.INFO) +logger = logging.getLogger(__name__) + + +class BiasMitigationBenchmark: + METRICS = METRICS + DATASETS = DATASETS + MITIGATORS = MITIGATORS + MODELS = MODELS + + def __init__(self, task_type: str, stage: str): + if task_type not in self.MODELS: + raise ValueError(f"Invalid task type. Choose from {list(self.DATASETS.keys())}") + self.task_type = task_type + self.metric = self.METRICS[task_type] + self.stage = stage + + def available_settings(self): + """ + Get the available settings for the benchmark. + + Returns + ------- + dict + Available settings. + """ + + data = list(self.DATASETS[self.task_type]) + mitigator = list(obj.__class__.__name__ for obj in self.MITIGATORS[self.task_type][self.stage]) # noqa: C400 + + print(f"Available datasets for {self.task_type}: {data}") # noqa: T201 + print(f"Available mitigators for {self.task_type} with {self.stage}: {mitigator}") # noqa: T201 + + def get_datasets(self): + """ + Get the datasets for the given task type. + + Returns + ------- + dict + Datasets. + """ + return self.DATASETS[self.task_type] + + def get_mitigators(self): + """ + Get the mitigators for the given task type and stage. + + Returns + ------- + dict + Mitigators. + """ + if self.stage: + if self.stage not in self.MITIGATORS[self.task_type]: + raise ValueError(f"Invalid stage. Choose from {list(self.MITIGATORS[self.task_type].keys())}") + return {self.stage: self.MITIGATORS[self.task_type][self.stage]} + return self.MITIGATORS[self.task_type] + + def get_table(self): + """ + Get the benchmark results as a table. + + Returns + ------- + pd.DataFrame + Benchmark results. + """ + data = pd.read_parquet( + f"https://huggingface.co/datasets/holistic-ai/bias_mitigation_benchmark/resolve/main/data/benchmark_{self.task_type}_{self.stage}.parquet", + ) + return data.set_index("Unnamed: 0").rename_axis("") + + def get_heatmap(self, fig_size=(10, 5), output_path=None): + """ + Create a heatmap based on the benchmark results. + + Returns + ------- + fig + Heatmap. + """ + plt.style.use("ggplot") + plt.rc("font", size=14) + + fig, ax = plt.subplots(figsize=fig_size) + data = self.get_table()[3:].T + sns.heatmap(data, annot=True, fmt=".4f", cmap=get_colors(len(data.columns)), linewidths=1, ax=ax) + # color bar title + cbar = ax.collections[0].colorbar + cbar.set_label("Balanced Fairness Score (mean)", fontsize=14) + if output_path: + plt.savefig(output_path, bbox_inches="tight") + plt.close() + return fig + + def get_plot(self, output_path=None): + """ + Create a bar plot based on the benchmark results. + + Returns + ------- + fig + Plot. + """ + results = self.get_table() + plt.style.use("ggplot") + plt.rc("font", size=14) + + results = results[3:] + n_datasets = len(results.index) + n_cols = min(4, n_datasets) # Maximum 4 columns + n_rows = math.ceil(n_datasets / n_cols) + fig, axes = plt.subplots(n_rows, n_cols, figsize=(5 * n_cols, 4 * n_rows), sharex=False) + + axes = axes.flatten() if n_datasets > 1 else [axes] + + for i, (dataset, row) in enumerate(results.iterrows()): + ax = axes[i] + + bars = ax.barh(row.index, row.values, alpha=0.8, color=get_colors(len(row.values)), label=row.index) + + ax.set_title(dataset, fontsize=12) + ax.set_xlabel("metric", fontsize=12) + ax.bar_label(bars, fmt="%.4f", padding=2, label_type="center", fontsize=14) + + if i % n_cols != 0: + ax.set_yticklabels([]) + else: + ax.set_ylabel("Model", fontsize=14) + + for j in range(i + 1, len(axes)): + fig.delaxes(axes[j]) + + plt.tight_layout() + fig.legend(results.columns, title="Mitigator", loc="center left", bbox_to_anchor=(1, 0.9), ncols=1, fontsize=14) + if output_path: + plt.savefig(output_path, bbox_inches="tight") + plt.close() + return fig + + def get_radar(self, figsize=(10, 5), output_path=None): + """ + Create a radar plot. + + Parameters + ---------- + figsize : tuple, optional + Size of the figure. The default is (10, 5). + + Returns + ------- + ax + Radar plot. + """ + tab = self.get_table() + tab_trans = tab[3:].T.reset_index(names="Mitigators") + + categories = list(tab_trans.columns)[1:] + N = len(categories) + angles = [n / float(N) * 2 * math.pi for n in range(N)] + angles += angles[:1] + + fig, ax = plt.subplots(figsize=figsize, subplot_kw={"polar": True}) + ax.set_theta_offset(math.pi / 2) + ax.set_theta_direction(-1) + + plt.xticks(angles[:-1], categories) + ax.set_rlabel_position(0) + + min_value = tab[3:].values.min() + max_value = tab[3:].values.max() + plt.ylim(min_value, max_value) + + # Add axis lines + ax.set_rlabel_position(0) + plt.yticks( + np.linspace(min_value, max_value, 5), + [f"{x:.4f}" for x in np.linspace(min_value, max_value, 5)], + color="grey", + size=7, + ) + plt.ylim(min_value, max_value) + + # Add subtle grid lines + ax.grid(True, linestyle="--", alpha=0.7) + + n_var = len(tab_trans) + colors = get_colors(n_var) + for i, color in enumerate(colors): + values = tab_trans.loc[i].drop("Mitigators").values.flatten().tolist() + values += values[:1] + ax.plot(angles, values, linewidth=2, linestyle="solid", label=tab_trans.loc[i]["Mitigators"], color=color) + ax.fill(angles, values, alpha=0.1, color=color) + + plt.legend(loc="center left", bbox_to_anchor=(1.3, 0.5)) + plt.tight_layout() + if output_path: + plt.savefig(output_path, bbox_inches="tight") + return ax + + def run(self, custom_mitigator: object = None, custom_dataset=None): + """ + Run the benchmark for the given task type and stage. + + Parameters + ---------- + custom_mitigator : object, optional + Custom mitigator to run the benchmark with. The default is None. + custom_dataset : object, optional + Custom dataset to run the benchmark with. The default is None. + + Returns + ------- + pd.DataFrame + Results of the benchmark. + """ + dataset = custom_dataset if custom_dataset is not None else self.get_datasets() + mitigator = custom_mitigator if custom_mitigator is not None else self.get_mitigators() + + is_custom_dataset = custom_dataset is not None + is_custom_mitigator = custom_mitigator is not None + + result = self._build_benchmark( + dataset, mitigator, custom_dataset=is_custom_dataset, custom_mitigator=is_custom_mitigator + ) + + if is_custom_mitigator: + bench = ( + self.get_table() + if not is_custom_dataset + else self._build_benchmark(dataset, self.get_mitigators(), custom_dataset=True, custom_mitigator=False) + ) + result = pd.concat([bench, result], axis=1) + result = result.sort_values(by="Mean Score", axis=1, ascending=False) + + return result + + def submit(self): + import webbrowser + + return webbrowser.open("https://forms.office.com/e/8nLjA7Y38w") + + def _retry_with_backoff(self, func, max_attempts=5, initial_wait=1, backoff_factor=2): + def wrapper(*args, **kwargs): + attempts = 0 + wait_time = initial_wait + while attempts < max_attempts: # noqa: RET503 + try: + return func(*args, **kwargs) + except Exception as e: + attempts += 1 + if attempts == max_attempts: + raise + print(f"Attempt {attempts} failed: {e}. Retrying in {wait_time} seconds...") # noqa: T201 + time.sleep(wait_time) + wait_time *= backoff_factor + + return wrapper + + def _build_benchmark(self, datasets, mitigators, custom_mitigator=None, custom_dataset=None, **kwargs): + """ + Run the benchmark for the given task type and stage. + + Parameters + ---------- + datasets : object + Datasets to run the benchmark with. + mitigators : object + Mitigators to run the benchmark with. + custom_mitigator : object, optional + Custom mitigator to run the benchmark with. The default is None. + custom_dataset : object, optional + Custom dataset to run the benchmark with. The default is None. + + Returns + ------- + pd.DataFrame + Results of the benchmark. + """ + results = pd.DataFrame({"dataset": [], "mitigator": [], "metric": [], "time": [], "value": []}) + + if not custom_dataset: + for dataset_name in datasets: + load_dataset = partial(_load_dataset_benchmark, dataset_name) + load_with_retry = self._retry_with_backoff(load_dataset) + try: + data = load_with_retry() + except Exception as e: # noqa: BLE001 + print(f"Failed to load dataset {dataset_name} after 10 attempts: {e}") # noqa: T201 + continue # Skip to the next dataset if all attempts fail + data_split = data.train_test_split(test_size=0.2, random_state=42) + train = data_split["train"] + test = data_split["test"] + + if custom_mitigator: + result_mitigator = self.run_mitigator(mitigators, train, test, dataset_name, **kwargs) + results = pd.concat([results, result_mitigator]) + else: + for mitigator in mitigators[self.stage]: + result_mitigator = self.run_mitigator(mitigator, train, test, dataset_name, **kwargs) + results = pd.concat([results, result_mitigator]) + else: + data = datasets + dataset_name = "Custom Dataset" + data_split = data.train_test_split(test_size=0.2, random_state=42) + train = data_split["train"] + test = data_split["test"] + if custom_mitigator: + result_mitigator = self.run_mitigator(mitigators, train, test, dataset_name, **kwargs) + results = pd.concat([results, result_mitigator]) + else: + for mitigator in mitigators[self.stage]: + result_mitigator = self.run_mitigator(mitigator, train, test, dataset_name, **kwargs) + results = pd.concat([results, result_mitigator]) + + df_time = results.pivot(index="mitigator", columns=["dataset"], values=["time"]) + df_result = results.pivot(index="mitigator", columns=["dataset"], values=["value"]) + df_result.columns = [f"{col[1]}" for col in df_result.columns] + df_result.insert(0, "Mean Score", df_result.mean(axis=1)) + df_result.insert(1, "Std Score", df_result.std(axis=1)) + df_result.insert(2, "Mean Time", df_time.mean(axis=1)) + df_result = df_result.sort_values(by="Mean Score", ascending=False) + return df_result.T + + def _create_pipeline(self, mitigator: Any, **kwargs) -> Pipeline: + steps = [("scalar", StandardScaler())] + + if self.stage == "preprocessing": + steps.extend([("bm_preprocessing", mitigator), ("estimator", self.MODELS[self.task_type])]) + elif self.stage == "postprocessing": + steps.extend([("estimator", self.MODELS[self.task_type]), ("bm_postprocessing", mitigator)]) + elif self.stage == "inprocessing": + in_mitigator = self._configure_inprocessing_mitigator(mitigator, **kwargs) + steps.append(("bm_inprocessing", in_mitigator)) + + return Pipeline(steps=steps) + + def _configure_inprocessing_mitigator(self, mitigator: Any, **kwargs) -> Any: + mitigator_name = mitigator.__name__ + config = { + "AdversarialDebiasing": lambda: mitigator( + features_dim=kwargs["train_shape"][1], + batch_size=512, + hidden_size=64, + adversary_loss_weight=3, + verbose=1, + use_debias=True, + seed=42, + ).transform_estimator(), + "GridSearchReduction": lambda: mitigator(**self._configure_grid_search_reduction()).transform_estimator( + self.MODELS[self.task_type] + ), + "PrejudiceRemover": lambda: mitigator( + maxiter=100, fit_intercept=True, verbose=1, print_interval=1 + ).transform_estimator(self.MODELS[self.task_type]), + "ExponentiatedGradientReduction": lambda: mitigator( + constraints="BoundedGroupLoss", loss="Square", min_val=-0.1, max_val=1.3, eps=0.01 + ).transform_estimator(self.MODELS[self.task_type]), + "FairKCenterClustering": lambda: mitigator(req_nr_per_group=(1, 1), nr_initially_given=0, seed=42), + } + try: + return config.get(mitigator_name, lambda: mitigator(**kwargs))() + except TypeError as e: + logger.exception(f"Error configuring {mitigator_name}: {e!s}") # noqa: TRY401 + raise + + def _configure_grid_search_reduction(self) -> Any: + if self.task_type == "binary_classification": + args = {"constraints": "EqualizedOdds", "grid_size": 20} + return args + if self.task_type == "regression": + args = { + "constraints": "BoundedGroupLoss", + "loss": "Square", + "min_val": -0.1, + "max_val": 0.1, + "grid_size": 50, + } + return args + return None + + def _predict_and_evaluate(self, pipeline: Pipeline, train: dict[str, Any], test: dict[str, Any]) -> tuple[Any, Any]: + if self.task_type == "clustering": + return self._handle_clustering(pipeline, train) + if self.task_type == "multiclass": + y_pred = pipeline.predict(test["X"], bm__group_a=test["group_a"], bm__group_b=test["group_b"]) + metric = self.METRICS[self.task_type](test["group_a"], y_pred, test["y"]) + else: + y_pred = pipeline.predict(test["X"], bm__group_a=test["group_a"], bm__group_b=test["group_b"]) + metric = self.METRICS[self.task_type](test["group_a"], test["group_b"], y_pred, test["y"]) + return y_pred, metric + + def _handle_clustering(self, pipeline: Pipeline, train: dict[str, Any]) -> tuple[Any, Any]: + if self.stage == "postprocessing": + y_pred = pipeline.predict( + train["X"], bm__group_a=train["group_a"], bm__group_b=train["group_b"], bm__centroids="cluster_centers_" + ) + centroids = pipeline["estimator"].cluster_centers_ + else: + y_pred = pipeline.predict(train["X"], bm__group_a=train["group_a"], bm__group_b=train["group_b"]) + centroids = self._get_centroids(pipeline, train) + + metric = self.METRICS[self.task_type](train["group_a"], train["group_b"], train["X"], centroids) + return y_pred, metric + + def _get_centroids(self, pipeline: Pipeline, train: dict[str, Any]) -> Any: + if self.stage == "inprocessing": + return pipeline["bm_inprocessing"].all_centroids + if self.stage == "preprocessing": + model = self.MODELS[self.task_type] + model.fit(train["X"]) + return model.cluster_centers_ + return None + + def run_mitigator( + self, mitigator: Any, train: dict[str, Any], test: dict[str, Any], dataset_name: str, **kwargs + ) -> pd.DataFrame: + """ + Run the benchmark for the given mitigator. + + Parameters + ---------- + mitigator : Any + Mitigator to run the benchmark with. + train : dict[str, Any] + Training data. + test : dict[str, Any] + Testing data. + dataset_name : str + Name of the dataset. + + Returns + ------- + pd.DataFrame + Results of the benchmark. + """ + mitigator_name = mitigator.__class__.__name__ if self.stage != "inprocessing" else mitigator.__name__ + start_time = time.time() + + try: + pipeline = self._create_pipeline(mitigator, train_shape=train["X"].shape, **kwargs) + pipeline.fit(train["X"], train["y"], bm__group_a=train["group_a"], bm__group_b=train["group_b"]) + + _, metric = self._predict_and_evaluate(pipeline, train, test) + + time_pipeline = time.time() - start_time + + return pd.DataFrame( + {"dataset": [dataset_name], "mitigator": [mitigator_name], "time": [time_pipeline], "value": [metric]} + ) + except Exception as e: + logger.exception(f"Error running mitigator {mitigator_name}: {e!s}") # noqa: TRY401 + return pd.DataFrame( + { + "dataset": [dataset_name], + "mitigator": [mitigator_name], + "time": [time.time() - start_time], + "value": [None], + "error": [str(e)], + } + ) diff --git a/src/holisticai/bias/metrics/__init__.py b/src/holisticai/bias/metrics/__init__.py index d99b9da1..197bb13e 100644 --- a/src/holisticai/bias/metrics/__init__.py +++ b/src/holisticai/bias/metrics/__init__.py @@ -10,6 +10,7 @@ abroca, accuracy_diff, average_odds_diff, + balanced_fairness_score, classification_bias_metrics, coefficient_of_variation, cohen_d, @@ -39,6 +40,7 @@ ) from holisticai.bias.metrics._multiclass import ( accuracy_matrix, + balanced_fairness_score_multiclass, confusion_matrix, confusion_tensor, frequency_matrix, @@ -70,6 +72,7 @@ from holisticai.bias.metrics._regression import ( avg_score_diff, avg_score_ratio, + balanced_fairness_score_error, correlation_diff, disparate_impact_regression, jain_index, @@ -87,6 +90,7 @@ # All bias functions and classes __all__ = [ "statistical_parity", + "balanced_fairness_score", "disparate_impact", "four_fifths", "cohen_d", @@ -95,6 +99,7 @@ "true_negative_rate_diff", "abroca", "statistical_parity_regression", + "balanced_fairness_score_error", "disparate_impact_regression", "success_rate_regression", "success_rate", @@ -143,6 +148,7 @@ "multiclass_average_odds", "multiclass_true_rates", "multiclass_statistical_parity", + "balanced_fairness_score_multiclass", "multiclass_bias_metrics", "z_test_diff", "z_test_ratio", diff --git a/src/holisticai/bias/metrics/_classification.py b/src/holisticai/bias/metrics/_classification.py index e79728c2..458bbdcb 100644 --- a/src/holisticai/bias/metrics/_classification.py +++ b/src/holisticai/bias/metrics/_classification.py @@ -39,6 +39,62 @@ def _group_success_rate(g, y): return y[g == 1].sum() / g.sum() # success rate group_a +def balanced_fairness_score(group_a, group_b, y_pred, y_true, alpha=0.5): + """Balanced Fairness Score (BFS). + + This function computes a metric that balances statistical parity and accuracy. + + Interpretation + -------------- + A higher value indicates better balance between fairness and accuracy. + The score ranges from 0 to 1, where 1 is the best possible score. + + Parameters + ---------- + group_a : array-like + Group membership vector (binary) + group_b : array-like + Group membership vector (binary) + y_pred : array-like + Predictions vector (binary) + y_true : array-like + True labels vector (binary) + alpha : float, optional + Weight for balancing fairness and accuracy (default is 0.5) + + Returns + ------- + float + Balanced Fairness Score + + Notes + ----- + BFS = (1 - alpha) * accuracy + alpha * (1 - abs(statistical_parity)) + + Where: + - accuracy is the overall prediction accuracy + - statistical_parity is as defined in the original function + - alpha is a parameter between 0 and 1 that determines the weight of fairness vs. accuracy + + Examples + -------- + >>> import numpy as np + >>> group_a = np.array([1, 1, 1, 1, 0, 0, 0, 0, 0, 0]) + >>> group_b = np.array([0, 0, 0, 0, 1, 1, 1, 1, 1, 1]) + >>> y_pred = np.array([1, 1, 1, 0, 1, 1, 0, 0, 0, 0]) + >>> y_true = np.array([1, 1, 1, 1, 1, 1, 0, 0, 0, 0]) + >>> balanced_fairness_score(group_a, group_b, y_pred, y_true) + 0.7916666666666666 + """ + group_a, group_b, y_pred, y_true = _classification_checks(group_a, group_b, y_pred, y_true) + + sp = statistical_parity(group_a, group_b, y_pred) + acc = accuracy_score(y_true, y_pred) + bfs = (1 - alpha) * acc + alpha * (1 - abs(sp)) + + return bfs + + def statistical_parity(group_a, group_b, y_pred): """Statistical parity. diff --git a/src/holisticai/bias/metrics/_multiclass.py b/src/holisticai/bias/metrics/_multiclass.py index 487bd723..65dc7488 100644 --- a/src/holisticai/bias/metrics/_multiclass.py +++ b/src/holisticai/bias/metrics/_multiclass.py @@ -4,6 +4,7 @@ # utils from holisticai.utils._validation import _multiclass_checks +from sklearn.metrics import accuracy_score def confusion_matrix(y_pred, y_true, classes=None, normalize=None): @@ -674,6 +675,68 @@ def multiclass_true_rates(p_attr, y_pred, y_true, groups=None, classes=None, agg return np.max(dist_mat) if aggregation_fun == "max" else np.sum(dist_mat) / (num_groups * (num_groups - 1) / 2) +def balanced_fairness_score_multiclass(p_attr, y_pred, y_true, groups=None, classes=None, alpha=0.5): + """Balanced Fairness Score Multiclass (BFSM). + + This function computes a metric that balances statistical parity and accuracy. + + Interpretation + -------------- + A higher value indicates better balance between fairness and accuracy. + The score ranges from 0 to 1, where 1 is the best possible score. + + Parameters + ---------- + group_a : array-like + Group membership vector (binary) + group_b : array-like + Group membership vector (binary) + y_pred : array-like + Predictions vector (binary) + y_true : array-like + True labels vector (binary) + alpha : float, optional + Weight for balancing fairness and accuracy (default is 0.5) + + Returns + ------- + float + Balanced Fairness Score + + Notes + ----- + BFS = (1 - alpha) * accuracy + alpha * (1 - abs(statistical_parity)) + + Where: + - accuracy is the overall prediction accuracy + - statistical_parity is as defined in the original function + - alpha is a parameter between 0 and 1 that determines the weight of fairness vs. accuracy + + Examples + -------- + >>> import numpy as np + >>> group_a = np.array([1, 1, 1, 1, 0, 0, 0, 0, 0, 0]) + >>> group_b = np.array([0, 0, 0, 0, 1, 1, 1, 1, 1, 1]) + >>> y_pred = np.array([1, 1, 1, 0, 1, 1, 0, 0, 0, 0]) + >>> y_true = np.array([1, 1, 1, 1, 1, 1, 0, 0, 0, 0]) + >>> balanced_fairness_score(group_a, group_b, y_pred, y_true) + 0.7916666666666666 + """ + p_attr, y_pred, y_true, groups, classes = _multiclass_checks( + p_attr=p_attr, + y_pred=y_pred, + y_true=y_true, + groups=groups, + classes=classes, + ) + + sp = multiclass_statistical_parity(p_attr, y_pred) + acc = accuracy_score(y_true, y_pred) + bfsm = (1 - alpha) * acc + alpha * (1 - abs(sp)) + + return bfsm + + def multiclass_statistical_parity(p_attr, y_pred, groups=None, classes=None, aggregation_fun="mean"): """Multiclass statistical parity diff --git a/src/holisticai/bias/metrics/_regression.py b/src/holisticai/bias/metrics/_regression.py index ed1d6b1b..39afe200 100644 --- a/src/holisticai/bias/metrics/_regression.py +++ b/src/holisticai/bias/metrics/_regression.py @@ -7,6 +7,7 @@ # utils from holisticai.utils._formatting import slice_arrays_by_quantile from holisticai.utils._validation import _check_non_empty, _regression_checks +from sklearn.metrics import mean_absolute_percentage_error def _calc_success_rate(group_membership: np.array, threshold=float): @@ -123,6 +124,68 @@ def disparate_impact_regression(group_a, group_b, y_pred, q=0.8): return np.squeeze(disp_impact)[()] +def balanced_fairness_score_error(group_a, group_b, y_pred, y_true, alpha=0.5): + """Balanced Fairness Score Error (BFSE) for Regression Tasks. + + This function computes a metric that balances statistical parity and prediction error + for regression tasks. + + Interpretation + -------------- + A higher value indicates better balance between fairness and accuracy. + The score ranges from 0 to 1, where 1 is the best possible score. + + Parameters + ---------- + group_a : array-like + Group membership vector (binary) + group_b : array-like + Group membership vector (binary) + y_pred : array-like + Predicted values (continuous) + y_true : array-like + True values (continuous) + alpha : float, optional + Weight for balancing fairness and accuracy (default is 0.5) + + Returns + ------- + float + Balanced Fairness Score Error + + Notes + ----- + BFSE = (1 - alpha) * MAPE + alpha * abs(statistical_parity) + + Where: + - MAPE is the Mean Absolute Percentage Error + - statistical_parity is adapted for regression tasks + - alpha is a parameter between 0 and 1 that determines the weight of fairness vs. accuracy + + Examples + -------- + >>> import numpy as np + >>> group_a = np.array([1, 1, 1, 1, 0, 0, 0, 0, 0, 0]) + >>> group_b = np.array([0, 0, 0, 0, 1, 1, 1, 1, 1, 1]) + >>> y_pred = np.array([2.5, 3.2, 2.8, 3.0, 1.8, 2.2, 1.5, 1.7, 2.0, 1.9]) + >>> y_true = np.array([2.3, 3.0, 2.9, 3.1, 2.0, 2.1, 1.6, 1.8, 1.9, 2.0]) + >>> balanced_fairness_score_error(group_a, group_b, y_pred, y_true) + 0.0875 + """ + group_a, group_b, y_pred, y_true, q = _regression_checks(group_a, group_b, y_pred, y_true) + + sp = statistical_parity_regression(group_a, group_b, y_pred) + mape = mean_absolute_percentage_error(y_true, y_pred) + + max_sp = np.max(y_pred) - np.min(y_pred) + normalized_sp = abs(sp) / max_sp if max_sp != 0 else 0 + normalized_mape = 2 / (1 + np.exp(-mape)) - 1 # sigmoid + + bfs = (1 - alpha) * normalized_mape + alpha * normalized_sp + + return 1 - bfs + + def statistical_parity_regression(group_a, group_b, y_pred, q=0.5): """Statistical Parity quantile (Regression version) diff --git a/src/holisticai/datasets/_dataloaders.py b/src/holisticai/datasets/_dataloaders.py index f2c9b063..44e7ef1b 100644 --- a/src/holisticai/datasets/_dataloaders.py +++ b/src/holisticai/datasets/_dataloaders.py @@ -44,6 +44,9 @@ def load_hai_datasets(dataset_name, data_home=None): - clinical_records - acsincome - acspublic + - acsemployment + - acsmobility + - acstraveltime - compass_two_year_recid - compass_is_recid - german_credit diff --git a/src/holisticai/datasets/_load_dataset.py b/src/holisticai/datasets/_load_dataset.py index aa0cb1f8..01979fab 100644 --- a/src/holisticai/datasets/_load_dataset.py +++ b/src/holisticai/datasets/_load_dataset.py @@ -623,6 +623,8 @@ def load_compas_two_year_recid_dataset( protected_attributes = ["race", "sex", "age"] output_column = "two_year_recid" + data["race"] = [1 if x == "Caucasian" else 0 for x in data["race"]] + df = data.copy() remove_columns = [*protected_attributes, output_column] df.reset_index(drop=True, inplace=True) @@ -644,7 +646,7 @@ def load_compas_two_year_recid_dataset( ga_label = "Causasian" gb_label = "Non-Caucasian" group_a = pd.Series(df["race"] == 1, name="group_a") - group_b = pd.Series(df["race"] == 2, name="group_b") + group_b = pd.Series(df["race"] == 0, name="group_b") else: raise ValueError( f"The protected attribute doesn't exist or not implemented. Please use: {protected_attribute}" @@ -677,6 +679,9 @@ def load_compas_is_recid_dataset(preprocessed=True, protected_attribute: Optiona data = load_hai_datasets(dataset_name="compas_is_recid") protected_attributes = ["race", "sex", "age"] output_column = "is_recid" + + data["race"] = [1 if x == "Caucasian" else 0 for x in data["race"]] + df = data.copy() remove_columns = [*protected_attributes, output_column] df.reset_index(drop=True, inplace=True) @@ -698,7 +703,7 @@ def load_compas_is_recid_dataset(preprocessed=True, protected_attribute: Optiona ga_label = "Causasian" gb_label = "Non-Caucasian" group_a = pd.Series(df["race"] == 1, name="group_a") - group_b = pd.Series(df["race"] == 2, name="group_b") + group_b = pd.Series(df["race"] == 0, name="group_b") else: raise ValueError( f"The protected attribute doesn't exist or not implemented. Please use: {protected_attribute}" @@ -780,13 +785,17 @@ def load_acsincome_dataset(preprocessed=True, protected_attribute: Optional[Lite If this parameter is set, the dataset will be returned with the protected attribute as a binary column group_a and group_b. Otherwise, the dataset will be returned with the protected attribute as a column p_attrs. + References + ---------- + [1] Ding, F., Hardt, M., Miller, J., & Schmidt, L. (2021). Retiring adult: New datasets for fair machine learning. Advances in neural information processing systems, 34, 6478-6490. + Returns ------- tuple A tuple with two lists containing the data, output variable, protected group A and protected group B """ data = load_hai_datasets(dataset_name="acsincome") - protected_attributes = ["AGEP", "RAC1P", "SEX"] + protected_attributes = ["RAC1P", "SEX"] output_column = "PINCP" # Total person's income data[output_column] = (data[output_column] > 50000).astype(int) # map income to binary @@ -810,9 +819,9 @@ def load_acsincome_dataset(preprocessed=True, protected_attribute: Optional[Lite group_b = pd.Series(df["SEX"] == 2, name="group_b") elif protected_attribute == "race": ga_label = "White" - gb_label = "Non_White" - group_a = pd.Series(df["RAC1P"] == "White", name="group_a") - group_b = pd.Series(df["RAC1P"] == "Non-White", name="group_b") + gb_label = "Non-White" + group_a = pd.Series(df["RAC1P"] == ga_label, name="group_a") + group_b = pd.Series(df["RAC1P"] == gb_label, name="group_b") else: raise ValueError( f"The protected attribute doesn't exist or not implemented. Please use: {protected_attribute}" @@ -836,13 +845,17 @@ def load_acspublic_dataset(preprocessed=True, protected_attribute: Optional[Lite If this parameter is set, the dataset will be returned with the protected attribute as a binary column group_a and group_b. Otherwise, the dataset will be returned with the protected attribute as a column p_attrs. + References + ---------- + [1] Ding, F., Hardt, M., Miller, J., & Schmidt, L. (2021). Retiring adult: New datasets for fair machine learning. Advances in neural information processing systems, 34, 6478-6490. + Returns ------- tuple A tuple with two lists containing the data, output variable, protected group A and protected group B """ data = load_hai_datasets(dataset_name="acspublic") - protected_attributes = ["AGEP", "RAC1P", "SEX"] + protected_attributes = ["RAC1P", "SEX"] output_column = "PUBCOV" # public health coverage : an individual's label is 1 if PUBCOV == 1 (with public health coverage), otherwise 0. df = data.copy() @@ -865,9 +878,77 @@ def load_acspublic_dataset(preprocessed=True, protected_attribute: Optional[Lite group_b = pd.Series(df["SEX"] == 2, name="group_b") elif protected_attribute == "race": ga_label = "White" - gb_label = "Non_White" - group_a = pd.Series(df["RAC1P"] == "White", name="group_a") - group_b = pd.Series(df["RAC1P"] == "Non-White", name="group_b") + gb_label = "Non-White" + group_a = pd.Series(df["RAC1P"] == ga_label, name="group_a") + group_b = pd.Series(df["RAC1P"] == gb_label, name="group_b") + else: + raise ValueError( + f"The protected attribute doesn't exist or not implemented. Please use: {protected_attribute}" + ) + + if protected_attribute is not None: + metadata = f"""{protected_attribute}: {{'group_a': '{ga_label}', 'group_b': '{gb_label}'}}""" + return Dataset(X=X, y=y, p_attrs=p_attrs, group_a=group_a, group_b=group_b, _metadata=metadata) + return Dataset(X=X, y=y, p_attrs=p_attrs) + +def load_mobility_dataset(preprocessed=True, protected_attribute: Optional[Literal["race", "sex", "dis"]] = "sex"): + """ + Processes the ACSMobility dataset and returns the data, output variable, protected group A and protected group B as numerical arrays or as dataframe if needed + + Parameters + ---------- + preprocessed : bool + Whether to return the preprocessed X and y. + protected_attribute : str + If this parameter is set, the dataset will be returned with the protected attribute as a binary column group_a and group_b. + Otherwise, the dataset will be returned with the protected attribute as a column p_attrs. + + References + ---------- + [1] Ding, F., Hardt, M., Miller, J., & Schmidt, L. (2021). Retiring adult: New datasets for fair machine learning. Advances in neural information processing systems, 34, 6478-6490. + + Returns + ------- + tuple + A tuple with two lists containing the data, output variable, protected group A and protected group B + """ + data = load_hai_datasets(dataset_name="acsmobility") + protected_attributes = ["RAC1P", "SEX", "DIS"] + output_column = "MIG" # an individual’s label is 1 if MIG == 1, and 0 otherwise. + + df = data.copy() + df["RAC1P"] = df["RAC1P"].astype(float) + df["SEX"] = df["SEX"].astype(float) + df["DIS"] = df["DIS"].astype(float) + + df["RAC1P"] = ["White" if x == 1 else "Non-White" for x in df["RAC1P"]] + df["DIS"] = ["With disability" if x == 1 else "Without disability" for x in df["DIS"]] + remove_columns = [*protected_attributes, output_column] + df.reset_index(drop=True, inplace=True) + X = df.drop(columns=remove_columns) + + if preprocessed: + X = pd.get_dummies(X, columns=X.select_dtypes(include=["category", "object"]).columns, dtype=float) + + p_attrs = df[protected_attributes] + y = df[output_column] + + if protected_attribute is not None: + if protected_attribute == "sex": + ga_label = "Male" + gb_label = "Female" + group_a = pd.Series(df["SEX"] == 1, name="group_a") + group_b = pd.Series(df["SEX"] == 2, name="group_b") + elif protected_attribute == "race": + ga_label = "White" + gb_label = "Non-White" + group_a = pd.Series(df["RAC1P"] == ga_label, name="group_a") + group_b = pd.Series(df["RAC1P"] == gb_label, name="group_b") + elif protected_attribute == "dis": + ga_label = "With disability" + gb_label = "Without disability" + group_a = pd.Series(df["DIS"] == ga_label, name="group_a") + group_b = pd.Series(df["DIS"] == gb_label, name="group_b") else: raise ValueError( f"The protected attribute doesn't exist or not implemented. Please use: {protected_attribute}" @@ -879,6 +960,143 @@ def load_acspublic_dataset(preprocessed=True, protected_attribute: Optional[Lite return Dataset(X=X, y=y, p_attrs=p_attrs) +def load_employment_dataset(preprocessed=True, protected_attribute: Optional[Literal["race", "sex", "dis"]] = "sex"): + """ + Processes the ACSEmployment dataset and returns the data, output variable, protected group A and protected group B as numerical arrays or as dataframe if needed + + Parameters + ---------- + preprocessed : bool + Whether to return the preprocessed X and y. + protected_attribute : str + If this parameter is set, the dataset will be returned with the protected attribute as a binary column group_a and group_b. + Otherwise, the dataset will be returned with the protected attribute as a column p_attrs. + + References + ---------- + [1] Ding, F., Hardt, M., Miller, J., & Schmidt, L. (2021). Retiring adult: New datasets for fair machine learning. Advances in neural information processing systems, 34, 6478-6490. + + Returns + ------- + tuple + A tuple with two lists containing the data, output variable, protected group A and protected group B + """ + data = load_hai_datasets(dataset_name="acsemployment") + protected_attributes = ["RAC1P", "SEX", "DIS"] + output_column = "ESR" # an individual’s label is 1 if ESR == 1, and 0 otherwise. + + df = data.copy() + df["RAC1P"] = df["RAC1P"].astype(float) + df["SEX"] = df["SEX"].astype(float) + df["DIS"] = df["DIS"].astype(float) + + df["RAC1P"] = ["White" if x == 1 else "Non-White" for x in df["RAC1P"]] + df["DIS"] = ["With disability" if x == 1 else "Without disability" for x in df["DIS"]] + remove_columns = [*protected_attributes, output_column] + df.reset_index(drop=True, inplace=True) + X = df.drop(columns=remove_columns) + + if preprocessed: + X = pd.get_dummies(X, columns=X.select_dtypes(include=["category", "object"]).columns, dtype=float) + + p_attrs = df[protected_attributes] + y = df[output_column] + + if protected_attribute is not None: + if protected_attribute == "sex": + ga_label = "Male" + gb_label = "Female" + group_a = pd.Series(df["SEX"] == 1, name="group_a") + group_b = pd.Series(df["SEX"] == 2, name="group_b") + elif protected_attribute == "race": + ga_label = "White" + gb_label = "Non-White" + group_a = pd.Series(df["RAC1P"] == ga_label, name="group_a") + group_b = pd.Series(df["RAC1P"] == gb_label, name="group_b") + elif protected_attribute == "dis": + ga_label = "With disability" + gb_label = "Without disability" + group_a = pd.Series(df["DIS"] == ga_label, name="group_a") + group_b = pd.Series(df["DIS"] == gb_label, name="group_b") + else: + raise ValueError( + f"The protected attribute doesn't exist or not implemented. Please use: {protected_attribute}" + ) + + if protected_attribute is not None: + metadata = f"""{protected_attribute}: {{'group_a': '{ga_label}', 'group_b': '{gb_label}'}}""" + return Dataset(X=X, y=y, p_attrs=p_attrs, group_a=group_a, group_b=group_b, _metadata=metadata) + return Dataset(X=X, y=y, p_attrs=p_attrs) + + +def load_traveltime_dataset(preprocessed=True, protected_attribute: Optional[Literal["race", "sex", "dis"]] = "sex"): + """ + Processes the ACSTravelTime dataset and returns the data, output variable, protected group A and protected group B as numerical arrays or as dataframe if needed + + Parameters + ---------- + preprocessed : bool + Whether to return the preprocessed X and y. + protected_attribute : str + If this parameter is set, the dataset will be returned with the protected attribute as a binary column group_a and group_b. + Otherwise, the dataset will be returned with the protected attribute as a column p_attrs. + + References + ---------- + [1] Ding, F., Hardt, M., Miller, J., & Schmidt, L. (2021). Retiring adult: New datasets for fair machine learning. Advances in neural information processing systems, 34, 6478-6490. + + Returns + ------- + tuple + A tuple with two lists containing the data, output variable, protected group A and protected group B + """ + data = load_hai_datasets(dataset_name="acstraveltime") + protected_attributes = ["RAC1P", "SEX", "DIS"] + output_column = "JWMNP" # an individual’s label is 1 if JWMNP > 20, and 0 otherwise. + + df = data.copy() + df["RAC1P"] = df["RAC1P"].astype(float) + df["SEX"] = df["SEX"].astype(float) + df["DIS"] = df["DIS"].astype(float) + + df["RAC1P"] = ["White" if x == 1 else "Non-White" for x in df["RAC1P"]] + df["DIS"] = ["With disability" if x == 1 else "Without disability" for x in df["DIS"]] + remove_columns = [*protected_attributes, output_column] + df.reset_index(drop=True, inplace=True) + X = df.drop(columns=remove_columns) + + if preprocessed: + X = pd.get_dummies(X, columns=X.select_dtypes(include=["category", "object"]).columns, dtype=float) + + p_attrs = df[protected_attributes] + y = df[output_column] + + if protected_attribute is not None: + if protected_attribute == "sex": + ga_label = "Male" + gb_label = "Female" + group_a = pd.Series(df["SEX"] == 1, name="group_a") + group_b = pd.Series(df["SEX"] == 2, name="group_b") + elif protected_attribute == "race": + ga_label = "White" + gb_label = "Non-White" + group_a = pd.Series(df["RAC1P"] == ga_label, name="group_a") + group_b = pd.Series(df["RAC1P"] == gb_label, name="group_b") + elif protected_attribute == "dis": + ga_label = "With disability" + gb_label = "Without disability" + group_a = pd.Series(df["DIS"] == ga_label, name="group_a") + group_b = pd.Series(df["DIS"] == gb_label, name="group_b") + else: + raise ValueError( + f"The protected attribute doesn't exist or not implemented. Please use: {protected_attribute}" + ) + + if protected_attribute is not None: + metadata = f"""{protected_attribute}: {{'group_a': '{ga_label}', 'group_b': '{gb_label}'}}""" + return Dataset(X=X, y=y, p_attrs=p_attrs, group_a=group_a, group_b=group_b, _metadata=metadata) + return Dataset(X=X, y=y, p_attrs=p_attrs) + def load_mw_medium_dataset(preprocessed=True, protected_attribute: Optional[Literal["race", "sex"]] = "race"): """ Processes the Minimum Wage Medium dataset and returns the data, output variable, protected group A and protected group B as numerical arrays or as dataframe if needed @@ -1008,6 +1226,9 @@ def load_mw_small_dataset(preprocessed=True, protected_attribute: Optional[Liter "diabetes", "acsincome", "acspublic", + "acsemployment", + "acsmobility", + "acstraveltime", "mw_medium", "mw_small", ] @@ -1073,6 +1294,12 @@ def load_dataset( return load_acsincome_dataset(preprocessed=preprocessed, protected_attribute=protected_attribute) if dataset_name == "acspublic": return load_acspublic_dataset(preprocessed=preprocessed, protected_attribute=protected_attribute) + if dataset_name == "acsemployment": + return load_employment_dataset(preprocessed=preprocessed, protected_attribute=protected_attribute) + if dataset_name == "acsmobility": + return load_mobility_dataset(preprocessed=preprocessed, protected_attribute=protected_attribute) + if dataset_name == "acstraveltime": + return load_traveltime_dataset(preprocessed=preprocessed, protected_attribute=protected_attribute) if dataset_name == "mw_medium": return load_mw_medium_dataset(preprocessed=preprocessed, protected_attribute=protected_attribute) if dataset_name == "mw_small": @@ -1106,6 +1333,7 @@ def _load_dataset_benchmark( NotImplementedError: If the specified dataset name is not supported. """ + if dataset_name == "adult_sex": return load_adult_dataset(preprocessed=preprocessed, protected_attribute="sex") if dataset_name == "adult_race": @@ -1128,7 +1356,6 @@ def _load_dataset_benchmark( if dataset_name == "us_crime_race": return load_us_crime_dataset(preprocessed=preprocessed, protected_attribute="race") - if dataset_name == "us_crime_multiclass_race": return load_us_crime_multiclass_dataset(preprocessed=preprocessed, protected_attribute="race") @@ -1146,33 +1373,49 @@ def _load_dataset_benchmark( if dataset_name == "compas_two_year_recid_sex": return load_compas_two_year_recid_dataset(preprocessed=preprocessed, protected_attribute="sex") - if dataset_name == "compas_two_year_recid_race": return load_compas_two_year_recid_dataset(preprocessed=preprocessed, protected_attribute="race") if dataset_name == "compas_is_recid_sex": return load_compas_is_recid_dataset(preprocessed=preprocessed, protected_attribute="sex") - if dataset_name == "compas_is_recid_race": return load_compas_is_recid_dataset(preprocessed=preprocessed, protected_attribute="race") if dataset_name == "diabetes_sex": return load_diabetes_dataset(preprocessed=preprocessed, protected_attribute="sex") - if dataset_name == "diabetes_race": return load_diabetes_dataset(preprocessed=preprocessed, protected_attribute="race") if dataset_name == "acsincome_sex": return load_acsincome_dataset(preprocessed=preprocessed, protected_attribute="sex") - if dataset_name == "acsincome_race": return load_acsincome_dataset(preprocessed=preprocessed, protected_attribute="race") if dataset_name == "acspublic_sex": return load_acspublic_dataset(preprocessed=preprocessed, protected_attribute="sex") - if dataset_name == "acspublic_race": return load_acspublic_dataset(preprocessed=preprocessed, protected_attribute="race") + + if dataset_name == "acsemployment_race": + return load_employment_dataset(preprocessed=preprocessed, protected_attribute="race") + if dataset_name == "acsemployment_sex": + return load_employment_dataset(preprocessed=preprocessed, protected_attribute="sex") + if dataset_name == "acsemployment_dis": + return load_employment_dataset(preprocessed=preprocessed, protected_attribute="dis") + + if dataset_name == "acsmobility_race": + return load_mobility_dataset(preprocessed=preprocessed, protected_attribute="race") + if dataset_name == "acsmobility_sex": + return load_mobility_dataset(preprocessed=preprocessed, protected_attribute="sex") + if dataset_name == "acsmobility_dis": + return load_mobility_dataset(preprocessed=preprocessed, protected_attribute="dis") + + if dataset_name == "acstraveltime_race": + return load_traveltime_dataset(preprocessed=preprocessed, protected_attribute="race") + if dataset_name == "acstraveltime_sex": + return load_traveltime_dataset(preprocessed=preprocessed, protected_attribute="sex") + if dataset_name == "acstraveltime_dis": + return load_traveltime_dataset(preprocessed=preprocessed, protected_attribute="dis") if dataset_name == "mw_medium_race": return load_mw_medium_dataset(preprocessed=preprocessed, protected_attribute="race") diff --git a/src/holisticai/utils/benchmark_config/__init__.py b/src/holisticai/utils/benchmark_config/__init__.py new file mode 100644 index 00000000..a4f28770 --- /dev/null +++ b/src/holisticai/utils/benchmark_config/__init__.py @@ -0,0 +1,13 @@ +from holisticai.utils.benchmark_config._bias import ( + DATASETS, + METRICS, + MITIGATORS, + MODELS, +) + +__all__ = [ + "DATASETS", + "METRICS", + "MITIGATORS", + "MODELS", +] diff --git a/src/holisticai/utils/benchmark_config/_bias.py b/src/holisticai/utils/benchmark_config/_bias.py new file mode 100644 index 00000000..90e8f226 --- /dev/null +++ b/src/holisticai/utils/benchmark_config/_bias.py @@ -0,0 +1,147 @@ +from sklearn.cluster import KMeans +from sklearn.linear_model import LinearRegression, LogisticRegression + +from holisticai.bias.metrics import ( + balanced_fairness_score, + balanced_fairness_score_error, + balanced_fairness_score_multiclass, + social_fairness_ratio, +) +from holisticai.bias.mitigation import ( + MCMF, + AdversarialDebiasing, + CalibratedEqualizedOdds, + CorrelationRemover, + DisparateImpactRemover, + EqualizedOdds, + ExponentiatedGradientReduction, + FairKCenterClustering, + FairletClusteringPreprocessing, + GridSearchReduction, + LearningFairRepresentation, + LPDebiaserBinary, + LPDebiaserMulticlass, + MLDebiaser, + PluginEstimationAndCalibration, + PrejudiceRemover, + RejectOptionClassification, + Reweighing, + WassersteinBarycenter, +) + +DATASETS = { + "binary_classification": [ + "compas_two_year_recid_sex", + "compas_two_year_recid_race", + "compas_is_recid_sex", + "compas_is_recid_race", + "adult_sex", + "adult_race", + "german_credit_sex", + "clinical_records_sex", + "bank_marketing_marital", + "law_school_sex", + "law_school_race", + "diabetes_sex", + "diabetes_race", + "census_kdd_sex", + "acsincome_sex", + "acsincome_race", + "acspublic_sex", + "acspublic_race", + "acsemployment_race", + "acsemployment_sex", + "acsemployment_disability", + "acsmobility_race", + "acsmobility_sex", + "acsmobility_disability", + "acstraveltime_race", + "acstraveltime_sex", + "acstraveltime_disability", + "mw_small_race", + "mw_small_sex", + "mw_medium_race", + "mw_medium_sex", + ], + "multiclass": [ + "us_crime_multiclass_race", + "student_multiclass_sex", + "student_multiclass_address", + ], + "regression": [ + "us_crime_race", + "student_sex", + "student_address", + ], + "clustering": [ + "clinical_records_sex", + "student_sex", + "german_credit_sex", + "compas_two_year_recid_sex", + "compas_is_recid_sex", + "adult_sex", + ], +} + +METRICS = { + "binary_classification": balanced_fairness_score, + "multiclass": balanced_fairness_score_multiclass, + "regression": balanced_fairness_score_error, + "clustering": social_fairness_ratio, +} + +MODELS = { + "binary_classification": LogisticRegression(random_state=42), + "multiclass": LogisticRegression(multi_class="multinomial", solver="newton-cg"), + "regression": LinearRegression(), + "clustering": KMeans(n_clusters=3, random_state=42), +} + + +MITIGATORS = { + "binary_classification": { + "preprocessing": [CorrelationRemover(), DisparateImpactRemover(), LearningFairRepresentation(), Reweighing()], + "inprocessing": [ + AdversarialDebiasing, + GridSearchReduction, + PrejudiceRemover, + ], + "postprocessing": [ + CalibratedEqualizedOdds(cost_constraint="fnr"), + EqualizedOdds(solver="highs", seed=42), + RejectOptionClassification(metric_name="Statistical parity difference"), + LPDebiaserBinary(), + MLDebiaser(), + ], + }, + "multiclass": { + "preprocessing": [CorrelationRemover(), DisparateImpactRemover(), Reweighing()], + "inprocessing": [], + "postprocessing": [ + LPDebiaserMulticlass(), + MLDebiaser(), + ], + }, + "regression": { + "preprocessing": [CorrelationRemover(), DisparateImpactRemover()], + "inprocessing": [ + ExponentiatedGradientReduction, + GridSearchReduction, + ], + "postprocessing": [ + PluginEstimationAndCalibration(), + WassersteinBarycenter(), + ], + }, + "clustering": { + "preprocessing": [ + FairletClusteringPreprocessing(seed=42), + ], + "inprocessing": [ + FairKCenterClustering, + ], + "postprocessing": [ + MCMF(metric="L1", solver="highs-ipm"), + ], + }, +} diff --git a/tutorials/benchmark/template_benchmark.ipynb b/tutorials/benchmark/template_benchmark.ipynb new file mode 100644 index 00000000..a46eb6ef --- /dev/null +++ b/tutorials/benchmark/template_benchmark.ipynb @@ -0,0 +1,703 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [], + "source": [ + "from holisticai.benchmark import BiasMitigationBenchmark" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Select the benchmark task type and the bias mitigation type" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "benchmark = BiasMitigationBenchmark(\"binary_classification\", \"preprocessing\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Check the available datasets and mitigators" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Available datasets for binary_classification: ['compas_two_year_recid_sex', 'compas_two_year_recid_race', 'compas_is_recid_sex', 'compas_is_recid_race', 'adult_sex', 'adult_race', 'german_credit_sex', 'clinical_records_sex', 'bank_marketing_marital', 'law_school_sex', 'law_school_race', 'diabetes_sex', 'diabetes_race', 'census_kdd_sex', 'acsincome_sex', 'acsincome_race', 'acspublic_sex', 'acspublic_race', 'acsemployment_race', 'acsemployment_sex', 'acsemployment_disability', 'acsmobility_race', 'acsmobility_sex', 'acsmobility_disability', 'acstraveltime_race', 'acstraveltime_sex', 'acstraveltime_disability', 'mw_small_race', 'mw_small_sex', 'mw_medium_race', 'mw_medium_sex']\n", + "Available mitigators for binary_classification with preprocessing: ['CorrelationRemover', 'DisparateImpactRemover', 'LearningFairRepresentation', 'Reweighing']\n" + ] + } + ], + "source": [ + "benchmark.available_settings()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Get the actual benchmark results" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
CorrelationRemoverReweighingDisparateImpactRemoverLearningFairRepresentation
Mean Score0.8865680.8750880.8645560.863185
Std Score0.0580840.0604480.0754340.079973
Mean Time1.6227134.59810733.05408316.696009
acsincome_race0.8763190.8394370.8423870.731448
acsincome_sex0.8666830.8634200.8679430.835478
acspublic_race0.8763790.8764590.8747020.851131
acspublic_sex0.8661630.8606150.8570400.838606
adult_race0.9222020.8784670.8848680.869641
adult_sex0.8774770.8753320.6662320.878220
bank_marketing_marital0.9482960.9436750.9348110.940838
census_kdd_sex0.9500830.9604400.9554570.970146
clinical_records_sex0.8416670.8083330.8416670.791667
compass_race0.7682450.7464270.7490150.706091
compass_sex0.7712250.7799310.7973510.766775
diabetes_race0.9442590.9442590.9442590.944259
diabetes_sex0.9442590.9442590.9442590.944259
german_credit_sex0.8471430.8461710.8441670.852500
law_school_race0.9445480.8965430.8924080.944952
law_school_sex0.9401350.9376460.9363270.944952
\n", + "
" + ], + "text/plain": [ + " CorrelationRemover Reweighing \\\n", + " \n", + "Mean Score 0.886568 0.875088 \n", + "Std Score 0.058084 0.060448 \n", + "Mean Time 1.622713 4.598107 \n", + "acsincome_race 0.876319 0.839437 \n", + "acsincome_sex 0.866683 0.863420 \n", + "acspublic_race 0.876379 0.876459 \n", + "acspublic_sex 0.866163 0.860615 \n", + "adult_race 0.922202 0.878467 \n", + "adult_sex 0.877477 0.875332 \n", + "bank_marketing_marital 0.948296 0.943675 \n", + "census_kdd_sex 0.950083 0.960440 \n", + "clinical_records_sex 0.841667 0.808333 \n", + "compass_race 0.768245 0.746427 \n", + "compass_sex 0.771225 0.779931 \n", + "diabetes_race 0.944259 0.944259 \n", + "diabetes_sex 0.944259 0.944259 \n", + "german_credit_sex 0.847143 0.846171 \n", + "law_school_race 0.944548 0.896543 \n", + "law_school_sex 0.940135 0.937646 \n", + "\n", + " DisparateImpactRemover LearningFairRepresentation \n", + " \n", + "Mean Score 0.864556 0.863185 \n", + "Std Score 0.075434 0.079973 \n", + "Mean Time 33.054083 16.696009 \n", + "acsincome_race 0.842387 0.731448 \n", + "acsincome_sex 0.867943 0.835478 \n", + "acspublic_race 0.874702 0.851131 \n", + "acspublic_sex 0.857040 0.838606 \n", + "adult_race 0.884868 0.869641 \n", + "adult_sex 0.666232 0.878220 \n", + "bank_marketing_marital 0.934811 0.940838 \n", + "census_kdd_sex 0.955457 0.970146 \n", + "clinical_records_sex 0.841667 0.791667 \n", + "compass_race 0.749015 0.706091 \n", + "compass_sex 0.797351 0.766775 \n", + "diabetes_race 0.944259 0.944259 \n", + "diabetes_sex 0.944259 0.944259 \n", + "german_credit_sex 0.844167 0.852500 \n", + "law_school_race 0.892408 0.944952 \n", + "law_school_sex 0.936327 0.944952 " + ] + }, + "execution_count": 4, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "benchmark.get_table()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Visualize the actual benchmark results" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "benchmark.get_plot(output_path=\"output.png\")" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "execution_count": 6, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "benchmark.get_heatmap(fig_size=(20, 5))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Create your own mitigator" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [], + "source": [ + "from __future__ import annotations\n", + "\n", + "import numpy as np\n", + "from typing import Optional\n", + "from holisticai.utils.transformers.bias import BMPreprocessing as BMPre\n", + "\n", + "class MyMitigator(BMPre):\n", + " def __init__(self):\n", + " self.sample_weight = None\n", + "\n", + " def fit(\n", + " self,\n", + " y: np.ndarray,\n", + " group_a: np.ndarray,\n", + " group_b: np.ndarray,\n", + " sample_weight: Optional[np.ndarray] = None,\n", + " ):\n", + " # Simple logic to assign weights based on group membership\n", + " self.sample_weight = np.ones_like(y, dtype=np.float32)\n", + " group_mask = (group_a == 1) & (group_b == 0)\n", + " self.sample_weight[group_mask] = 2.0 # Assign higher weight to this group\n", + " \n", + " return self\n", + "\n", + " def transform(self, X: np.ndarray):\n", + " return X # If needed, apply some transformation to X\n", + "\n", + " def fit_transform(\n", + " self,\n", + " X: np.ndarray,\n", + " y: np.ndarray,\n", + " group_a: np.ndarray,\n", + " group_b: np.ndarray,\n", + " sample_weight: Optional[np.ndarray] = None,\n", + " ):\n", + " return self.fit(y, group_a, group_b, sample_weight).transform(X)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Evaluate your mitigator in the benchmark setup" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "ename": "", + "evalue": "", + "output_type": "error", + "traceback": [ + "\u001b[1;31mThe Kernel crashed while executing code in the current cell or a previous cell. \n", + "\u001b[1;31mPlease review the code in the cell(s) to identify a possible cause of the failure. \n", + "\u001b[1;31mClick here for more info. \n", + "\u001b[1;31mView Jupyter log for further details." + ] + } + ], + "source": [ + "my_mitigator = MyMitigator()\n", + "\n", + "my_results = benchmark.run(custom_mitigator=my_mitigator)\n", + "my_results" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Do you have nice results? Share them with us!" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "True" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "benchmark.submit()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Create and evaluate your own dataset" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "INFO:jax._src.xla_bridge:Unable to initialize backend 'rocm': module 'jaxlib.xla_extension' has no attribute 'GpuAllocatorConfig'\n", + "INFO:jax._src.xla_bridge:Unable to initialize backend 'tpu': INTERNAL: Failed to open libtpu.so: libtpu.so: cannot open shared object file: No such file or directory\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
mitigatorLearningFairRepresentationDisparateImpactRemoverReweighingCorrelationRemover
Mean Score0.850000.7861110.7750000.763889
Std Score0.000000.0000000.0000000.000000
Mean Time0.746460.0184520.0241060.369011
Custom Dataset0.850000.7861110.7750000.763889
\n", + "
" + ], + "text/plain": [ + "mitigator LearningFairRepresentation DisparateImpactRemover \\\n", + "Mean Score 0.85000 0.786111 \n", + "Std Score 0.00000 0.000000 \n", + "Mean Time 0.74646 0.018452 \n", + "Custom Dataset 0.85000 0.786111 \n", + "\n", + "mitigator Reweighing CorrelationRemover \n", + "Mean Score 0.775000 0.763889 \n", + "Std Score 0.000000 0.000000 \n", + "Mean Time 0.024106 0.369011 \n", + "Custom Dataset 0.775000 0.763889 " + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "from sklearn import datasets\n", + "from holisticai.datasets import Dataset\n", + "\n", + "np.random.seed(42)\n", + "\n", + "# Prepare the data (preprocessing)\n", + "iris = datasets.load_iris()\n", + "iris.keys()\n", + "\n", + "X = pd.DataFrame(iris['data'], columns=iris['feature_names'])\n", + "y = pd.Series(iris['target'], name='target')\n", + "y = y.apply(lambda x: 1 if x == 1 else 0) # binary target variable\n", + "\n", + "protected = np.random.choice(['a', 'b'], X.shape[0]) # illustrative protected attribute\n", + "group_a = pd.Series(protected == 'a', name='group_a') # binary group_a attribute\n", + "group_b = pd.Series(protected == 'b', name='group_b') # binary group_b attribute\n", + "\n", + "# Create a custom dataset object\n", + "my_dataset = Dataset(X=X, y=y, group_a=group_a, group_b=group_b)\n", + "\n", + "# Run the benchmark with the custom dataset\n", + "my_results = benchmark.run(custom_dataset=my_dataset)\n", + "my_results" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Evaluate your own dataset and mitigator" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
mitigatorLearningFairRepresentationDisparateImpactRemoverReweighingMyMitigatorCorrelationRemover
Mean Score0.8500000.7861110.7750000.77500.763889
Std Score0.0000000.0000000.0000000.00000.000000
Mean Time0.2121120.0154820.0151520.01090.008937
Custom Dataset0.8500000.7861110.7750000.77500.763889
\n", + "
" + ], + "text/plain": [ + "mitigator LearningFairRepresentation DisparateImpactRemover \\\n", + "Mean Score 0.850000 0.786111 \n", + "Std Score 0.000000 0.000000 \n", + "Mean Time 0.212112 0.015482 \n", + "Custom Dataset 0.850000 0.786111 \n", + "\n", + "mitigator Reweighing MyMitigator CorrelationRemover \n", + "Mean Score 0.775000 0.7750 0.763889 \n", + "Std Score 0.000000 0.0000 0.000000 \n", + "Mean Time 0.015152 0.0109 0.008937 \n", + "Custom Dataset 0.775000 0.7750 0.763889 " + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "my_mitigator = MyMitigator()\n", + "my_dataset = Dataset(X=X, y=y, group_a=group_a, group_b=group_b)\n", + "\n", + "my_results = benchmark.run(custom_mitigator=my_mitigator, custom_dataset=my_dataset)\n", + "my_results" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "holisticai", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.14" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/tutorials/bias/mitigating_bias/clustering/demos/preprocessing.ipynb b/tutorials/bias/mitigating_bias/clustering/demos/preprocessing.ipynb index 8503dab9..4764e4b7 100644 --- a/tutorials/bias/mitigating_bias/clustering/demos/preprocessing.ipynb +++ b/tutorials/bias/mitigating_bias/clustering/demos/preprocessing.ipynb @@ -15,7 +15,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ @@ -39,7 +39,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 2, "metadata": {}, "outputs": [ { @@ -192,6 +192,11 @@ " align-items: center;\n", "}\n", "\n", + ".generic-object-container,\n", + ".generic-object-header,\n", + ".attribute-list {\n", + " color: #000000 !important; /* Establece el color del texto a negro */\n", + "}\n", " \n", "
\n", "
\n", @@ -200,7 +205,7 @@ " [Dataset]\n", "
\n", "
\n", - "
Instances: 299
Features: X , y , p_attrs , group_a , group_b
Metadata: sex: {'group_a': '0', 'group_b': '1'}
\n", + "
Instances: 16644
Features: X , y , p_attrs , group_a , group_b
Metadata: race: {'group_a': 'Causasian', 'group_b': 'Non-Caucasian'}
\n", " \n", "
\n", "
\n", @@ -222,16 +227,16 @@ " " ], "text/plain": [ - "" + "{\"dtype\":\"Dataset\",\"attributes\":{\"Instances\":16644,\"Features\":[\"X , y , p_attrs , group_a , group_b\"]},\"metadata\":\"race: {'group_a': 'Causasian', 'group_b': 'Non-Caucasian'}\"}" ] }, - "execution_count": 5, + "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "dataset = load_dataset('clinical_records',protected_attribute=\"sex\")\n", + "dataset = load_dataset('compass',protected_attribute=\"race\")\n", "train_test = dataset.train_test_split(test_size=0.2, random_state=42)\n", "\n", "train = train_test['train']\n", @@ -242,7 +247,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 3, "metadata": {}, "outputs": [ { @@ -278,32 +283,32 @@ " \n", " \n", " Cluster Balance\n", - " 0.711297\n", + " 0.319848\n", " 1\n", " \n", " \n", " Minimum Cluster Ratio\n", - " 0.500000\n", + " 1.064137\n", " 1\n", " \n", " \n", " Cluster Distribution Total Variation\n", - " 0.057983\n", + " 0.048022\n", " 0\n", " \n", " \n", " Cluster Distribution KL Div\n", - " 0.021202\n", + " 0.009896\n", " 0\n", " \n", " \n", " Social Fairness Ratio\n", - " 1.217368\n", + " 1.042285\n", " 1\n", " \n", " \n", " Silhouette Difference\n", - " -0.007195\n", + " 0.004703\n", " 0\n", " \n", " \n", @@ -313,15 +318,15 @@ "text/plain": [ " Value Reference\n", "Metric \n", - "Cluster Balance 0.711297 1\n", - "Minimum Cluster Ratio 0.500000 1\n", - "Cluster Distribution Total Variation 0.057983 0\n", - "Cluster Distribution KL Div 0.021202 0\n", - "Social Fairness Ratio 1.217368 1\n", - "Silhouette Difference -0.007195 0" + "Cluster Balance 0.319848 1\n", + "Minimum Cluster Ratio 1.064137 1\n", + "Cluster Distribution Total Variation 0.048022 0\n", + "Cluster Distribution KL Div 0.009896 0\n", + "Social Fairness Ratio 1.042285 1\n", + "Silhouette Difference 0.004703 0" ] }, - "execution_count": 10, + "execution_count": 3, "metadata": {}, "output_type": "execute_result" } @@ -358,7 +363,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -394,32 +399,32 @@ " \n", " \n", " Cluster Balance\n", - " 0.946014\n", + " 0.922569\n", " 1\n", " \n", " \n", " Minimum Cluster Ratio\n", - " 0.507042\n", + " 1.000000\n", " 1\n", " \n", " \n", " Cluster Distribution Total Variation\n", - " 0.037510\n", + " 0.013492\n", " 0\n", " \n", " \n", " Cluster Distribution KL Div\n", - " 0.002869\n", + " 0.000419\n", " 0\n", " \n", " \n", " Social Fairness Ratio\n", - " 1.337623\n", + " 1.041066\n", " 1\n", " \n", " \n", " Silhouette Difference\n", - " -0.064977\n", + " 0.132821\n", " 0\n", " \n", " \n", @@ -429,15 +434,15 @@ "text/plain": [ " Value Reference\n", "Metric \n", - "Cluster Balance 0.946014 1\n", - "Minimum Cluster Ratio 0.507042 1\n", - "Cluster Distribution Total Variation 0.037510 0\n", - "Cluster Distribution KL Div 0.002869 0\n", - "Social Fairness Ratio 1.337623 1\n", - "Silhouette Difference -0.064977 0" + "Cluster Balance 0.922569 1\n", + "Minimum Cluster Ratio 1.000000 1\n", + "Cluster Distribution Total Variation 0.013492 0\n", + "Cluster Distribution KL Div 0.000419 0\n", + "Social Fairness Ratio 1.041066 1\n", + "Silhouette Difference 0.132821 0" ] }, - "execution_count": 11, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" } @@ -471,7 +476,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -507,32 +512,32 @@ " \n", " \n", " Cluster Balance\n", - " 0.946014\n", + " 0.932819\n", " 1\n", " \n", " \n", " Minimum Cluster Ratio\n", - " 0.507042\n", + " 1.022222\n", " 1\n", " \n", " \n", " Cluster Distribution Total Variation\n", - " 0.037510\n", + " 0.011651\n", " 0\n", " \n", " \n", " Cluster Distribution KL Div\n", - " 0.002869\n", + " 0.000312\n", " 0\n", " \n", " \n", " Social Fairness Ratio\n", - " 1.337623\n", + " 1.041066\n", " 1\n", " \n", " \n", " Silhouette Difference\n", - " -0.064977\n", + " 0.131541\n", " 0\n", " \n", " \n", @@ -542,15 +547,15 @@ "text/plain": [ " Value Reference\n", "Metric \n", - "Cluster Balance 0.946014 1\n", - "Minimum Cluster Ratio 0.507042 1\n", - "Cluster Distribution Total Variation 0.037510 0\n", - "Cluster Distribution KL Div 0.002869 0\n", - "Social Fairness Ratio 1.337623 1\n", - "Silhouette Difference -0.064977 0" + "Cluster Balance 0.932819 1\n", + "Minimum Cluster Ratio 1.022222 1\n", + "Cluster Distribution Total Variation 0.011651 0\n", + "Cluster Distribution KL Div 0.000312 0\n", + "Social Fairness Ratio 1.041066 1\n", + "Silhouette Difference 0.131541 0" ] }, - "execution_count": 7, + "execution_count": 5, "metadata": {}, "output_type": "execute_result" } @@ -559,7 +564,7 @@ "mitigator = FairletClusteringPreprocessing(seed=42)\n", "\n", "# set the pipeline\n", - "pipeline = Pipeline(steps=[('bm_preprocessing', mitigator), ('model', KMeans(n_clusters = 3, random_state=42))])\n", + "pipeline = Pipeline(steps=[('bm_preprocessing', mitigator), ('model', model)])\n", "pipeline.fit(train['X'], bm__group_a = train['group_a'], bm__group_b = train['group_b'])\n", "\n", "# predict the clusters and get the centroids\n", @@ -580,7 +585,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 6, "metadata": {}, "outputs": [ { @@ -627,38 +632,38 @@ " \n", " \n", " Cluster Balance\n", - " 0.946014\n", - " 0.946014\n", + " 0.922569\n", + " 0.932819\n", " 1\n", " \n", " \n", " Minimum Cluster Ratio\n", - " 0.507042\n", - " 0.507042\n", + " 1.000000\n", + " 1.022222\n", " 1\n", " \n", " \n", " Cluster Distribution Total Variation\n", - " 0.037510\n", - " 0.037510\n", + " 0.013492\n", + " 0.011651\n", " 0\n", " \n", " \n", " Cluster Distribution KL Div\n", - " 0.002869\n", - " 0.002869\n", + " 0.000419\n", + " 0.000312\n", " 0\n", " \n", " \n", " Social Fairness Ratio\n", - " 1.337623\n", - " 1.337623\n", + " 1.041066\n", + " 1.041066\n", " 1\n", " \n", " \n", " Silhouette Difference\n", - " -0.064977\n", - " -0.064977\n", + " 0.132821\n", + " 0.131541\n", " 0\n", " \n", " \n", @@ -669,15 +674,15 @@ " Traditional Pipeline \n", " Value Value Reference\n", "Metric \n", - "Cluster Balance 0.946014 0.946014 1\n", - "Minimum Cluster Ratio 0.507042 0.507042 1\n", - "Cluster Distribution Total Variation 0.037510 0.037510 0\n", - "Cluster Distribution KL Div 0.002869 0.002869 0\n", - "Social Fairness Ratio 1.337623 1.337623 1\n", - "Silhouette Difference -0.064977 -0.064977 0" + "Cluster Balance 0.922569 0.932819 1\n", + "Minimum Cluster Ratio 1.000000 1.022222 1\n", + "Cluster Distribution Total Variation 0.013492 0.011651 0\n", + "Cluster Distribution KL Div 0.000419 0.000312 0\n", + "Social Fairness Ratio 1.041066 1.041066 1\n", + "Silhouette Difference 0.132821 0.131541 0" ] }, - "execution_count": 12, + "execution_count": 6, "metadata": {}, "output_type": "execute_result" } @@ -703,7 +708,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.11.9" + "version": "3.10.14" }, "orig_nbformat": 4 }, diff --git a/tutorials/bias/mitigating_bias/regression/examples/example_us_crime.ipynb b/tutorials/bias/mitigating_bias/regression/examples/example_us_crime.ipynb index 686d3399..abf190b0 100644 --- a/tutorials/bias/mitigating_bias/regression/examples/example_us_crime.ipynb +++ b/tutorials/bias/mitigating_bias/regression/examples/example_us_crime.ipynb @@ -49,7 +49,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 1, "metadata": { "id": "cb0SfHEYJfNo" }, @@ -71,7 +71,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 2, "metadata": { "id": "3X1keA2QORTA" }, @@ -104,7 +104,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 3, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -129,7 +129,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 4, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -289,6 +289,11 @@ " align-items: center;\n", "}\n", "\n", + ".generic-object-container,\n", + ".generic-object-header,\n", + ".attribute-list {\n", + " color: #000000 !important; /* Establece el color del texto a negro */\n", + "}\n", " \n", "
\n", "
\n", @@ -319,10 +324,10 @@ " " ], "text/plain": [ - "" + "{\"dtype\":\"Dataset\",\"attributes\":{\"Instances\":1993,\"Features\":[\"X , y , p_attrs , group_a , group_b\"]},\"metadata\":\"race: {'group_a': 'racePctWhite>0.5', 'group_b': 'racePctWhite<=0.5'}\"}" ] }, - "execution_count": 8, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" } @@ -341,7 +346,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -361,7 +366,7 @@ "Length: 101, dtype: int64" ] }, - "execution_count": 9, + "execution_count": 5, "metadata": {}, "output_type": "execute_result" } @@ -384,7 +389,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 6, "metadata": { "id": "ivo9UpXeJqDA" }, @@ -395,7 +400,7 @@ "" ] }, - "execution_count": 10, + "execution_count": 6, "metadata": {}, "output_type": "execute_result" }, @@ -426,7 +431,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 7, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -442,7 +447,7 @@ "" ] }, - "execution_count": 11, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" }, @@ -472,7 +477,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 8, "metadata": { "id": "e1FVWNLPJ86z" }, @@ -497,7 +502,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 9, "metadata": { "id": "VLM933LQJ_2l" }, @@ -541,7 +546,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 10, "metadata": { "id": "8jBCOnh5CpVd" }, @@ -676,7 +681,7 @@ "Correlation Difference 0.002428 0" ] }, - "execution_count": 14, + "execution_count": 10, "metadata": {}, "output_type": "execute_result" } @@ -732,7 +737,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 11, "metadata": { "id": "UbylXQcaKVfo" }, @@ -743,7 +748,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 12, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -753,13 +758,6 @@ "outputId": "f177e10a-e285-41fa-a8d6-a4ef90b4a1b9" }, "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "An NVIDIA GPU may be present on this machine, but a CUDA-enabled jaxlib is not installed. Falling back to cpu.\n" - ] - }, { "data": { "text/html": [ @@ -848,7 +846,7 @@ " \n", " \n", " RMSE Ratio Q80\n", - " 0.483447\n", + " 0.483448\n", " 1\n", " \n", " \n", @@ -884,13 +882,13 @@ "Max Statistical Parity 0.216814 0\n", "Statistical Parity AUC 0.107243 0\n", "RMSE Ratio 0.374976 1\n", - "RMSE Ratio Q80 0.483447 1\n", + "RMSE Ratio Q80 0.483448 1\n", "MAE Ratio 0.332584 1\n", "MAE Ratio Q80 0.434959 1\n", "Correlation Difference 0.002576 0" ] }, - "execution_count": 16, + "execution_count": 12, "metadata": {}, "output_type": "execute_result" } @@ -932,7 +930,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 13, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -945,7 +943,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "RMS error: 0.1876979543987722\n" + "RMS error: 0.18769793124732634\n" ] } ], @@ -968,7 +966,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 14, "metadata": {}, "outputs": [], "source": [ @@ -977,7 +975,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 15, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1117,7 +1115,7 @@ "Correlation Difference -0.051509 0" ] }, - "execution_count": 19, + "execution_count": 15, "metadata": {}, "output_type": "execute_result" } @@ -1157,7 +1155,7 @@ }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 16, "metadata": {}, "outputs": [ { @@ -1184,7 +1182,7 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": 17, "metadata": {}, "outputs": [], "source": [ @@ -1193,7 +1191,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 18, "metadata": {}, "outputs": [ { @@ -1326,7 +1324,7 @@ "Correlation Difference 0.011022 0" ] }, - "execution_count": 22, + "execution_count": 18, "metadata": {}, "output_type": "execute_result" } @@ -1363,7 +1361,7 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 19, "metadata": {}, "outputs": [ { @@ -1397,7 +1395,7 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 20, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1535,7 +1533,7 @@ " \n", " RMSE Ratio Q80\n", " 0.966421\n", - " 0.483447\n", + " 0.483448\n", " 0.845568\n", " 0.429231\n", " 1\n", @@ -1582,7 +1580,7 @@ "Max Statistical Parity 0.767257 0.216814 \n", "Statistical Parity AUC 0.452860 0.107243 \n", "RMSE Ratio 0.614069 0.374976 \n", - "RMSE Ratio Q80 0.966421 0.483447 \n", + "RMSE Ratio Q80 0.966421 0.483448 \n", "MAE Ratio 0.486594 0.332584 \n", "MAE Ratio Q80 0.937572 0.434959 \n", "Correlation Difference 0.002428 0.002576 \n", @@ -1624,7 +1622,7 @@ "Correlation Difference 0 " ] }, - "execution_count": 24, + "execution_count": 20, "metadata": {}, "output_type": "execute_result" } @@ -1680,7 +1678,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.11.9" + "version": "3.10.14" } }, "nbformat": 4,